Liver Molecular Mechanisms Involved in Type 2 Diabetes

Transcription

Liver Molecular Mechanisms Involved in Type 2 Diabetes:
An Investigation Using Roux-en Y Gastric Bypass as a Human
Model of Diabetes Remission
By
Vinko Besic
Wakefield Biomedical Research Unit
Department of Pathology and Molecular Medicine
The University of Otago (Wellington)
A thesis submitted to the University of Otago for the
degree of Doctor of Philosophy
2013
ii
Abstract
Type 2 diabetes mellitus is a chronic disease characterised by progressive insulin
resistance and loss of β-cell function. An incomplete understanding of its
pathogenesis is hindering the effective treatment of this disease. Roux-en Y gastric
bypass surgery (RYGB); however, causes rapid remission of liver insulin resistance
and type 2 diabetes, and therefore affords us an opportunity to examine some
fundamental characteristics of these conditions. Gathering evidence suggests that liver
insulin resistance may be a crucial contributor to development of diabetes. In this
thesis, we used liver biopsies taken before, and in some individuals after, RYGB
surgery to explore or identify several molecular processes involved in the
pathogenesis of type 2 diabetes. The study cohort included individuals with normal
glucose tolerance and others with type 2 diabetes.
Ecto-nucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1) may cause insulin
resistance through its inhibitory action on insulin signalling. This thesis provides
evidence to the contrary, suggesting that liver ENPP1 is not a contributor to liver
insulin resistance. We found liver ENPP1 protein abundance was lower in individuals
with type 2 diabetes than in those with normal glucose tolerance, and increased after
RYGB surgery in those individuals who had remission of diabetes. ENPP1 positively
correlated with insulin sensitivity at the liver which is contrary to what others have
reported in muscle and adipose tissue. We reasoned that our findings are likely due to
the hypothesized role of ENPP1 as a natural modulator of insulin signalling and the
unique role the liver has in insulin processing.
Changes in the expression ratio of insulin receptor (IR) isoform A (IR-A) and B (IRB) have previously been implicated in the pathogenesis of type 2 diabetes. The
metabolically active IR-B isoform has been shown to predominate in the liver, with
the liver IR-B:A ratio being reported to be 9.8. By assaying levels of IR-A and IR-B
mRNA expression we found that the ratio of liver IR-B:A was abnormally low in
individuals with type 2 diabetes (5.2) and increased with remission of diabetes (5.4 to
8.6).
iii
The change in ratio was due to a diminished IR-A expression following remission of
diabetes. Further work with an in vitro cell model showed that insulin’s ability to
inhibit gluconeogenesis in Hep G2 cells overexpressing IR-A was reduced, suggesting
that the altered liver IR-B:A ratio observed in diabetes may have a detrimental effect
on glucose homeostasis.
Access to liver tissue before and after RYGB surgery afforded a rare opportunity to
observe changes in liver gene expression before and after remission of diabetes. Using
gene microarray analysis, we showed that the majority of genes that were
differentially regulated after RYGB surgery are involved in lipotoxicity,
inflammation, and ER stress, particularly in those individuals who had remission of
type 2 diabetes. Although there were many significant observations, the apparent
inter-organ communication between the liver and the pancreas presents a hitherto
unconfirmed relationship whereby the liver can not only mediate pancreatic β cell
size, but can also regulate insulin secretion. This work has identified a mechanism
which may be exploited to develop novel treatments for type 2 diabetes.
In conclusion this thesis describes several novel findings with respect to the
pathogenesis of type 2 diabetes. It provides novel data on the role of ENPP1 and IR
isoforms in insulin signalling which have furthered our understanding of the insulin
signalling pathway. In addition, microarray gene analysis in liver tissue from before
and after improvements in insulin resistance and remission of type 2 diabetes has
allowed us to identify several candidate genes that are worthy of further investigation.
iv
Acknowledgements
First I would like to thank Dr Mark Hayes and Professor Richard Stubbs for the
support and guidance they have offered over the course of my study. They offered
freedom rarely afforded to other PhD students which allowed me to develop into an
independent and capable scientist. Dr Hayes’s assistance regarding experimental
design and study direction was invaluable and critical to the completion of this work.
Additional thanks are given to Professor Richard Stubbs, without whom none of the
work presented here would be possible. His vision and dedication to the scientific
method formed the extensive database and collection of human samples that was the
fundamental basis for my thesis.
I would also like to acknowledge our research collaborators that have contributed
significantly to the completion of my work. I would like to thank Professor C. Ronald
Kahn and staff at the Joslin Diabetes Centre, Boston, MA, US for providing the IR-B
isoform expression vector and Professor Conan Fee and the staff at the Biomolecular
Interaction Centre for sequencing and validating said vector. Professor Eric Hoffman
and the staff at the Research Centre for Genetic Medicine, Children’s Research
Hospital, Washington DC, US for conducting the microarray gene analysis and
Professor Parry Guilford at the Cancer Genetics Laboratory, Department of
Biochemistry, University of Otago, Dunedin for allowing me to use their ABI Prism®
7900HT Sequence Detection System. I would also like to thank the many
organisations that have provided funding, without which none of my work would be
possible. They include; the Wakefield Biomedical Research Unit, the Wakefield
Gastroenterology Research Trust and the Wellington Medical Research Foundation.
My education in the Life Sciences has been broad and that is due in no small part to
knowledgeable and supportive staff of the Wakefield Biomedical Research Unit, all of
whom were always happy to answer my incessant questions. Fiona, Dan, John, Sarah,
Chandra, I will be forever grateful for the patience you have shown. I would
especially like to thank Hong Jun Shi (Sifu) as without him large tracts of this thesis
would not be possible.
v
I would also like to thank the Department of Pathology and Molecular Medicine. The
time after the closure of the WBRU was particularly difficult and all the staff at the
Path Department were supportive and made me feel welcome. I would like to thank
Professor Brett Delahunt in particular. The support and guidance he offered went a
long way in helping me finish this work and just knowing he was there if I needed
him was comforting. I would also like to express my appreciation to Jane, Robyn, St
John and Coleen, who were always happy to listen to me whinge during the arduous
process of writing up. I cannot overstate how necessary they were.
Finally, I would like to express my deepest gratitude to my family, without whom I
would have failed. Thank you to Andro and Ana Besic for the money they spent on
raising and educating me and for the support they offered during my studies. Most of
all I would like to thank Jennie Besic. Of all the people mentioned here she put up
with the most, never complained and was always there when I needed her.
vi
Dissemination of Work
Publications
Besic V, Stubbs R.S, Hayes M.T. ENPP1 protein increases in liver tissue of individuals who
have a remission of type 2 diabetes after Roux-en Y gastric bypass surgery. Submitted to
Journal of Clinical Endocrinology and Metabolism
Besic V, Stubbs R.S, Shi HJ, Hayes M.T. Liver insulin receptor isoform A expression
decreases with remission of type 2 diabetes after Roux-en Y gastric bypass surgery. Submitted
to Journal of Endocrinology
Besic V, Stubbs R.S, Hayes M.T. Liver gene expression after Roux-en Y gastric bypass
surgery and remission of type 2 diabetes: novel insights into type 2 diabetes pathogenesis.
Manuscript in preparation.
Conference Proceedings and Presentations
Besic V, Hayes M.T, Stubbs R.S. Investigation of the relative abundance of insulin receptor
isotype mRNA in human liver tissue from obese insulin sensitive and diabetic patients. 7th
International Diabetes Federation Western Pacific Regions Congress, Wellington, New
Zealand, March 30-April 3 2008.
Besic V, Hayes M.T, Stubbs R.S. Relative abundance of insulin receptor isoform mRNA and
protein from Diabetic and Non-diabetic individuals. Queenstown Molecular Biology Meeting,
Queenstown, New Zealand, August 31-Sepetember6 2008.
Besic V, Hayes M.T, Stubbs R.S. Candidate gene expression studies in liver samples from
obese patients before and after Roux-en-Y gastric bypass surgery. Pre-Diabetes and Metabolic
Syndrome Congress, Madrid, Spain, April 4-9 2011.
Besic V, Shi H.J, Stubbs R.S, Hayes M.T. Changes in expression of insulin receptor isoforms
present a potential mechanism for driving insulin resistance at the liver. Health Sciences
Divisional Research Forum, Dunedin, New Zealand, 17 September 2012.
Besic V, Hayes M.T, Stubbs R.S. Changes in insulin receptor isoform expression presents a
potential mechanism for driving insulin resistance at the liver. Wellington Health and
Biomedical Research Society Young New Investigator of the Year (2012) –Richard Stewart
Memorial Prize
vii
viii
Table of Contents
Abstract ........................................................................................................................ iii
Acknowledgements........................................................................................................v
Dissemination of Work ................................................................................................vii
Table of Contents..........................................................................................................ix
List of Tables ..............................................................................................................xvi
List of Figures ............................................................................................................xvii
List of Abbreviations ..................................................................................................xxi
CHAPTER ONE: The pathology of type 2 diabetes ....................................................1
1.1 Type 2 diabetes mellitus: a disease for modern man ...........................................2
1.2 Insulin resistance and impaired insulin secretion ................................................4
1.3 Insulin action: from secretion to binding of the insulin receptor.........................5
1.3.1 Insulin secretion............................................................................................5
1.3.2 Hepatic insulin clearance ..............................................................................7
1.3.3 Molecules involved in hepatic insulin clearance ..........................................8
1.3.4 Classical insulin target tissues and consequence of signal transmission ......9
1.3.5 Non-classical insulin target tissue...............................................................11
1.4 Insulin resistance in the classical insulin target tissues .....................................11
1.4.1 Measurement of insulin resistance..............................................................12
1.4.2 Insulin resistance in peripheral tissue .........................................................13
1.4.3 Insulin resistance at the liver ......................................................................13
1.5 Insulin target tissue relevance in regulating glucose homeostasis.....................15
1.5.1 Transgenic animal models of tissue specific insulin resistance..................15
1.5.2 Interplay between key insulin target tissues in progression to type 2
diabetes: does the liver have overriding importance?..........................................17
1.6 Insulin Signalling ...............................................................................................19
1.6.1 Insulin receptor ...........................................................................................19
1.6.2 Insulin receptor substrate (IRS) proteins ....................................................20
1.6.3 Negative regulation of IR activity ..............................................................21
1.6.4 PI3-K/AKT pathway...................................................................................22
1.6.5 PI3-K/AKT-mediated regulation of glycogen synthesis ............................22
1.6.6 PI3-K/AKT-mediated regulation of hepatic glucose output.......................24
ix
1.6.7 PI3-K/AKT-mediated regulation of lipid homeostasis ...............................24
1.6.8 Ras/MAPK pathway ...................................................................................25
1.7 Obesity and type 2 diabetes: the search for a common pathway .......................25
1.7.1 Obesity and β-cell dysfunction ...................................................................25
1.7.2 Obesity and insulin resistance.....................................................................26
1.7.3 Lipotoxicity.................................................................................................27
1.7.4 Molecular mechanism behind lipotoxicity .................................................28
1.7.5 Inflammation...............................................................................................29
1.7.6 Molecular mechanisms behind inflammation-mediated insulin resistance 30
1.7.7 The unfolded protein response: a link between lipotoxicity and
inflammation ........................................................................................................32
1.8 A potential model detailing progression from insulin resistance to type 2
diabetes ....................................................................................................................34
1.8.1 Lipotoxicity and inflammation: a feed-forward mechanism ......................34
1.8.2 Primary hepatic insulin resistance in animal models..................................35
1.8.3 The liver as a key organ in the progression to type 2 diabetes ...................37
1.8.4 From insulin resistance to type 2 diabetes: a potential hypothesis .............38
1.9 Insulin receptor isoforms and their role in selective insulin signalling .............40
1.9.1 Metabolic vs mitogenic outcomes of IR signalling ....................................42
1.9.2 Hybrid receptors .........................................................................................44
1.9.3 Insulin receptor and its isoforms in diabetes: an incomplete story.............46
1.10 Bariatric surgery: A model of type 2 diabetes remission.................................47
1.10.1 Clinical outcomes of bariatric surgery......................................................49
1.10.2 Defining the normalization of glucose homoeostasis after bariatric surgery
.............................................................................................................................51
1.10.3 Weight -independent anti-diabetic effects of RYGB surgery...................52
1.10.4 Mechanisms behind remission of diabetes after bariatric surgery............53
1.10.5 Tissue specific changes in insulin action after RYGB surgery.................54
1.11 Hypothesis and Aims .......................................................................................56
CHAPTER TWO: Roux-en Y gastric bypass surgery can be used to study type 2
diabetes ........................................................................................................................59
2.1 Introduction: using models to study type 2 diabetes..........................................60
2.1.1 Experimental strategy .................................................................................61
2.2 Research Design and Methods...........................................................................62
x
2.2.1 Pre operative data collection and type 2 diabetes diagnosis .......................62
2.2.2 Subject selection and grouping ...................................................................63
2.2.3 Tissue sampling and follow up data collection...........................................64
2.2.4 Biochemical testing.....................................................................................64
2.2.4.1 Glucose assay and method principle....................................................65
2.2.4.2 HbA1c assay.........................................................................................65
2.2.4.3 Insulin assay and method principle......................................................66
2.2.5 Estimation of insulin resistance ..................................................................66
2.2.6 Statistical analysis.......................................................................................67
2.3 Results................................................................................................................68
2.3.1 Weight loss after bariatric surgery..............................................................68
2.3.2 Improvements in insulin resistance after RYGB surgery ...........................69
2.3.3 Remission of diabetes after RYGB surgery................................................71
2.3.4 Metabolic status during liver biopsy...........................................................72
2.4 Discussion ..........................................................................................................75
CHAPTER THREE: Investigation of molecules involved in the modulation of insulin
signalling strength and hepatic insulin clearance.........................................................79
3.1 Introduction........................................................................................................80
3.1.1 Experimental strategy .................................................................................81
3.2 Research Design and Methods...........................................................................82
3.2.1 RNA extraction ...........................................................................................82
3.2.2 RNA quality control....................................................................................83
3.2.3 First strand cDNA synthesis .......................................................................83
3.2.4 Analysis of gene expression with RT-qPCR ..............................................84
3.2.5 EXPRESS qPCR SuperMix........................................................................84
3.2.6 TaqMan gene expression assays .................................................................85
3.2.7 RT-qPCR method principle and analysis of data........................................85
3.2.8 Relative quantification of gene expression.................................................86
3.2.9 Reference genes ..........................................................................................88
3.2.10 Protein extraction......................................................................................88
3.2.11 Protein resolution with E-PAGE™...........................................................89
3.2.12 Sample preparation and loading ...............................................................89
3.2.13 Electrophoretic transfer to PVDF and western blotting............................89
3.2.14 Stripping....................................................................................................90
xi
3.2.15 Relative quantification of protein abundance ...........................................90
3.2.16 Statistical analysis.....................................................................................91
3.3 Results................................................................................................................92
3.3.1 Liver ENPP1 levels in diabetes ..................................................................92
3.3.2 Liver ENPP1 levels before and after RYGB surgery .................................93
3.3.3 Liver CEACAM-1 and IDE levels in diabetes ...........................................96
3.3.4 Liver CEACAM-1 and IDE levels before and after RYGB surgery ..........97
3.4 Discussion ..........................................................................................................99
3.4.1 ENPP1.......................................................................................................100
3.4.2 CEACAM-1 and IDE................................................................................102
CHAPTER FOUR: Liver insulin receptor isoforms and their role in the pathogenesis
of type 2 diabetes .......................................................................................................105
4.1 Introduction......................................................................................................106
4.1.1 Experimental strategy ...............................................................................107
4.2 Research Design and Methods.........................................................................108
4.2.1 Analysis of IR-A and IR-B mRNA expression using RT-qPCR..............108
4.2.2 Analysis of IR protein abundance using E-PAGE electrophoresis and
western blotting..................................................................................................109
4.2.3 Functional studies of insulin receptor and its isoforms in cultured liver cells
...........................................................................................................................110
4.2.3.1 Propagation and storage.....................................................................110
4.2.3.2 In Vitro experiments ..........................................................................111
4.2.3.3 RNA extraction from cultured cells...................................................111
4.2.3.4 RNA quantification............................................................................111
4.2.3.5 RT-qPCR of IR-A/IR-B and PCK1 in Hep G2 cells .........................112
4.2.3.6 Protein extraction ...............................................................................112
4.2.3.7 SDS-PAGE ........................................................................................113
4.2.3.8 Electrophoretic transfer to PVDF and western blotting.....................113
4.2.3.9 Stripping.............................................................................................113
4.2.4 Transient Knockdown of IR-A and IR-B isoform expression using siRNA
...........................................................................................................................114
4.2.4.1 siRNA transfection.............................................................................114
4.2.4.2 Reverse transfection of siRNA ..........................................................115
4.2.5 IR-A and IR-B Isoform overexpression vector.........................................116
4.2.5.1 Transformation...................................................................................116
4.2.5.2 Vector propagation and extraction.....................................................117
4.2.5.3 Plasmid enzyme digestion..................................................................117
4.2.5.4 PCR of linear plasmid........................................................................118
xii
4.2.5.5 Attempted cloning of IR-A ................................................................120
4.2.5.6 OriGene IR-A expression vector .......................................................121
4.2.5.7 Plasmid transfection...........................................................................121
4.2.6 Insulin treatment of Hep G2 cells .............................................................122
4.2.7 Insulin-induced AKT phosphorylation in cells overexpressing IR-A or IRB.........................................................................................................................123
4.2.8 Insulin-induced inhibition of PCK1 transcription in cells overexpressing
IR-A or IR-B ......................................................................................................123
4.2.9 Statistical analysis.....................................................................................124
4.3 Results..............................................................................................................125
4.3.1 Liver insulin receptor protein abundance in individuals with type 2 diabetes
...........................................................................................................................125
4.3.2 Liver insulin receptor protein abundance after RYGB surgery ................126
4.3.3 IR-B:A mRNA ratio in individuals with type 2 diabetes..........................127
4.3.4 IR-B:A mRNA ratio before and after RYGB surgery ..............................128
4.3.5 IR abundance and relative isoform expression in Hep G2 cells treated with
supraphysiological levels of insulin...................................................................130
4.3.6 Functional studies of IR isoform specific insulin signalling ....................131
4.3.7 Overexpression of IR-B and IR-A in Hep G2 cells ..................................132
4.3.8 Confirmation of isoform specific over expression....................................133
4.3.9 AKT phosphorylation in Hep G2 cells overexpressing IR-A or IR-B .....134
4.3.10 Insulin-induced PCK1 mRNA expression in Hep G2 cells overexpressing
IR-A or IR-B ......................................................................................................136
4.4 Discussion ........................................................................................................137
4.4.1 Changes in IR isoform expression affect insulin signalling in Hep G2 cells
...........................................................................................................................139
CHAPTER FIVE: Changes in global hepatic gene expression after RYGB Surgery
....................................................................................................................................141
5.1 Introduction......................................................................................................142
5.1.1 Experimental strategy ...............................................................................143
5.2 Research Design and Methods.........................................................................144
5.2.1 RNAstable transport to Children’s Research Hospital .............................144
5.2.2 cRNA synthesis.........................................................................................144
5.2.3 Illumina microarray ..................................................................................145
5.2.4 GeneomeStudio™ Gene Expression Module data output ........................147
5.2.5 Analysis of microarray data with BRB ArrayTools .................................147
xiii
5.2.5.1 Processing, collating and filtering the data ........................................148
5.2.5.2 Data normalisation .............................................................................148
5.2.6 Gene annotation ........................................................................................150
5.2.7 Class comparison ......................................................................................150
5.2.8 Multivariate Permutation Test and the Random Variance Model ............150
5.2.9 TaqMan® Low Density Array confirmatory RT-qPCR...........................152
5.2.9.1 TaqMan® Array Micro Fluidic Card loading, sealing and analysis..153
5.2.9.2 Statistical analysis of TLDA RT-qPCR data .....................................154
5.2.10 Analysis of LGALS4 protein abundance using E-PAGE electrophoresis
and western blotting...........................................................................................155
5.3 Results..............................................................................................................156
5.3.1 Differential gene expression after RYGB surgery....................................156
5.3.1.1 Glucose metabolism...........................................................................156
5.3.1.2 Lipid metabolism and amino acid metabolism ..................................164
5.3.1.3 Inflammation, immunity and blood circulation .................................164
5.3.1.4 Transcriptional regulation..................................................................165
5.3.1.5 Cell growth and differentiation..........................................................165
5.3.1.6 Xenobiotic metabolism ......................................................................166
5.3.1.7 Cellular transport and structure..........................................................166
5.3.2 Differential gene expression between individuals with normal glucose
tolerance or type 2 diabetes................................................................................167
5.4 Validation of microarray profiles with TLDA RT-qPCR................................169
5.4.1 Differentially expressed genes after RYGB surgery that were confirmed by
TLDA RT-qPCR................................................................................................171
5.4.2 Differentially expressed genes between individuals with normal glucose
tolerance and type 2 diabetes that were confirmed by TLDA RT-qPCR ..........175
5.4.3 LGALS4 protein abundance is elevated in individuals with type 2 diabetes
...........................................................................................................................176
5.5 Discussion ........................................................................................................177
CHAPTER SIX: Global changes in hepatic gene expression after RYGB surgery
reflect the pathology of type 2 diabetes .....................................................................183
6.1 Glucose metabolism.........................................................................................185
6.2 Regulation of metabolic homeostasis via leptin signalling..............................186
6.3 Lipotoxicity and non-alcoholic fatty liver disease...........................................187
6.4 Inflammation and blood coagulation ...............................................................189
6.5 Endoplasmic reticulum stress ..........................................................................192
6.6 The liver and inter organ communication........................................................193
6.7 Circadian rhythm .............................................................................................195
xiv
6.8 Concluding remarks .........................................................................................197
CHAPTER SEVEN: Final discussion and experimental limitations........................199
7.1 Insulin processing at the liver: a new role for ENPP1 .....................................202
7.2 Insulin receptor isoforms in diabetes: detrimental affects of IR-A
overexpression on regulation of glucose homeostasis ...........................................204
7.3 Changes in gene expression after remission of diabetes reveal new roles for the
liver in regulating metabolic homeostasis..............................................................207
7.4 Conclusion .......................................................................................................209
REFRENCES............................................................................................................211
APPENDIX...............................................................................................................275
9.1 Western Blots for Chapter 3 ............................................................................275
9.2.1 AKT Phosphorylation Western Blots .......................................................281
xv
List of Tables
Table 1-1. Transgenic tissue specific mouse models of insulin resistance..................16
Table 1-2. Metabolic disease state after traditional bariatric surgery. Data adapted
from systematic review and meta-analysis by Buchwald et al. (2004). Data in brackets
is updated remission rates from Buchwald et al. (2009)..............................................50
Table 1-3. Treatment outcomes for type 2 diabetes as defined by Buse et al. (2009). 52
Table 2-1. Anthropometric data of 55 obese individuals classified by study group....63
Table 2-2. Metabolic status at the time of initial liver biopsy .....................................73
Table 2-3. Metabolic status of individuals at the time of initial and subsequent liver
biopsy...........................................................................................................................73
Table 3-1. Mean RNA integrity number for all groups being compared.....................83
Table 3-2. Thermocycling conditions for RT-qPCR reactions using EXPRESS qPCR
SuperMix......................................................................................................................84
Table 3-3. TaqMan gene expression assay amplification efficiency...........................88
Table 4-1. IR-A and IR-B siRNA sequence ..............................................................114
Table 4-2. Touchdown PCR thermocycling parameters for insulin receptor. ...........119
Table 4-3. Phusion High Fidelity PCR thermocycling parameters for insulin receptor
....................................................................................................................................120
Table 5-1. Illumina HumanHT-12_v4_BeadChip expression content ......................146
Table 5-2. Illumina microarray differentially expressed (≥ 1.35 fold) liver genes ~16
months after RYGB surgery (n=15). .........................................................................157
Table 5-3. Illumina microarray differentially expressed genes between the sNGT
(n=8) and sT2DM (n=7) groups.................................................................................168
Table 5-4. Confirmed genes that were differentially expressed after RYGB surgery.
Highlighted genes are likely affected by changes in diabetic or insulin-resistant status.
....................................................................................................................................172
xvi
List of Figures
Figure 1.2-1. Relationship between pancreatic beta cell function and insulin
sensitivity. ......................................................................................................................4
Figure 1.3-1. Schematic diagram of biphasic insulin secretion after square wave
increase in plasma glucose concentration. .....................................................................7
Figure 1.3-2. Direct effect of insulin on the three classical insulin target tissues of
liver, muscle and fat.....................................................................................................10
Figure 1.4-1. A brief depiction of the key organs responsible for the unregulated
hyperglycaemia seen in type 2 diabetes.......................................................................14
Figure 1.6-1. A brief depiction of insulin signalling and its outcomes in the
hepatocyte. ...................................................................................................................23
Figure 1.8-1. A hypothesized model describing the sequence of events leading to
development of Type 2 diabetes. .................................................................................39
Figure 1.9-1. Schematic diagram of the different insulin receptor hybrids and ligands.
......................................................................................................................................45
Figure 1.10-1. Typical bariatric surgical procedures. ..................................................48
Figure 1.10-2. Improvement in fasting plasma glucose concentration and HOMA-IR
after RYGB surgery. ....................................................................................................55
Figure 2.2-1. Glucose method principle. .....................................................................65
Figure 2.3-1. Mean BMI ± SD after RYGB surgery in all individuals and the subset
that had a second liver biopsy......................................................................................68
Figure 2.3-2. Change in mean fasting plasma insulin concentrations and HOMA-IR ±
SD in the NGT, IGT and T2DM groups (A and C) and in the sNGT and sT2DM
groups (B and D) up to one year after RYGB surgery................................................70
Figure 2.3-3. Change in mean fasting plasma glucose concentrations and HbA1c ± SD
in the NGT, IGT and T2DM groups (A and C) and in the sNGT and sT2DM groups
(B and D) up to one year after RYGB surgery............................................................71
Figure 3.2-1. 2-∆∆Ct formula for calculating fold changes in gene expression between a
calibrator and test sample.............................................................................................87
Figure 3.3-1. Mean ENPP1 levels in liver tissue of morbidly obese individuals. NGT;
Normal glucose tolerance (n=19), T2DM; Type 2 diabetes (n=27). ...........................93
xvii
Figure 3.3-2. Liver ENPP1 gene expression and protein abundance in liver tissue of
morbidly obese individuals before and after RYGB surgery (● sNGT, n=8; ○ sT2DM,
n=8). .............................................................................................................................94
Figure 3.3-3. (A) Pearson product-moment correlation of liver ENPP1 protein relative
abundance (log) vs fasting plasma insulin concentration (log). (B) Pearson productmoment correlation of liver ENPP1 protein relative abundance (log) vs HOMA-IR
(log)..............................................................................................................................95
Figure 3.3-4 Mean + SE CAECAM-1 mRNA expression (A) and IDE mRNA
expression (B) presented as 2-∆Ct normalised to 18S. NGT; Normal glucose tolerance
(n=19), T2DM; Type 2 diabetes (n=27). .....................................................................96
Figure 3.3-5. CEACAM-1 and IDE protein abundance in liver tissue of morbidly
obese individuals. NGT; Normal glucose tolerance (n=17), T2DM; Type 2 diabetes
(n=26)...........................................................................................................................97
Figure 3.3-6. Liver CEACAM-1 and IDE gene expression and protein abundance in
liver tissue of morbidly obese individuals before and after RYGB surgery (● sNGT,
n= 8; ○ sT2DM, n=8)...................................................................................................98
Figure 4.2-1. IR-A and IR-B amplified region including MGM probe binding area.109
Figure 4.2-2. Map of pcDNA3.1+ vector containing human IR-B clone. .................118
Figure 4.2-3. Forward and reverse primer sequences for confirmation of IR-B cDNA
clone and further cloning of IR-A cDNA designed using the insulin receptor
consensus coding sequence (INSR CCDS)................................................................118
Figure 4.2-4. Map of pCMV6-XL4 vector containing human IR-A clone................121
Figure 4.3-1. (A) Mean ± SE IR protein abundance in liver tissue of morbidly obese
individuals at the time of RYGB surgery. NGT; Normal glucose tolerance (n=17),
T2DM; Type 2 diabetes (n=26). ................................................................................125
Figure 4.3-2. (A) Pearson product-moment correlation of liver IR protein relative
abundance (log) vs fasting plasma insulin concentration (log). (B) Pearson productmoment correlation of liver IR protein relative abundance (log) vs HOMA-IR (log).
....................................................................................................................................126
Figure 4.3-3. (A) Mean ± SE IR protein abundance in liver tissue of morbidly obese
individuals before and after RYGB surgery (● sNGT, n = 6; ○ sT2DM, n = 7). ......127
Figure 4.3-4. Insulin receptor A and B isoform mRNA expression levels in liver tissue
of morbidly obese individuals. NGT; Normal glucose tolerance (n=19), T2DM; Type
2 diabetes (n=27). ......................................................................................................128
xviii
Figure 4.3-5. IR isoform relative expression in liver tissue of morbidly obese
individuals before and after RYGB surgery (● sNGT, n = 8; ○ sT2DM, n = 8). ......129
Figure 4.3-6. Mean + SE Insulin Receptor protein abundance (A) and isoform
expression presented as 2-∆Ct normalised to 18S (B) in Hep G2 cells treated with
insulin (n=4)...............................................................................................................130
Figure 4.3-7. siRNA-mediated knockdown of insulin receptor A and B isoform.....131
Figure 4.3-8. (A) Insulin receptor PCR using linear pcDNA3.1 IR-B ......................132
Figure 4.3-9. Insulin receptor exon 10-12 fragment PCR in Hep G2 cells
overexpressing IR-A or IR-B isoform. ......................................................................134
Figure 4.3-10. AKT phosphorylation in cells overexpressing IR isoform A or B.....135
Figure 4.3-11. Mean + SE PCK1 mRNA expression presented as 2-∆Ct normalised to
18S in HepG2 cells overexpressing IR-A or IR-B after treatment with insulin (n=6).
....................................................................................................................................136
Figure 5.2-1. Box and whisker plots of expression data from all 30 arrays before and
after Quantile Normalisation......................................................................................149
Figure 5.2-2. TaqMan® Array Micro Fluidic Card. ..................................................153
Figure 5.4-1. Comparison of Illumina Microarray and TLDA RT-qPCR fold changes
of 43 genes that were found to be differentially expressed .......................................170
Figure 5.4-2 Mean ± SE INHBE, IGFB2, ATF3, BHLHE40, CDKN1A, and IL1RAP
expression presented as 2-∆Ct normalised to 18S in the sNGT (●, n=8) and the sT2DM
(○, n=8) group before and after RYGB surgery. .......................................................174
Figure 5.4-3. Mean ± SE CYP2C19, DBH, IGFALS, IGJ, LGALS4, PYROXD2, and
SAA1 expression (2-ΔCt) in the sNGT (●, n=8) and the sT2DM (○, n=8) before and
after RYGB surgery. ..................................................................................................175
Figure 5.4-4. (A) Mean + SE liver LGALS4 protein abundance (relative to actin) of
NGT (n=17) and T2DM (n=26) groups. See Appendix for complete western blot. .176
Figure 9.1-1. Figure 9.1-2. ENPP1 (110-130kda) western blot using protein extracted
from liver tissue biopsied from individuals at the time of RYGB surgery................275
Figure 9.1-3. IDE (118kda) western blot using protein extracted from liver tissue
biopsied from individuals at the time of RYGB surgery. ..........................................276
Figure 9.1-4. CEACAM-1 (160-180kda) western blot using protein extracted from
liver tissue biopsied from individuals at the time of RYGB surgery.........................276
Figure 9.1-5. Actin (43kda) western blot using protein extracted from liver tissue
xix
Figure 9.1-6. ENPP1 (110-130kda) western blot using protein extracted from liver
tissue biopsied from individuals both at the time of RYGB surgery and again at a
subsequent operation..................................................................................................277
Figure 9.1-7. IDE (118kd) western blot using protein extracted from liver tissue
biopsied from individuals both at the time of RYGB surgery and again at a subsequent
operation. ...................................................................................................................278
Figure 9.1-8 CEACAM-1 (160-180) western blot using protein extracted from liver
tissue biopsied from individuals both at the time of RYGB surgery and again at a
subsequent operation..................................................................................................278
Figure 9.1-9 Actin-1 (43kda) western blot using protein extracted from liver tissue
biopsied from individuals both at the time of RYGB surgery and again at a subsequent
operation. ...................................................................................................................279
Figure 9.2-1. Insulin Receptor β subunit (95kda) western blot using protein extracted
from liver tissue biopsied from individuals at the time of RYGB surgery................280
Figure 9.2-2. Insulin Receptor β subunit (95kda) western blot using protein extracted
from liver tissue biopsied from individuals both at the time of RYGB surgery and
again at a subsequent operation. ................................................................................280
Figure 9.2-3. AKT phosphorylation in HepG2 cells overexpressing IR-A (Hep G2 IRA) or IR-B (Hep G2 IR-B). Empty vector was used as control (Hep G2 EV). .........281
Figure 9.3-1. LGALS4 (36kda) western blot using protein extracted from liver tissue
biopsied from individuals in the sNGT and sT2DM group at the at the time of RYGB
surgery........................................................................................................................282
Figure 9.3-2. LGALS4 (36kda) western blot using protein extracted from liver tissue
xx
List of Abbreviations
Study cohort quick reference
NGT
Individuals who had normal glucose tolerance (n=19)
IGT
Individuals who had impaired glucose tolerance (n=9)
T2DM
Individuals who had type 2 diabetes (n=27)
sNGT
Individuals who had a second liver biopsy and normal glucose tolerance (n=8)
sT2DM
Individuals who had a second liver biopsy and type 2 diabetes (n=8)
Abbreviations
A2M
Alpha-2-macroglobulin
ACSL4
Long-chain-fatty-acid--CoA ligase 4
AGXT2L1
Ethanolamine-phosphate phospho-lyase
AHCY
Adenosylhomocysteinase
AP-1
Activator protein 1
ARNTL
Aryl hydrocarbon receptor nuclear translocator-like protein 1
ASS1
Argininosuccinate synthase
ATF3
Activating transcription factor 3
ATF5
Cyclic AMP-dependent transcription factor ATF-5
AXUD1
Cysteine/serine-rich nuclear protein 1
BHLHB2
Basic helix-loop-helix family, member e40 (BHLHE40)
βIRKO
β-cell insulin receptor knockout mice
BMI
Body mass index
BPD
Biliopancreatic diversion
BRSK2
Serine/threonine-protein kinase BRSK2
CEACAM-1 Carcinoembryonic antigen-related cell adhesion molecule 1
CCDS
Consensus coding sequence
CCRN4L
Nocturnin
CDKN1A
Cyclin-dependent kinase inhibitor 1
CNTNAP2
Putative uncharacterized protein CNTNAP2
CRP
C-reactive protein
xxi
CRYAA
Alpha-crystallin A chain
CXCR7
C-X-C chemokine receptor type 7
CYP1A2
Cytochrome P450 1A2
CYP2C19
Cytochrome P450 2C19
DBH
Dopamine beta-hydroxylase
DDIT4
DNA-damage-inducible transcript 4
DEPC
Diethylpyrocarbonate
DJB
Duodenal-jejunal bypass surgery
DM1
Myotonic Dystrophy type 1
DMEM
Dulbecco's Modified Eagle Medium
DMSO
Dimethyl sulfoxide
DUSP1
Dual specificity phosphatase 1
EEF1G
Eukaryotic translation elongation factor 1 gamma
ENO3
Enolase 3 (beta, muscle)
ENPP1
Ecto-nucleotide pyrophosphatase/phosphodiesterase
ERK1/2
Extracellular signal regulated kinase 1/2
FADS1
Fatty acid desaturase 1
FASN
Fatty acid synthase
FCS
Fetal calf serum
FIRKO
Fat insulin receptor knockout mice
FFA
Free fatty acid
FGL1
Fibrinogen-like protein 1
FNDC5
Fibronectin type III domain containing 5
FOSB
FBJ murine osteosarcoma viral oncogene homolog B
FOXO1
Forkhead box protein O1
GADD45B
Growth arrest and DNA damage-inducible protein GADD45 beta
GO
Gene ontology
HbA1c
Glycated hemoglobin
HBA2
Hemoglobin, alpha 2
HDL
High density lipoprotein
H19
H19, imprinted maternally expressed transcript
HEPACAM
Hepatocyte cell adhesion molecule
HOMA-IR
Homeostatic model assessment-insulin resistance
HPRT1
Hypoxanthine-guanine phosphoribosyltransferase
xxii
HPS5
Hermansky-Pudlak syndrome 5 protein
HSD17B11
Estradiol 17-beta-dehydrogenase 11
IDE
Insulin degrading enzyme
IDF
International Diabetes Federation
IFI6
Interferon alpha-inducible protein 6
IGF-IR
Insulin-like growth factor 1 receptor
IGF-1
Insulin-like growth factor 1
IGF-II
Insulin-like growth factor 2
IGFALS
Insulin-like growth factor-binding protein complex acid labile subunit
IGFBP2
Insulin-like growth factor-binding protein 2
IGFBP5
Insulin-like growth factor-binding protein 5
IGJ
Immunoglobulin J chain
IGLL1
Immunoglobulin lambda-like polypeptide 1
IL-1
Interleukin 1
IL32
Interleukin-32
IL1RN
Interleukin-1 receptor antagonist protein
IL1RAP
Interleukin-1 receptor accessory protein
iLIRKO
Inducible liver insulin receptor knockout mice
IRKO
Insulin receptor knockout mice
INHBE
Inhibin, beta E
IR
Insulin receptor
IRS
Insulin receptor substrate
IR-A
Insulin receptor isoform A
IR-B
Insulin receptor isoform B
IR-B:A
Insulin receptor isoform B to A ratio
JUN
Jun proto-oncogene
JUND
Transcription factor jun-D
KLF6
Kruppel-like factor 6
LAGB
Lap Band adjustable gastric banding
LEPR
Leptin receptor
LGALS4
Galectin-4
LDL
Low density lipoprotein
LDLR
Low density lipoprotein receptor
LIRKO
Liver insulin receptor knockout mice
xxiii
LPA
Apolipoprotein(a)
MAPK
Mitogen activated protein kinase
MAP2K1
Dual specificity mitogen-activated protein kinase kinase 1
MetS
Metabolic Syndrome
MIRKO
Muscle insulin receptor knockout mice
MLLT10
Protein AF-10
MPT
Multivariate permutation test
MTE
Metallothionein
MT1A
Metallothionein-1A
MT1F
Metallothionein 1F
MT1G
Metallothionein-1G
MT1H
Metallothionein 1H
MT1M
Metallothionein 1M
MT2A
Metallothionein-2
NAFLD
Non-alcoholic fatty liver disease
NIRKO
Neuron insulin receptor knockout mice
NR4A2
Nuclear receptor subfamily 4 group A member 2
OSTalpha
OSTalpha protein
PBS
Phosphate buffered saline
PCK1
Phosphenolpyruvate carboxykinase 1 (gene)
PEI
Polyethylenimine
PEPCK
Phosphenolpyruvate carboxykinase 1 (protein)
PI3K
Phosphoinositide 3-kinase
PLIN2
Perilipin-2
PVDF
Polyvinylidene difluoride
P4HA1
Prolyl 4-hydroxylase subunit alpha-1
PYROXD2
Pyridine nucleotide-disulfide oxidoreductase domain-containing protein 2
RCAN1
Calcipressin-1
RCT
Randomized controlled trial
REST
Relative expression software tool
RFTN1
RFTN1 protein
RHOB
Rho-related GTP-binding protein RhoB
RND3
Rho-related GTP-binding protein RhoE
RPL10A
60S ribosomal protein L10
xxiv
RT-qPCR
Reverse transcription quantitative real-time polymerase chain reaction
RVM
Random variance model
RYGB
Roux-en Y gastric bypass
SAA1
Serum amyloid A-1 protein
SDS
Serine dehydratase
SDS-PAGE
Sodium dodecyl sulphate polyacrylamide gel electrophoresis
SERPINE1
Serpine peptidase inhibitor, clade E, member 1
SERPINA11 Serpin A11
SGK1
Serine/threonine-protein kinase
SHBG
Sex hormone-binding globulin
SIK1
siRNA
Small interfering RNA
SLC2A3
Solute carrier family 2, facilitated glucose transporter member 3
SLC25A25
Calcium-binding mitochondrial carrier protein SCaMC-2
SLC29A4
Equilibrative nucleoside transporter 4
SLC3A1
Neutral and basic amino acid transport protein rBAT
SLC22A10
Solute carrier family 22 member 10
SNAI2
Zinc finger protein SNAI2
SPTBN1
Spectrin beta chain, non-erythrocytic 1
SQLE
Squalene monooxygenase
SRD5A2
3-oxo-5-alpha-steroid 4-dehydrogenase 2
SYT7
Synaptotagmin-7
TEMED
Tetramethylethylenediamine
TLDA
Taqman low density array
TMEM45B
Transmembrane protein 45B
TMEM154
Transmembrane protein 154
TRIB1
Tribbles homolog 1
TP53INP1
Tumor protein p53 inducible nuclear protein 1
UPP2
Uridine phosphorylase 2
UPR
Unfolded protein response
VBG
Vertical banded gastroplasty
VLDL
Very low density lipoprotein
xxv
xxvi
CHAPTER ONE: The pathology of type 2 diabetes
1
1.1 Type 2 diabetes mellitus: a disease for modern man
The need to feed is a fundamental driving force of life. Humans need a steady intake
of nutrients to provide the fuel needed to keep our central nervous system performing.
Because of the irregular availability of nutrients we have developed a mechanism
which promotes anabolism when nutrient supply exceeds demand or catabolism when
nutrient intake can not meet demand. Insulin secretion and action are integral parts of
this mechanism. Secreted by the pancreatic β-cells in response to nutrient intake,
insulin promotes carbohydrate uptake and storage for immediate use and change of
carbohydrate and protein to lipids for long term storage. For millennia this system has
served development of Homo sapiens well. But recently this powerful survival
mechanism has turned on humanity. Because of our ability to alter our environment in
shorter time frames, we have gone from having low nutrient availability and high
physical demand to abundant nutrient supply and relatively little physical demand.
Obesity and type 2 diabetes mellitus are now endemic and the world is grappling with
an imminent crisis. The global prevalence of type 2 diabetes is predicted to increase
from ~285 million in 2010 to ~552 million by the year 2030.1,
2
A chronic and
progressive disease, type 2 diabetes is a manifestation of a much broader underlying
disorder called Metabolic Syndrome (MetS). The International Diabetes Federation
(IDF) has defined MetS as consisting of abdominal obesity, impaired glucose
tolerance, hyperinsulinaemia, dyslipidaemia and hypertension.3 Although there have
been various definitions of MetS due to its association with both diabetes and
cardiovascular disease, a recent joint consensus statement has recommended
guidelines for clinical diagnosis.4 Nonetheless, the IDF classification of metabolic
syndrome is more predictive for development of impaired fasting glucose than
classifications by other organisations like the National Cholesterol Education
Program-Third Adult Treatment Panel (NCEP-ATPIII).5
The goal of conventional treatment for type 2 diabetes is strict glycaemic control that
is meant to prevent and/or reduce the risk of progression to end stage organ disease
such as myocardial infarctions, stroke, retinopathy and nephropathy.6
2
Even so, studies have shown that in the United States only 50% of individuals with
type 2 diabetes reach glycaemic control targets suggesting many go on to develop end
stage organ disease.7 Additionally, of the three large scale clinical trials (VADT,
ACCORD and ADVANCE)8-10 that used intensive therapy to lower glycated
hemoglobin levels (HbA1c), only one showed a 10% reduction of cardiovascular
events.10 The burden this will place on already struggling national healthcare systems
is enormous. In New Zealand, expenditure on treating type 2 diabetes is expected to
increase from 600 million in 2007 to 1.3 billion by 2016.11
A drastic shift in the way we approach treatment of type 2 diabetes and its
complications is needed. Rather than controlling the symptoms, treatments that
address the underlying pathogenesis must be developed. Unfortunately, despite
decades of intensive research into diabetes the exact mechanisms and precise
sequence of events behind its pathogenesis are still unclear. Although extensive use of
animal models and in vitro cell culture has elucidated the large majority of the
pathways involved in insulin signalling, the molecular lesions that cause the
characteristic insulin resistance that contributes to the development of type 2 diabetes
are still unknown.
However, the scientific community is becoming increasingly aware of what bariatric
surgeons have known since 1995.12 That bariatric surgery can cause remission of type
2 diabetes in 50-95% of individuals depending on procedure type.13,
14
Bariatric
surgery is so effective that in 2011 the IDF released a position statement
recommending bariatric surgery as a treatment for type 2 diabetes in morbidly obese
individuals.15
The work presented here is centred around the effects bariatric surgery has on type 2
diabetes. Specifically, gastric bypass surgery was used as a human model to provide
novel insights on molecular processes previously implicated in the pathogenesis of
type 2 diabetes and to search for as yet unknown mechanisms that may contribute to
this disease. The following literature review justifies the work done and it begins with
a description of the two main abnormalities present in type 2 diabetes, namely insulin
resistance and impaired insulin secretion.
3
1.2 Insulin resistance and impaired insulin secretion
Type 2 diabetes is thought to be the culmination of progressive insulin resistance and
loss of pancreatic β-cell function. In normal individuals a square wave increase in
plasma glucose concentration causes insulin to be released from the pancreas which
acts to increase glucose uptake and storage while decreasing endogenous glucose
production. Insulin resistance describes the impaired biological response to insulin
action and is characterized by progressively higher insulin concentrations, which is a
hallmark feature of type 2 diabetes.16, 17
Insulin resistance is one of the earliest detectable pathologies in individuals who
progress from normal glucose tolerance to impaired glucose tolerance and finally to
type 2 diabetes.18 It contributes to the characteristic fasting and postprandial
hyperglycaemia seen in type 2 diabetes by causing dysregulated hepatic glucose
production and impaired glucose uptake respectively.19,
20
However, even though
insulin resistance can be present 10-20 years prior to diabetes,18 fasting
hyperglycaemia does not manifest until there are defects in insulin secretion (Figure
1.2-1).20-23
Figure 1.2-1. Relationship between pancreatic beta cell function and insulin sensitivity. As the beta
cells fail to compensate for the decreased insulin sensitivity by increasing insulin secretion individuals
will develop impaired glucose tolerance and eventually type 2 diabetes (adapted from Kahn et al.
1993)24. IGT; Impaired glucose tolerance, NGT; Normal glucose tolerance, T2DM; type 2 diabetes mellitus.
4
Impaired insulin secretion is caused by abrogation of the frequency and amplitude of
insulin release in response to glucose. It has been observed in both individuals with
normal fasting plasma glucose and individuals with impaired glucose tolerance
suggesting it can be present well before development of overt hyperglycaemia.25
Consequently, failure of the β-cell to compensate for the insulin resistance is thought
to be a pre-requisite step for development of impaired glucose tolerance and type 2
diabetes.26-28
Insulin resistance is often used solely to describe the inability of insulin to regulate
glucose homeostasis. However, the actions of insulin are broad and require
consideration in order to fully appreciate the importance of insulin resistance and its
role in type 2 diabetes. The following sections will detail insulin action from secretion
through to binding of the insulin receptor and transmission of the signal. It will
outline the rationale of why the liver may have overriding importance in type 2
diabetes and the possible mechanisms behind insulin resistance.
1.3 Insulin action: from secretion to binding of the insulin receptor
The insulin signalling system can be divided into three sequential components; 1)
insulin production and secretion, 2) hepatic insulin clearance and 3) insulin binding of
the insulin receptor and signal transmission.
1.3.1 Insulin secretion
Insulin is produced in the β-cells of the pancreatic islets of Langerhans as a 5808 Da
polypeptide hormone. It is first synthesized as a single polypeptide preproinsulin and
shortly after is processed in the endoplasmic reticulum to proinsulin before it is sent to
the trans-Golgi network.25 Proinsulin consists of an A and B chain linked together by
disulfide bonds and a C-peptide bridge. Endopeptidases cleave off the C-peptide
bridge creating mature insulin which is packed into granules and stored on the plasma
membrane. Metabolic signals result in exocytosis of the granules and release of
insulin into the circulation. The main mechanism for eliciting insulin secretion is the
5
phosphorylation of glucokinase, which is mediated by glucose upon its entry into the
β-cells through the GLUT 2 channel.25
There are two distinct mechanisms involved in insulin secretion from the pancreas
into the hepatic portal vein. Firstly, insulin secretion occurs in discrete secretory
pulses every four minutes.29,
30
The oscillatory nature of insulin secretion has been
shown to be important in insulin action.31-33 Consequently, a decreased pulse mass is
thought to be a contributing factor to the impaired insulin secretion seen in type 2
diabetes.34-36
Insulin secretion is also biphasic in response to a glucose stimulus (Figure 1.3-1).37-39
The initial spike in glucose-induced insulin release is referred to as the rapid or firstphase. It is characterized by a rapid increase in insulin secretion (or insulin pulse
mass) which can last for up to 10 minutes.40 Continued exposure of the β-cell to
glucose leads to a slower but sustained second-phase of insulin secretion that occurs
within 10-20 minutes after glucose exposure and can last for several hours.40 It is
thought that insulin-containing granules exist as two functionally distinct pools that
are distinguished by their proximity to the plasma membrane.41 The pool closest to the
plasma membrane (<5% of mature insulin-containing granules) is thought to be
responsible for the glucose-induced first-phase insulin release,41,
42
whereas the
sustained second-phase is a consequence of stored insulin-containing granules and de
novo insulin synthesis. 41, 43, 44
First-phase insulin secretion is important in maintaining normoglycaemia and is
thought to be central to the metabolic shift from a fasting to a postprandial state.40
Studies have shown that it is required for the efficient suppression of hepatic glucose
production45-47 and may also prime peripheral insulin sensitive-tissues to increase
glucose uptake.39 An abrogated first-phase response has been shown to be associated
with impaired glucose tolerance48-51 and may be an early event in the development of
type 2 diabetes.40, 52, 53
6
Figure 1.3-1. Schematic diagram of biphasic insulin secretion after square wave increase in plasma
glucose concentration. Insulin secretion occurs in discrete secretory pulses and is oscillatory in nature.
A square wave increase in glucose concentration causes a rapid increase in insulin secretion (first
phase) followed by slower but sustained secretion (second phase) (adapted from Pratley and Weyer,
2001).40
Furthermore the suppression of hepatic glucose production is thought to be primarily
due to a decrease in gluconeogenesis,54 which suggests that the first-phase insulin
secretion has more of an effect on the liver as opposed to peripheral tissue.
Considering hepatic gluconeogenesis is a characteristic pathology in type 2 diabetes23
the importance of the liver as a key organ in regulating insulin action and glucose
homeostasis should not be overlooked.
1.3.2 Hepatic insulin clearance
Once released from the pancreas, insulin travels in pulses via the hepatic portal vein to
the liver where it binds to the insulin receptor and is either degraded or released into
the systemic circulation. The liver is the only organ which is exposed to insulin
concentrations that can range in amplitude between 1,000-5,000 pmol/l.55 The level of
insulin in the systemic circulation is only ~1% of that presented at the liver.55
Although the lower insulin concentration in the systemic circulation is due in part to
the five fold dilution of hepatic portal vein blood by the systemic circulation, the liver
has also been shown to have a key role in extracting insulin from the circulation.56
7
Insulin clearance at the liver is crucial to maintaining an appropriate systemic insulin
concentration.56 Accordingly, the liver is thought to respond minute by minute to
changes in insulin concentration by decreasing hepatic insulin clearance in response to
increased demand for insulin and vice versa.56 Hepatic insulin clearance has been
found to be diminished in obesity and type 2 diabetes.57-60 Consequently, it has been
suggested that diminished insulin clearance may be a potential compensatory
mechanism that maintains the high systemic insulin levels needed to maintain
normoglycaemia in insulin resistant states.61
1.3.3 Molecules involved in hepatic insulin clearance
Insulin clearance by the liver is a complex process that is a function of insulin
internalization and degradation. The primary mechanism for hepatic insulin uptake is
an insulin receptor-mediated process. However, insulin that is removed from the
circulation is not necessarily degraded, with a significant amount of receptor bound
insulin being released back into the circulation.62 Generally, 80% of total insulin in
the body is bound to the insulin receptors on the liver, which extracts ~50% of the
secreted insulin during first pass transit.56, 62, 63 In the last 15 years, major strides have
been made in uncovering some of the molecules involved in insulin internalization
and degradation. Although some of the functional characteristic of carcinoembryonic
antigen-related cell adhesion molecule 1 (CEACAM-1) and insulin degrading enzyme
(IDE) have been identified, their involvement in the pathological process of type 2
diabetes remains to be profiled.
CEACAM-1 is involved in mediating the internalization of the insulin-insulin
receptor complex.
64-66
In addition to its role in insulin internalization and cell-cell
adhesion it has also been found to suppress tumour growth in vivo
67
and to
downregulate the mitogenic effects of insulin. 68 Insulin binding to the insulin receptor
causes the IR tyrosine kinase to phosphorylate CEACAM-1.69 Surprisingly this action
is specific to IR as its close relative the insulin like growth factor-1 receptor does not
illicit a similar response68. Specifically it is the C terminus of the β subunit of the
insulin receptor that is required for insulin stimulated phosphorylation of CEACAM1.
8
Liver-specific overexpression of a dominant negative (S503A) mutant of CEACAM1,
which
is
phosphorylation-deficient,
impairs
insulin
clearance
causing
hyperinsulinaemia. This leads to secondary insulin resistance, abnormal glucose
tolerance, increased fatty acids and visceral adiposity.
65, 70
Likewise a CEACAM-1
null mutation (Cc-1- ) in mice causes impaired insulin clearance resulting in
hyperinsulinaemia, hepatic steatosis and visceral adiposity.64 Evidently, CEACAM-1
has an important role in insulin action via regulation of its internalization and
consequent hepatic clearance.
Hepatic insulin clearance is also dependent on IDE-mediated insulin degradation.
Insulin degradation is a regulated process that ultimately inactivates and removes
insulin from the circulation. Inhibitors of IDE have previously been proposed as a
potential anti-diabetic therapy.71 In animal models of diabetes (Goto-Kakazaki rats)
IDE was shown to be involved in the diabetic phenotype by contributing to insulin
resistance and hyperglycaemia through an impaired ability to degrade insulin.72
Accordingly, mice lacking IDE developed hyperinsulinaemia leading to progressive
insulin resistance and ultimately glucose intolerance,73 thus reflecting the progressive
nature of insulin resistance and type 2 diabetes. Consequently, in a prospective casecohort study, two IDE polymorphisms were associated with type 2 diabetes risk and
were found to exert influence on insulin degradation, secretion and sensitivity.74 Like
CEACAM-1, IDE is a major component of hepatic insulin clearance, which has
previously been implicated in the pathogenesis of type 2 diabetes.
1.3.4 Classical insulin target tissues and consequence of signal transmission
Insulin exerts its metabolic effects by binding to the insulin receptor. As a multipotent
hormone, insulin regulates several major biological processes which include glucose
metabolism, lipid metabolism and protein turnover. In humans, maintaining a constant
supply of glucose is crucial for continued biological functions due to the reliance of
the brain on glucose as its major fuel source. Organs like the brain and cells like
erythrocytes consume glucose evenly in fed and fasted states.75 Consequently,
maintaining normoglycaemia in the face of variable availability of glucose is achieved
by regulating glucose uptake and endogenous glucose production (Figure 1.3-2).
9
Figure 1.3-2. Direct effect of insulin on the three classical insulin target tissues of liver, muscle and fat.
In fasting conditions insulin levels are low resulting in increased glucose output from the liver. In a
postprandial state, glucose levels are increased leading to insulin secretion, inhibition of hepatic
glucose output and stimulation of muscle and fat glucose uptake (adapted from DeFronzo, 2009 and
Konrad et al. 2005).76, 77
The three classic target tissues of insulin action are liver, muscle and fat (Figure 1.32).75 In fasting conditions the liver releases glucose via glycogenolysis and
gluconeogenesis. In the postprandial state insulin levels rise which inhibits break
down of glycogen via inhibition of glycogenolysis in muscle and liver tissue in
addition to inhibiting gluconeogenesis in the liver.78 At the same time insulin
stimulates glucose uptake by muscle and fat tissue.75 In addition to stimulating
glucose uptake, insulin increases glucose storage as either glycogen or lipids in
various tissues. The liver stores both glycogen and lipids, whereas muscle and adipose
tissue store glycogen and lipid respectively.
10
Insulin also has indirect effects on glucose metabolism by decreasing the release of
gluconeogenic precursors via regulation of lipid metabolism and protein turnover.
Insulin inhibition of lipolysis at adipose tissue reduces the release of free fatty acids
(FFA) and glycerol which contributes to the inhibition of hepatic glucose production.
Likewise, insulin inhibition of proteolysis at muscle tissue decreases the release of
amino acids necessary for gluconeogenesis. 79, 80
1.3.5 Non-classical insulin target tissue
Although the classical insulin target tissues are the major regulators of insulin action,
other tissue such as the central nervous system, vascular cells and pancreatic β-cells
have also been found to have roles in insulin signalling.76 Insulin signalling in the
brain is thought to be involved in appetite suppression,81, 82 and has been shown to be
abrogated in obese, insulin-resistant, normal glucose tolerant individuals.83 In vascular
cells it is thought to stimulate vasodilatation and capillary recruitment thus enhancing
glucose delivery and increasing muscle glucose uptake.84 In the β-cell insulin has
been shown to activate transcription of both the insulin and glucokinase genes.85
1.4 Insulin resistance in the classical insulin target tissues
Insulin resistance is defined as a reduced biological effect for any given concentration
of insulin.86 The impact of insulin resistance extends far beyond its effect on glucose
metabolism and is one of the fundamental physiological changes that lead to the
cluster of abnormalities (type 2 diabetes, obesity, hypertension, dyslipidaemia,
cardiovascular disease) seen in metabolic syndrome X.17 The measurement of insulin
resistance in humans is a key component of managing metabolic syndrome and type 2
diabetes and is often approximated by methodologies that measure the biological
action of insulin (insulin sensitivity). Insulin sensitivity is a quantitative measure of a
specific action of insulin and is the reciprocal of insulin resistance.87
11
1.4.1 Measurement of insulin resistance
Although the different methodologies employed for measuring insulin resistance are
numerous and complex, they can be divided into two major groups.86 What follows is
a brief description of two most common techniques used.
Non-dynamic methodologies take measurements during the steady state and can be
based on glucose and insulin measurements alone. One such measurement is the
Homeostasis model assessment (HOMA), which takes measurements during fasting
conditions.88,
89
HOMA is a mathematical model that generates values for insulin
sensitivity from fasting plasma glucose and insulin concentrations.88 As HOMA is
based on the assumption of a feedback loop between the liver and the pancreas and is
used in the fasting state, where normoglycaemia is regulated by hepatic glucose
output, it is more indicative of hepatic insulin resistance. HOMA-IR values less than
or equal to 2.4 have been shown to be representative of an insulin sensitive, normal
glucose tolerant population.89-91
Dynamic measurements of insulin sensitivity are based on artificial disruption of the
steady state and evaluation of the return to the steady state. An example technique is
the euglycemic clamp technique.86 With this technique glucose levels are maintained
constant with a variable intravenous infusion and insulin levels are elevated using a
constant intravenous insulin infusion. Once the glucose levels are ‘clamped’ at a
predetermined concentration the insulin resistance is inversely related to the glucose
infusion rate necessary to maintain the required glucose concentration.92 Although the
euglycaemic clamp is the gold standard for measuring insulin sensitivity in vivo, its
execution is time consuming, and labour intensive.86 Consequently, because of its
simplicity, HOMA is often used to estimate insulin resistance. Estimates of insulin
resistance from HOMA correlated highly with estimates of insulin resistance using the
euglycaemic clamp in two different studies (Rs =0.88, p<0.000188, Rs =0.85,
p<0.0001)93 suggesting HOMA is a robust and appropriate method for estimating
changes in insulin resistance.
12
1.4.2 Insulin resistance in peripheral tissue
Insulin resistance in the periphery can affect glucose disposal and lipolysis. Muscle
insulin resistance is present when insulin-mediated glucose disposal through the
GLUT4 channel is reduced to the lowest quartile of control subjects.94 Muscle tissue
is responsible for 90% of glucose disposal in peripheral tissue.76,
95, 96
Because the
euglycaemic clamp technique uses the glucose disposal rate to estimate insulin
resistance, muscle tissue has been considered to be central to the development of type
2 diabetes. The progressive hyperinsulinaemia required to overcome the insulin
resistance at the muscle is also thought to play a role in the development of metabolic
syndrome.97 Furthermore, insulin resistance at adipose tissue leads to unregulated
lipolysis and an increase in circulating FFAs, which promotes hepatic glucose
output.98-101 This has lead some to suggest that the inability of insulin to inhibit
lipolysis may be a major contributor to the increased hepatic glucose output seen in
diabetes.102
1.4.3 Insulin resistance at the liver
Of the three classical insulin target tissues, the liver is the primary organ that directly
regulates metabolic homeostasis. Not only does it control insulin levels in the
periphery via hepatic insulin clearance, but it is also the major organ that regulates
normoglycaemia during fasting conditions103 and has a major role in lipid
homeostasis. Uncontrolled hepatic gluconeogenesis has been shown to contribute to
the fasting and overnight hyperglycaemia seen in individuals with type 2 diabetes.19,
20, 23, 104, 105
Following an overnight fast, the liver of normal individuals produces glucose at an
approximate rate of 2mg/kg per minute, whereas the liver of an individual with type 2
diabetes produces glucose at a rate of 2.5mg/kg per minute.19, 76 As the diabetic state
worsens there is a correlated increase between hepatic glucose production and fasting
plasma glucose concentrations, despite the progressive (2.5-3 fold) increase in insulin
concentration.76 Consequently, hepatic insulin resistance has been shown to adversely
affect the direct suppressive effect of insulin on hepatic glucose production in type 2
13
diabetes.106, 107 An abrogated first phase insulin response is also likely involved as it
has been associated with impaired suppression of hepatic glucose production.45, 46
Insulin resistance at the liver has multiple effects, of which, the ability to regulate
lipid homeostasis has become increasingly relevant with regards to its role in
metabolic syndrome. The dyslipidaemia that is often present in individuals with type 2
diabetes is characterized by hypertriglyceridaemia, raised LDL-cholesterol and a low
HDL cholesterol profile.108 The overstimulation of lipogenesis at the liver due to
hyperinsulinaemic conditions is thought to be a critical component of the
overproduction of VLDL particles seen in type 2 diabetes.109 Consequently, the
abnormal fat storage and ectopic fat deposition in other insulin target tissues has been
suggested to have a role in the progressive nature of insulin resistance.110
All three of the classical insulin target tissues have critical roles in regulating glucose
homeostasis (Figure 1-4.1). However, the question of which insulin signalling tissues
(if any) have overriding importance in the progression to type 2 diabetes is yet to be
answered. Although in vivo studies in humans are useful, they are ultimately limited.
The use of animal models allows researchers to target specific tissue and observe the
sequential molecular changes that occur as one passes from normal glucose tolerance
to type 2 diabetes.
Figure 1.4-1. A brief depiction of the key organs responsible for the unregulated hyperglycaemia seen
in type 2 diabetes. A combination of impaired insulin secretion and insulin resistance causes increased
glucose output in the liver, increased lipolysis in fat tissue and decreased glucose uptake in muscle
tissue. (Adapted from DeFronzo, 2009)76
14
1.5 Insulin target tissue relevance in regulating glucose homeostasis
Transgenic animal models have focused our understanding of the relevance of insulin
target tissues in glucose homeostasis. Although muscle and fat tissue still have a
fundamental role in the progression to type 2 diabetes, compelling data have
highlighted the importance of the liver, pancreatic β-cell and brain.111
1.5.1 Transgenic animal models of tissue specific insulin resistance
Studies by Kahn and colleagues have been useful in determining the importance of
specific insulin target tissue (Table 1-1).112-116 In these studies, tissue specific total
insulin resistance was created by ablating the insulin receptor using Cre-loxP
technology. Cre-loxP uses a bacterial recombinase (Cre) to excise a portion of DNA
that has been tagged by two recognition sequences (lox).117 In the case of the insulin
receptor gene, transgenic mice with lox sequences inserted either side of exon 4 of the
insulin receptor are bred with transgenic mice expressing Cre recombinase. Cre
recombinase expression is driven by a tissue specific promoter (e.g. albumin promoter
in liver tissue),116 resulting in offspring that have a missense hepatic insulin receptor
mRNA that codes for a non-signalling insulin receptor fragment.
Because of the importance of muscle glucose uptake in glucose homeostasis it was
one of the first tissues to be investigated. Surprisingly, muscle insulin receptor
knockout mice (MIRKO)114 did not have aberrant glucose homeostasis despite
elevated plasma triglycerides and FFA. The inability of MIRKO muscle tissue to take
up glucose was compensated by adipose tissue, which resulted in increased adiposity
and weight gain.
Similarly, fat insulin receptor knockout mice (FIRKO)112 also had an unexpected
phenotype. Not only did FIRKO mice have normal glucose homeostasis in addition to
having improved insulin sensitivity; they were also protected against obesity-induced
glucose intolerance and had increased longevity.
15
Table 1-1. Transgenic tissue specific mouse models of insulin resistance.
Insulin
Receptor
knockout
Muscle
(MIRKO)114
Insulin
Secretion
Insulin
action
Glucose
Homeostasis
Lipid
homeostasis
Obesity
Glucose
tolerance
Normal
Normal
Normal
Obese
Normal
Fat
(FIRKO)112
Normal
Improved
insulin
sensitivity
Normal↑
↑ FFA
↑
Triglycerides
↓Triglycerides
Lean*
Improved
LIVER
(LIRKO)116
Increased
Markedly
insulin
resistant
Hyperglycaemia
↓ FFA
↓Triglycerides
-
Severely
Impaired
Β-cell
(βIRKO)115
Impaired
Mildly
insulin
resistant
Normal
-
-
Impaired
Neuronal
(NIRKO)113
Increased
Normal
↑
Triglycerides
Obese
Normal
Mildly
insulin
resistant
* Protected against diet-induced obesity.
Liver insulin receptor knockout mice (LIRKO)116 showed severe glucose intolerance
and insulin resistance. They had mild fasting hyperglycaemia and marked
postprandial hyperglycaemia, which was thought to stem from the inability of insulin
to inhibit hepatic glucose production. LIRKO mice also exhibited increased insulin
secretion and impaired hepatic insulin clearance (due to loss of receptor-mediated
insulin internalization and degradation) resulting in marked hyperinsulinaemia.
Although this caused secondary whole body insulin resistance, as the mice aged the
impaired glucose tolerance resolved. This suggested even though the liver appears to
have more relevance than muscle and fat in insulin resistance, failure of pancreatic βcell function is still necessary for progression to overt type 2 diabetes.
Pancreatic β-cell insulin receptor knockout (βIRKO)115 mice are characterized by one
of the earliest defects found in individuals with type 2 diabetes, namely loss of
glucose stimulated first phase secretion. As with humans, loss of the first-phase
insulin response in βIRKO mice leads to impaired glucose tolerance that worsens with
age. This was later shown to be caused by the inability of insulin to increase
glucokinase gene transcription in the pancreas.85 The data from the βIRKO mouse
shows that insulin resistance at the β-cell may be as critical as hepatic insulin
resistance appears to be in the development of type 2 diabetes.
16
The brain has also been shown to have a critical role in regulating glucose
homeostasis. Neuronal insulin receptor knockout mice (NIRKO)113 have whole body
insulin resistance, hyperphagia and obesity. Likewise, decreasing insulin receptors in
the rat hypothalamus not only induced hyperphagia and insulin resistance but also
negatively affected the ability of insulin to regulate hepatic glucose output.118
1.5.2 Interplay between key insulin target tissues in progression to type 2
diabetes: does the liver have overriding importance?
The transgenic animal models of Kahn and colleagues give great insight into tissue
specific regulation of glucose homeostasis. One major conclusion from these animal
models is that although some tissues (liver and pancreas) appeared more relevant in
the pathogenesis of type 2 diabetes than others (muscle and fat), insulin resistance at a
single tissue is not enough to generate diabetes. Consequently, several animal models
have been developed that have defects in multiple tissues.
An animal model with targeted insulin resistance in both muscle and adipose tissue
showed impaired glucose tolerance and impaired insulin secretion, but did not
develop diabetes.119 Compounding β-cell insulin resistance with muscle insulin
resistance (βIRKO-MIRKO) improved rather than worsened glucose tolerance.120
These data suggest that peripheral insulin resistance, even with impaired insulin
secretion, is not enough to produce diabetes.
Simultaneous knockout of insulin receptors in liver and pancreatic β-cells; however,
generated markedly diabetic mice providing further evidence of the importance of
these tissues in type 2 diabetes.111 By the same token, animals that have whole body
knockout of the insulin receptor (IRKO) can be rescued from death by reconstituting
insulin receptor in certain tissues. Shortly after birth, IRKO mice develop severe
hyperinsulinaemia followed by β-cell failure, diabetic ketoacidosis and death.121
Reconstituting insulin receptor in the liver, pancreatic β-cells, and brain rescues IRKO
mice from neonatal death and prevents development of diabetes.122 Given these
points, it is becoming increasingly apparent that insulin resistance at these three
tissues is necessary for the development of type 2 diabetes.
17
Collectively, the data from the LIRKO, βIRKO and other animal models points to a
tantalizingly unified theory in which insulin resistance at the liver causes whole body
insulin resistance and ultimately type 2 diabetes. To test this hypothesis Escribano et
al. (2009)123 created an inducible liver specific insulin receptor knockout mice model
(iLIRKO). iLIRKO mice are born with functioning liver insulin receptor that was
gradually deleted resulting in progressive hepatic and peripheral insulin resistance.
Notably, peripheral insulin resistance developed as a secondary effect of the primary
hepatic insulin resistance. This phenotype is consistent with the hypothesis that
hyperinsulinaemia itself can cause insulin resistance in peripheral tissues.124
The progressive insulin resistance seen in iLIRKO mice was associated with
hyperinsulinaemia and increased β-cell mass; while eventual failure of the β-cells to
secrete enough insulin lead to uncontrolled diabetes. Admittedly, all insulin target
tissues are relevant with regards to insulin resistance. But, the progressive nature of
insulin resistance stresses the need to find the organ in which the genesis of insulin
resistance ultimately leads to type 2 diabetes. Evidence from the study by Escribano et
al. (2009) supports a hypothesis in which the liver may be the primary organ in the
pathogenesis of type 2 diabetes.
18
1.6 Insulin Signalling
Insulin signalling through the insulin receptor regulates a wide variety of biological
processes that are dependent on tissue type. In addition to the classical regulation of
glucose/lipid homeostasis, insulin is also responsible for regulating cell growth and
differentiation. Although animal studies with tissue specific ablation of the insulin
receptor are useful in defining tissue relevance in insulin-resistant states, in reality
complete insulin resistance is only seen in rare genetic diseases like leprechaunism
and type A insulin resistance syndrome.125 Most cases of insulin resistance are due to
a combination of factors that affect some, but not all outcomes of insulin action. To
understand how “selective” insulin resistance may play a role in type 2 diabetes it is
necessary to understand the molecular mechanisms behind insulin signalling.111
Intensive research into the insulin signalling pathway has elucidated a highly
integrated and complex network of signalling components. Moreover, most of the
major insulin signalling components are also involved in regulating signalling
pathways initiated by other biological stimuli.126 The specific signalling initiated by
insulin through what appear to be signalling hubs is likely mediated by multiple
isoforms of the same regulatory protein.127 Although insulin can elicit many
biological outcomes, the more established processes are presented below.
1.6.1 Insulin receptor
The insulin receptor (IR) is the first major component in the insulin signalling
pathway. A member of the receptor tyrosine kinase superfamily,128 the IR and its
close homologue the insulin-like growth factor-1 receptor (IGF-1R),129-131 are large
heterodimers that consist of two α and β subunits. The α-subunit (Mr 140 kDa), which
resides completely in the extracellular matrix, is linked to the intracellular β-subunit
(Mr 95 kDa) via disulfide bonds. Insulin binding to the α-subunit leads to
conformational changes that induce autophosphorylation of several tyrosine residues
on the β-subunit.132-134 Autophosphorylation of the β-subunit activates the receptor’s
protein tyrosine kinase, which activates intracellular substrates responsible for the
metabolic consequences of insulin signalling.
19
1.6.2 Insulin receptor substrate (IRS) proteins
To date, 11 intracellular substrates of the IR have been identified,127 six of which are
structurally similar and have been termed insulin receptor substrate proteins (IRS1-6).
135-140
Although the IR tyrosine kinase interacts with other intracellular targets
including Shc,141 Cbl,142 p62dok,143and Gab-1,144 the IRS proteins are responsible for
mediating activation of the two main pathways in insulin signalling. Namely, the
phosphatidylinositol 3-kinase (PI3K)-AKT pathway and the Ras-mitogen activated
protein kinase (MAPK) pathway.127 IRS-mediated signalling is particularly important
in regulation of glucose homeostasis in the liver.
IRS proteins contain a pleckstrin homology (PH) domain, a protein tyrosine binding
domain (PTB) and tyrosine residues that can be phosphorylated by the IR tyrosine
kinase.140 Phosphorylation of IRS tyrosine residues creates recognition sites for
intracellular molecules with a src-homology-2 (SH2) domain. Thus, IRS proteins
function as docking molecules in order to facilitate propagation of the insulin signal
throughout the cell.
The tissue distribution of IRS proteins varies considerably: IRS-1 and IRS-2 are
widely distributed,145 IRS-3 has predominant expression in liver and lung tissue,146
IRS-4 is primarily expressed in embryonic tissue136 while IRS-5/6 appear to have
limited tissue expression.135 Despite the high level of homogeneity between the IRS
proteins, data from animal studies suggests that IRS-1 and 2 may have increased
importance in glucose homeostasis. IRS-1 knockout mice have glucose intolerance
and peripheral insulin resistance,147,
148
whereas IRS-2 knockout mice develop
diabetes because of hepatic insulin resistance and lack of pancreatic β-cell
compensatory response.149, 150 Conversely, IRS-3151 and IRS-4152 knockout mice have
little to no perturbations in glucose homeostasis. These data suggest IRS function may
be dependant on tissue and subcellular location.
20
1.6.3 Negative regulation of IR activity
Left unchecked, insulin activation of the IR would lead to extreme metabolic changes.
Insulin-induced down regulation of IR is a well established feedback mechanism153-158
that regulates the strength of insulin signalling and is common in hyperinsulinaemic
states.159 However, the contribution insulin-induced down regulation of the IR has to
the pathogenesis of type 2 diabetes is unclear. Other mechanisms that effect negative
regulation of IR signalling include: protein tyrosine phosphatase-mediated
deactivation of the IR tyrosine kinase, steric blocking of IR-IRS interaction and
inhibition of insulin binding to the IR.
Of the protein tyrosine phosphatases, PTP1B is the most well studied.160 PTP1B acts
to dephosphorylate the IR tyrosine kinase thus terminating the insulin signal.161
Animals with liver specific deletion of PTP1B have improved insulin sensitivity and
glucose homeostasis162 leading some to suggest PTP1B inhibitors as a possible
treatment for type 2 diabetes.163 Similarly, suppressors of cytokine signalling-1 and 3
(SOCS-1 and 3) proteins are thought to be involved in insulin resistance through
inhibition of tyrosine kinase phosphorylation of IRS proteins.164 Notably, SOCS
proteins have been shown to be elevated in obesity,165 a state often associated with
insulin resistance.
Of particular interest is inhibition of insulin binding to the IR α-subunit by
ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1 also referred to as
plasma cell membrane glycoprotein 1 or PC-1).166 Increased ENPP1 protein levels
have been found in skeletal muscle,167 adipose tissue168 and skin fibroblasts169,
170
from insulin-resistant individuals. Subsequently, high levels of ENPP1 coincided with
decreased insulin receptor activity implicating ENPP1 in the pathogenesis of insulin
resistance and type 2 diabetes.171 Overexpressing ENPP1 in vitro170 and in vivo172 led
to impaired insulin signalling and action while selectively suppressing the levels of
ENPP1 in the liver of mice improved insulin sensitivity.173 Moreover, there is
compelling evidence suggesting a role for ENPP1 in genetic forms of insulin
resistance. Genotype-phenotype association studies have identified a gain of function
mutation (K121Q), which increases the inhibitory function of ENPP1 and may have a
role in insulin resistance and type 2 diabetes. 174, 175
21
1.6.4 PI3-K/AKT pathway
Phosphatidylinositol 3-kinase (PI3-K) is a pivotal kinase involved in mediating
metabolic outcomes of insulin signalling.176 It consists of a regulatory p85177 and a
catalytic p110 subunit,178 both of which occur in several isoforms.127 The p85 subunit
of PI3-K has two SH2 domains that interact with the phosphotyrosine residues on
IRS-1.179 Although PI3-K is involved in multiple biological process,180 production of
PI-3P, PI-(3,4)P2 or PIP2 and PI-(3,4,5,)P3 or PIP3, from the phosphorylation of
phosphoinositides at the 3-position, are the most important in insulin signalling.181, 182
PIP2 and PIP3 go on to bind the PH domains of other molecules thus activating them
or changing their subcellular location.183
Although PIP3 regulates three main classes of signalling molecules, the AGC family
of serine/threonine protein kinases is the most studied and pertinent to glucose
metabolism.182 Of the AGC kinases, phosphoinositide-dependant kinase 1 (PDK1) is
the best characterized and along with PIP3, is involved in activating a serine/threonine
kinase called AKT.127, 184 PIP3 activates PDK1, which phosphorylates and increases
the catalytic activity of AKT. But, because PDK1 resides at the plasma membrane,
PIP2 and PIP3 must first facilitate the interaction of PDK1 and AKT by recruiting it to
the plasma membrane via its PH domain.185 Once activated, AKT is thought to
mediate the majority of the hepatic metabolic outcomes stimulated by insulin.127
1.6.5 PI3-K/AKT-mediated regulation of glycogen synthesis
Glucose entering the cell through the glucose channels is converted to glycogen by
glycogen synthase (GS). Insulin activates glycogen synthase by promoting its
dephosphorylation via inhibition of glycogen synthase kinase 3 (GSK-3). Active AKT
phosphorylates and deactivates GSK-3, thus decreasing the rate of phosphorylation of
GS and increasing its activity.186 Insulin has also been shown to target protein
phosphatase 1 (PP1) which further increases the dephosphorylation and activation of
GSK-3. Rather than globally activating PP1, insulin is thought to stimulate PP1
activity that is localized to glycogen particles.187 Although there is no direct evidence
of the mechanisms behind insulin activation of PP1, inhibitors of PI3-K block this
effect suggesting the involvement of the PI3-K/AKT pathway.182
22
Figure 1.6-1. A brief depiction of insulin signalling and its outcomes in the hepatocyte. Insulin binding to the IR
at the hepatocytes regulates glucose and lipid homeostasis and cell growth and differentiation. Activation of the
tyrosine kinase catalyses the phosphorylation of proteins like the IRS and Shc. These proteins then interact with
signalling molecules through their SH2 domains resulting in activation of PI3-K and the Ras-MAPK cascade.
PI3-K activates AKT, which regulates two of the major mechanisms involved in glucose homoeostasis in the
liver. First, AKT phosphorylates and deactivates GSK-3, which stops it from phosphorylating GS thus
increasing its activity and glycogen synthesis. Second, AKT phosphorylates FOXO-1, thus preventing it from
entering the nucleus and activating transcription of gluconeogenic genes (G6P and PEPCK) thus decreasing
gluconeogenesis. Finally, although the exact mechanisms are unclear, insulin increase the levels of both
SREBP-1c transcript and its mature nuclear form. Nuclear SREBP-1c increases transcription of lipogenic genes
thus increasing lipogenesis. IRS: insulin receptor substrate; PI3-K: Phosphatidylinositol 3-kinase; MAPK:
mitogen activated protein kinase; GSK-3: glycogen synthase kinase 3; GS: glycogen synthase; FOXO-1:
forkhead transcription factor 1; G6P: glucose-6-phosphate; PEPCK: phosphoenolpyruvate carboxy kinase;
SREBP-1c: sterol regulatory element binding protein 1c. (adapted from Saltiel and Kahn, 2001)182
23
1.6.6 PI3-K/AKT-mediated regulation of hepatic glucose output
Insulin regulates hepatic glucose output through direct effects on gene expression188
and indirect regulation of substrate availability.182 Insulin inhibits transcription of
gluconeogenic genes that code for phosphoenolpyruvate carboxy kinase (PEPCK),189,
190
glucose-6-phosphatase191,
192
and fructose-1,6-bisphophatase, while increasing
transcription of genes that code for glycolytic enzymes such as glucokinase193 and
pyruvate kinase.194 Current evidence suggests that forkhead transcription factor 1
(FOXO-1)192,
195
and the transcriptional co-activator PGC-1196 are involved in
regulating gluconeogenesis in the liver. Specifically, AKT phosphorylates FOXO-1
which prevents it from entering the nucleus and activating transcription of genes that
code for glucose-6-phosphatase and PEPCK, an enzyme that catalyzes the rate
limiting step in gluconeogenesis.191, 197
1.6.7 PI3-K/AKT-mediated regulation of lipid homeostasis
In addition to the regulatory action on glucose homeostasis, insulin also has a critical
role in lipid homeostasis. Insulin deactivates hormone sensitive lipase (HSL) through
the PI3-K/AKT pathway thus inhibiting lipolysis in adipose tissue and depriving the
liver of fuel (FFA) for glucose production.198 In the liver, insulin promotes lipogenesis
by increasing transcription of lipogenic genes like acetyl-CoA carboxylase.199 In the
same way insulin’s effects on glucose metabolism are regulated by FOXO-1, sterol
regulatory element binding protein 1c (SREBP-1c) mediates many of insulin’s effects
on lipogenesis.200 Induction of SREBP-1c transcription appears to be dependant both
on AKT201-203 and liver X receptor activation,204 although the exact mechanism is
unknown. Unlike FOXO-1; however, SREBP-1c regulation of lipogenic gene
expression is a complex process and is subject to a sterol sensing element within the
cell. Insulin increases levels of SREBP-1c transcript and nuclear protein.205,
206
SREBP-1c transcripts encode a membrane bound precursor, which is held in the
endoplasmic reticulum by Insig proteins. Sterol depletion causes the SREBP cleavage
activating protein (SCAP) to transport SREBP-1c to the Golgi apparatus, where it
undergoes cleavage by two proteases. Once cleaved, the N-terminal fragment of
SREBP-1c is released and is transported to the nucleus, where it induces transcription
of lipogenic genes that promote triglyceride synthesis.
24
1.6.8 Ras/MAPK pathway
The Ras/MAPK pathway is the other major signalling network that can be stimulated
by insulin and is involved in mediating cell growth, differentiation and survival.
Insulin causes tyrosine phosphorylation of IRS-1, Gab1 and Shc, thus facilitating
binding of growth factor receptor binding protein (Grb-2).207 Grb2 recruits the guanyl
nucleotide exchange protein (SOS) to the plasma membrane resulting in the activation
of G protein Ras.207 Activated Ras induces a phosphorylation cascade that begins with
Raf and ends with activation of the Mitogen activated protein kinase (MAPK)/
Extracellular signal regulated kinases (ERK) pathway. The activated ERKs
phosphorylate various intracellular substrates and translocate to the nucleus to
phosphorylate transcription factors, like Elk-1, that promote gene expression. Studies
with Ras or SOS dominant negative mutants and with IRS-1 siRNA/antibodies have
shown that the Ras/MAPK pathway is important for insulin’s effect on cell growth
and DNA synthesis.208-211
1.7 Obesity and type 2 diabetes: the search for a common pathway
Genetic and environmental factors can act alone or in unison to cause insulin
resistance. Hereditary diseases of insulin resistance such as leprechaunism,212 RabsonMendenhall syndrome213,
214
and Type A insulin resistance125 are rare and often
involve mutations of the IR gene. Environmental factors such as obesity; however, are
commonly associated with insulin resistance and type 2 diabetes.215 Like diabetes,
obesity is a worldwide epidemic that shows no signs of abating. Recent estimates
show that in the United States, more than one third of adults (35.7%) and 17% of
children were considered obese.216 In New Zealand, 27.8% of adults and 8.3% of
children were classified as obese by two nationwide health surveys.217, 218
1.7.1 Obesity and β-cell dysfunction
Obesity has a complex relationship with insulin resistance and type 2 diabetes.
Indeed, ~25-30% of obese individuals remain metabolically normal retaining normal
levels of insulin sensitivity, blood pressure, lipid profile and inflammatory profile.21925
223
Further,
many
obese
insulin-resistant
individuals
do
not
develop
hyperglycaemia.215 This is because the pancreatic β-cells compensate for decreased
insulin sensitivity by increasing insulin secretion thus maintaining physiological
glucose concentrations. Type 2 diabetes is prevented as long as the β-cells can secrete
enough insulin to overcome the insulin resistance. Increased insulin secretion from the
β-cells appears to be mediated by both increased β-cell mass224, 225 and enhanced βcell function.226, 227 However at a certain point, β-cells fail to produce enough insulin
resulting in impaired glucose tolerance and ultimately frank hyperglycaemia.
228
A
recent review by De Fronzo and Abdul-Ghani (2011) has suggested that preservation
of 20-30% of β-cell function may prevent individuals from progressing to type 2
diabetes.229
Several lines of evidence suggest that β-cell failure has a genetic component. Only a
subset of individuals progress to type 2 diabetes, while abnormal insulin secretion has
been observed in first-degree relatives of individuals with type 2 diabetes.
33, 230
.
Longitudinal data from the Pima Indians, who have the highest prevalence of diabetes
than any other group in the world, shows that progression from impaired glucose
tolerance to diabetes coincided with β-cell failure.26 Although there is a genetic
component, obesity and the concomitant insulin resistance have a critical role to play
in diabetes progression.
1.7.2 Obesity and insulin resistance
The recent emergence of adipose tissue as a secretory organ has revolutionized the
way researchers consider obesity and metabolic disease. Adipose tissue can influence
glucose homeostasis by release of metabolites (FFA), hormones (leptin, adiponectin,
resistin, visfatin) and proinflammatory cytokines (IL-1, TNFα, MCP-1).231,
232
Consequently, obesity has been associated with an increase in the production of many
of these products, with adipokine-mediated release of proinflammatory factors being
implicated in insulin resistance.233-235
Although the mechanisms behind the affect of obesity on insulin resistance are
incompletely understood, it has been suggested that it may be due to the distribution
of fat. Fat distribution can be generally classified as lower body, abdominal
26
subcutaneous (underneath the skin), overall coverage or visceral fat (located among
the organs of the abdomen).236 Visceral fat is thought to increase the risk of insulin
resistance because of its anatomical location and venous drainage of secreted factors
into the hepatic portal vein.233, 236, 237 Consequently, it has been suggested that high
visceral fat content leads to increased FFA deposition in the liver and insulin
resistance.238-240 However, other studies have presented data to the contrary
suggesting there is no direct causative link between visceral fat tissue and hepatic
insulin resistance.241-243 In individuals with lipodystrophy242 and a fatless mouse
model,243 hepatic insulin resistance and marked liver steatosis was present despite a
lack of visceral adiposity. Although intra-abdominal fat mass has been observed to
have a detrimental affect on insulin sensitivity by increasing both ectopic lipid
deposition and inflammation, its exact role in mediating insulin resistance is unclear.
Obesity associated insulin resistance is often caused by external factors that interfere
in the insulin signalling pathway. Several mechanisms have been implicated in the
pathogenesis of type 2 diabetes, which are: glucotoxicity, lipotoxicity, oxidative
stress, endoplasmic reticulum stress and inflammation. Each is a distinct mechanism
and they are thought to equally contribute to the development of the disease.
However, a unifying pathway that can be sequentially followed from inception of
insulin resistance to type 2 diabetes has yet to be elucidated. Nonetheless, recent
advances in the field have increased the significance of three of these mechanisms
(lipotoxicity, ER stress and inflammation) and their contributions to development of
insulin resistance.
1.7.3 Lipotoxicity
It is now generally accepted that increased levels of FFAs have a detrimental effect on
insulin action244 as they have been shown to cause insulin resistance and impaired
insulin secretion. Obese individuals with or without type 2 diabetes have increased
levels of FFAs.245, 246 A short term (48 hours) sustained increase of FFAs was shown
to cause impaired insulin secretion in response to mixed meals and intravenous
glucose.247 The evident toxic effects of elevated FFAs led to the coining of the term
‘Lipotoxicity’.248
27
Lipotoxicity in muscle tissue impairs the insulin signalling cascade and insulinmediated glucose uptake causing insulin resistance.249-251 Specifically, lipid
accumulation, rather than the circulating plasma lipid concentration, is responsible for
affecting insulin action.252,
253
Non-alcoholic fatty liver disease (NAFLD) is the
clinical term for lipid accumulation in the liver. NAFLD can range from simple
steatosis to steatohepatitis, advanced fibrosis and cirrhosis, and is commonly found in
individuals with obesity and type 2 diabetes.254 Fabbrini et al. (2009) demonstrated
that intra-hepatic fat was associated with the metabolic complications of obesity.255
Reversing NAFLD with a strict hypocaloric diet in individuals with type 2 diabetes
improved liver insulin sensitivity (but not peripheral) and normalized fasting
hyperglycaemia.256 Similarly, a recent study suggested strict reduction of dietary
energy intake could reverse the underlying abnormalities present in individuals with
type 2 diabetes and insulin resistance. Notably, decreased pancreatic and liver lipid
accumulation was associated with improved hepatic insulin sensitivity, insulin
secretion (first phase response) and normalization of fasting hyperglycaemia.257
1.7.4 Molecular mechanism behind lipotoxicity
Increased FFA delivery to tissue results in increased intracellular content of fatty acid
metabolites such as diacylglycerol (DAG), fatty acyl-coenzyme A (fatty acyl-CoA)
and ceramides, which can act as secondary messengers in key signalling pathways.244
Consequently, high levels of ceramides and DAG in liver and muscle tissues were
associated with development of insulin resistance.258
DAGs mediate insulin resistance via an interaction with several molecules belonging
to the protein kinase C family (PKC).244,
259
The PKC family belongs to the AGC
kinase superfamily and includes: conventional PKCs (cPKC: α, βI, βI I, γ), novel
PKCs (nPKC: ε, θ, η, δ) and atypical PKCs (aPKC: ζ and λ).244 Specifically, DAGmediated activation of PKCθ has been shown to cause phosphorylation of IRS-1. This
in turn blocks insulin-induced phosphorylation of IRS-1 and inhibits signalling
through the PI3-K/AKT pathway.260,
261
PKCε has also been shown to impair IR
tyrosine kinase activity in liver tissue. Rats that were fed a high fat diet developed
hepatic steatosis and hepatic insulin resistance, which was associated with increased
28
levels of PKCε. Subsequent knockdown of PKCε reversed the fat-induced effects on
insulin signalling and protected the rats from hepatic insulin resistance.262
Evidently, DAG-mediated activation of nPKCs can impair insulin signalling and
action in key tissues like the liver and muscle. Similarly, ceramide accumulation in
tissue is thought to mediate insulin resistance by inhibiting activation of AKT.263, 264
Ceramide accumulates in membrane microdomains where it recruits PKCζ.265,
266
PKCζ phosphorylates AKT and prevents it being bound by PIP3. 267 This inhibits the
actions of insulin through the PI3-K/AKT pathway causing insulin resistance with
regards to glucose metabolism.
1.7.5 Inflammation
Chronic low-grade systemic inflammation has a complex but critical role in insulin
resistance.235 Initial observations of elevated acute-phase reactants and cytokines in
obesity and type 2 diabetes implied that stimulation of the innate immune response
had a pathological role in this disease.268-271 Elevated levels of interleukin-1β (IL-1β),
IL-6 and CRP were found to be predictive of diabetes development.272,
273
Furthermore, serum concentrations of IL-1 receptor antagonist (LI-1RA) were
increased in obesity274 and continue to increase in the years preceding diabetes.275
Over the last decade, obesity has been characterized by a broad array of inflammatory
factors that can impact on insulin signalling.234 TNFα provided the first molecular link
between obesity, inflammation and insulin resistance when it was found to be
increased in adipose tissue of insulin-resistant rodents.276 Similarly, increased TNFα
levels were observed in adipose tissue of obese individuals270, 271 and muscle tissue of
insulin-resistant individuals with type 2 diabetes.277 Subsequent studies of the
mechanism behind TNFα-mediated insulin resistance suggested that TNFα induces
serine phosphorylation (Ser 307) of IRS-1 thus inhibiting insulin signalling through
the IR.278, 279
Although the exact mechanisms behind obesity-induced inflammation are currently
unclear, recruitment of adipose tissue macrophages (ATMs) by chemokine signalling
is thought to play a key role.280-283 ATMs secrete many inflammatory factors and they
were found to be responsible for most of the adipose tissue associated TNFα
29
secretion.281 Overexpression of monocyte chemo attractant protein-1 (MCP-1) in
adipose tissue increased ATM infiltration leading to development of insulin resistance
and hepatic steatosis; whereas expression of a dominant negative mutant of MCP-1
improved insulin resistance in db/db mice and mice fed a high fat diet.284
Furthermore, research over the last five years has implicated T lymphocytes, natural
killer T cells, mast cells, and B cells in obesity-related insulin resistance.282
However, a recent study by Meijer et al. (2011) has shown that primary adipocytes
can secrete cytokines and chemokines independently of macrophages.285 This led the
authors to suggest that adipocyte metabolic dysfunction is the primary event leading
to chronic inflammation, which can be aggravated by ATM infiltration. Nonetheless,
ATM infiltration was shown to be a strong predictor of insulin sensitivity in obese
individuals.286 Collectively, these data suggest that infiltration of adipose tissue by
ATMs is a part of a broader inflammatory response associated with hepatic lipid
accumulation and insulin resistance.
1.7.6 Molecular mechanisms behind inflammation-mediated insulin resistance
Inflammation contributes to insulin resistance primarily through inhibition of down
stream insulin signalling. Jun-terminal kinase (JNK) and inhibitor of nuclear factor-κβ
kinases (IKK) are activated by stress stimuli present in metabolic disease (cytokines
and FFAs) and have been implicated in insulin resistance.287, 288
IKK can inhibit insulin signalling both directly and indirectly. Mice which lack IKK
were partially protected from high-fat diet or lipid infusion-induced insulin
resistance.289,
290
Consequently, IKK was shown to inhibit insulin signalling by
disrupting the IR-IRS-1 interaction via direct phosphorylation of IRS-1 on serine
residues.288, 291 Not surprisingly, mice with hepatic IKK knockout retain liver insulin
sensitivity, although they do develop muscle and fat insulin resistance in response to a
high fat diet.292 IKKβ also indirectly impairs insulin signalling by activating NF-κβ
via phosphorylation of inhibitor of NF-κβ (Iκβ). NF-κβ is a transcription factor that
stimulates production of multiple inflammatory factors, including TNFα and IL-6.293
Transgenic mice overexpressing IKK in liver tissue exhibited hyperglycaemia,
marked hepatic insulin resistance and moderate systemic insulin resistance, which was
30
associated with an increased liver production of IL-6, IL-1 and TNFα.294 Although
activation of the IKK pathway in liver tissue appears to have a critical role in obesityinduced insulin resistance, it may not be as important in muscle tissue.292, 294, 295
JNKs belong to the MAPK family and regulate transcription via activation of the
activating protein-1 (AP-1) complex.296 JNK is now thought to have a central role in
obesity-induced insulin resistance.297 Like TNFα and IKKβ, JNK activation by
cytokines and free fatty acids leads to phosphorylation of IRS-1 which inhibits IRS-1
interaction with the IR and thus insulin signalling.279,
288, 298
Overexpressing a
dominant-negative type JNK in liver of obese diabetic mice markedly improved
insulin sensitivity and glucose tolerance, suggesting the JNK liver pathway may have
a critical role in regulating glucose homeostasis.299 Consequently, inhibition of JNK
signalling has been suggested as a viable treatment for diabetes.300
Another key inflammatory system involved in obesity-induced insulin resistance is
the IL-1 pathway. Elevated circulating levels of IL-1β are associated with an
increased risk of developing type 2 diabetes, whereas blockage of IL-1 signalling in
individuals with type 2 diabetes improved glycaemic control and β-cell function.301
Excessive IL-1β production is thought to be regulated by FFAs,302 with recent work
suggesting this effect is mediated by the inflamasome.303 The IL-1β pathway is
similar to IKK in that it affects insulin signalling in multiple ways. IL-1β has toxic
effects on pancreatic β-cells and has been implicated in β-cell deterioration caused by
glucotoxicity in type 2 diabetes, suggesting a role in impairment of insulin
secretion.304 It was found to inhibit insulin signalling in adipose cells by inducing
down regulation of IRS-1 gene expression.305 Furthermore, IL-1β stimulates
proinflammatory actions through the IL-1 receptor that in turn recruits the IL-1
receptor accessory protein. This receptor complex initiates a signalling cascade that
results in the activation of multiple inflammatory factors, which include, JNK, IKK
and AP-1, all of which have been shown to inhibit insulin signalling.306 Accordingly,
animal models that lack IL-1β receptor were protected against high fat diet-induced
insulin resistance.307
31
1.7.7 The unfolded protein response: a link between lipotoxicity and
inflammation
The emergence of lipotoxicity and obesity-related inflammation has refocused the
way the scientific community looks at metabolic disease and type 2 diabetes.
Although we are aware of some of the major factors involved in these two processes,
there are still many unanswered questions. It is interesting to note that lipid infusion
or high fat diets are a common method of producing insulin resistance and type 2
diabetes in animal models. Conversely, animals that had ablation of inflammatory
factors were protected against insulin resistance, while treatment of humans with
drugs that target inflammatory pathways improves insulin sensitivity. The apparent
intimate relationship between lipotoxicity and inflammation points to a unifying
mechanism involved in mediating the pathological processes of insulin resistance.
Endoplasmic reticulum (ER) stress is emerging as a key process in obesity and
diabetes and may fill this role.308
The ER is responsible for protein folding, maturation and trafficking, and as such is
central in many intracellular biological processes. Stressful conditions (like
hyperinsulinaemia) cause overproduction of proteins, thus increasing the amount of
unfolded proteins in the ER lumen, which leads to consequent initiation of the
unfolded protein response (UPR).309
Activation of the UPR leads to reduced protein synthesis, increased protein
degradation and production of chaperones that mediate protein folding. Insufficient
reduction of ER stress by the UPR leads to compromised cell functionality and
apoptosis. The actions of the UPR are mediated by three ER-membrane associated
proteins that regulate three mechanistically distinct UPR pathways: PKR-like
eukaryotic initiation factor 2α kinase (PERK), inositol requiring enzyme 1 (IRE1),
and activating transcription factor-6 (ATF6).309 In the absence of ER stress, these
three proteins are inactive due to sequestration via binding of BiP/GRP78.310
Accumulation of unfolded proteins in the ER disassociates BiP/GRP78, inducing
activation of PERK and IRE1 and translocation of ATF6 to the Golgi apparatus.309
The UPR has been linked to inflammation via the JNK and IKK pathways.311,
312
During ER stress, the IRE1 arm of the UPR was shown to promote inflammatory
32
signalling through activation of JNK.313 Similarly, activation of the ATF6 arm of the
UPR was also associated with NF-κβ activation.314 Interestingly, induction of ER
stress in macrophages increased levels of proinflammatory factors such as TNFα and
IL-6, which was shown to be mediated by free cholesterol activation of the IKK and
JNK pathways.315 Activation of these inflammatory pathways and induction of both
cytokines was reliant on cholesterol trafficking to the ER, suggesting a possible link
between lipotoxicity and UPR-mediated inflammation.
Current data suggests the ER is an integral organelle in the regulation of glucose and
lipid metabolism. Hepatic gluconeogenesis was shown to be regulated by CREBH, an
ER-bound transcription factor.316 Consequently, animal models have shown that
disrupting the PERK pathway of the UPR resulted in hypoglycaemia associated with
defective gluconeogenesis.317,
318
Interestingly, inhibiting PERK signalling in mice
reduced insulin sensitivity in muscle and adipose tissue, suggesting a possible
crosstalk mechanism between the liver and peripheral insulin target tissue.319
ER-mediated regulation of lipid homeostasis is emerging as a possible link between
lipotoxicity and insulin resistance. As discussed previously, stimulation of lipogenesis
occurs via insulin-mediated SREBP-1c protein activation, which resides on the ER.320
Interestingly, inhibiting the UPR in liver tissue inhibited SREBP-1c cleavage and
expression, thus reducing hepatic triglyceride and cholesterol levels and improving
insulin sensitivity.321 The effects of the UPR on lipid metabolism are thought to be
regulated by IRE1α and ATF6-mediated pathways. IRE1α represses expression of key
metabolic enzymes, including those involved in triglyceride biosynthesis and lipid
secretion.322 Although mice that have complete ablation of liver IRE1α are not
significantly different from wild type littermates, they do develop profound lipid
accumulation after induction of ER stress with chemical agents.322 Similarly, ATF6
knockout mice are phenotypically normal until chemical induction of ER stress causes
lipid droplet accumulation and hepatic steatosis.323 Taken together, these studies
suggest that the actions of the UPR are critical in regulating glucose and lipid
homeostasis.
However, there is no clear consensus on whether the UPR directly interferes with
insulin signalling or is simply a cellular response to metabolic changes.244
Nonetheless, ER stress has been implicated in obesity-related insulin resistance in
33
animal models and humans. Insulin resistance in leptin-deficient ob/ob mice was
associated with activation of the UPR, which led to inhibition of insulin of signalling
via JNK-mediated phosphorylation of IRS-1.324, 325 Similarly, reduction of ER stress
was observed in individuals who had substantial weight loss after bariatric surgery.326
Consequently, treating obese mice that had type 2 diabetes with chemical agents that
reduce ER stress normalized hyperglycaemia, restored insulin sensitivity and resolved
NAFLD.327 A similar effect was observed in humans,328 although others have
suggested that these agents have confounding effects on lipogenesis329 and thyroid
hormone regulation of energy expenditure.330
Although there is compelling evidence supporting an important role for the UPR in
insulin resistance, much is still not known regarding the three key pathway of the
UPR. Evidently, certain components of the UPR regulate lipid (IRE1α and ATF) and
glucose metabolism (PERK), while activation of the UPR may have role in NAFLD.
1.8 A potential model detailing progression from insulin resistance
to type 2 diabetes
An important feature of type 2 diabetes and its associated complications is the
progressive nature of the disease. A brief look at some of the mechanisms behind the
factors involved in insulin resistance reveals several feed-forward mechanisms that
become progressively exacerbated as time passes on.
1.8.1 Lipotoxicity and inflammation: a feed-forward mechanism
FFAs and other metabolites like glucose are thought to have a role in activating the
inflamasome.303 Several studies have demonstrated FFA activation of toll like
receptors,331,
332
which can induce production of NF-κβ and other inflammatory
factors.333 Treatment of macrophage cells with palmitate (FFA) was shown to induce
expression of TNFα and IL-1β among others.334 Many of these cytokines can further
activate the JNK and IKK pathways in a feed-forward manner, thus exponentially
increasing the effect of the inflammatory response.
34
On the other hand, several cytokines were observed to have a lipolytic effect in
adipose tissue thus increasing circulating FFAs. Although inflammation is thought to
have a regulatory role in lipid homeostasis, the underlying mechanisms are unclear.
IL-6 and TNFα were shown to stimulate lipolysis in adipose tissue335-338 and while IL6 was shown to increase hormone sensitive lipase (HSL) mRNA expression, it did not
have an effect on HSL protein activity.337 TNFα is also thought to indirectly promote
lipolysis by decreasing expression of perilipin (PLIN), a protein involved in forming
lipid droplets that protect triglycerides from lipases.339-341 Accordingly, lack of PLIN
expression was associated with markedly increased lipolysis.342, 343 Recently, a newly
identified lipid droplet associated protein, fat specific protein 27 (FSP27), was found
to be down regulated by TNFα, IL-1β and IFNγ.344 Like PLIN, FSP27 is a lipid
droplet stabilizer and decreasing its expression mirrors the lipolytic effects of
TNFα.344 Taken together, these data suggest a self-feeding cycle that leads to
exacerbated lipotoxicity and inflammation.
The development of insulin resistance thereafter appears to be a matter of time and
although the exact sequence of events that lead to type 2 diabetes are yet to be
determined, several animal models provide clues as to the possible mechanisms.
1.8.2 Primary hepatic insulin resistance in animal models
Recently, several experiments using microarray technology have confirmed variable
hepatic gene expression consistent with a key role of inflammation in high fat dietinduced insulin resistance. In insulin-resistant, hyperglycaemic mice, genes that
regulate inflammation were up regulated early and were consistently elevated.
Additionally, variable expression of genes that regulate lipid homeostasis were
concomitant with dysregulated lipid metabolism.345-347 Notably, Kleeman et al. (2010)
used the same method to investigate variable gene expression in liver, muscle and fat
tissue in a time resolved manner.348 In mice fed a high fat diet there was a rapid up
regulation of genes involved in carbohydrate (only liver tissue) and lipid (liver and fat
tissues) metabolism, whereas these genes were gradually down regulated in skeletal
muscle tissue.348
35
Interestingly, Kleeman et al. (2010) observed that a high fat diet-induced insulin
resistance primarily in the liver (week six) and then in adipose tissue (week 12), while
skeletal muscle remained sensitive. The mechanisms behind this are currently unclear
although inflammation in adipose tissue is thought to shift toward the liver resulting
in hepatic steatosis.349 Recent evidence presented by Nov et al. (2013) has suggested a
possible IL-1β crosstalk mechanism between adipose tissue and the liver.350 IL-1β
levels were elevated preferentially in portal blood of high-fat diet mice and were also
linked to regulation of hepatic gluconeogenic flux.
The manifestation of early hepatic insulin resistance and subsequent peripheral insulin
resistance is consistent with animal studies done by Kahn and colleagues.116,
123
Despite the methodological differences (hepatic IR ablation versus high-fat diet), in
both cases primary hepatic insulin resistance seems to be a precursor for development
of whole body insulin resistance. A recent study on the pathological properties of
hepatic ER stress revealed a possible role for the UPR in mediating cross talk between
the liver and peripheral organ insulin sensitivity.319
Mice fed a high fat diet developed hepatic steatosis and insulin resistance that was
associated with augmented signalling through the PERK arm of the UPR. Inhibition
of PERK arm signalling decreased hepatic glucose production, but was also
associated with impaired insulin sensitivity in muscle and adipose tissue. This was
suggested to be partially due to increased levels of circulating IGFBP-3.319
Interestingly, iLIRKO mice, which develop peripheral insulin resistance in muscle
tissue after primary hepatic insulin resistance due to IR ablation, also had elevated
levels of IGBP-3.123 Collectively, these data suggest that a compensatory mechanism
that attempts to control the abnormal production of glucose in the liver may be
inadvertently causing peripheral insulin resistance in other tissue via IGFBP3.
36
1.8.3 The liver as a key organ in the progression to type 2 diabetes
Insulin signalling at the liver may be a principal process involved in the development
of type 2 diabetes. In 1992, Dennis McGarry was the first to suggest that
hyperinsulinaemia can cause peripheral insulin resistance due to over-stimulated
lipogenesis at the liver.124 McGarry hypothesized that overproduction of insulin by the
pancreas may cause hyperinsulinaemia that drives hepatic lipogenesis and VLDL
synthesis. Consequently, the increased flux of triglycerides from the liver would
accumulate in muscle tissue causing insulin resistance and mild hyperglycaemia. This
would further stimulate insulin secretion from the pancreas causing even greater
hyperinsulinaemia eventually leading to β-cell failure and diabetes.124 More than a
decade later, our understanding of the pathogenesis of type 2 diabetes has increased
considerably and McGarry’s opinion that lipid homeostasis has a crucial role in type 2
diabetes turned out to be prophetic.
The fundamental characteristic behind McGarry’s hypothesis is the multi-stimulatory
nature of insulin action. The ability of insulin to regulate glucose homeostasis, lipid
homeostasis (particularly at the liver), and cell growth naturally led to the concept of
“selective insulin resistance”.351 Several animal models have provided evidence for
the importance of selective insulin resistance in type 2 diabetes. Lipodystrophic and
ob/ob mice, which have hyperglycaemia and hyperinsulinaemia, were found to be
insulin resistant in the FOXO-1 pathway. However, insulin sensitivity was maintained
in the SREBP-1c pathway, which caused increased fatty acid synthesis and
triglyceride accumulation in the liver.352
The LIRKO mouse, characterized by absolute ablation of hepatic insulin signalling
through the IR, is a prime example of selective insulin signalling. In these mice,
insulin cannot suppress gluconeogenesis leading to hyperglycaemia, but by the same
token insulin also cannot activate SREBP-1c resulting in low levels of hepatic
triglycerides and circulating VLDL. Interestingly, the hyperglycaemia seen in LIRKO
mice resolves with age due to increased β-cell hyperplasia. The absence of diabetes in

Dennis McGarry (1940-2002), PhD, was a scientist who made a series of groundbreaking
contributions to mammalian metabolism. His seminal paper published in Science entitled “What if
Minkowski Had Been Ageusic?” revolutionized how researchers think about type 2 diabetes and led to
the eventual identification of lipotoxicity as a key process in insulin resistance.
37
these mice was suggested to be due to the lack of lipotoxic effects on the pancreatic βcell caused by the non-existent insulin-mediated lipogenesis at the liver.351 β-cell
failure is thought to be critical in obesity-related progression to type 2 diabetes.215
Although the mechanisms governing this process are currently an active field of
research a definitive mechanism that leads to β-cell failure is yet to be identified.
Inflammation,235 lipotoxicity,248, 353 ER stress354 and glucotoxicity355, 356 have all been
implicated in impairing β-cell function.
1.8.4 From insulin resistance to type 2 diabetes: a potential hypothesis
Taken together, the evidence presented over the last 30 years allows us to form a
potential hypothesis that explains the pathology behind type 2 diabetes (Figure 1.8-1).
Increased levels of metabolites like lipids (lipotoxicity) and glucose (glucotoxicity),
often seen in obesity, lead to inflammation in adipose and other insulin sensitive
tissue (Figure 1.8-1A). Lipotoxicity and inflammation feed into each other increasing
the potential for ectopic lipid deposition and exacerbation of the immune response,
both of which have been implicated in mediating hepatic insulin resistance. Selective
hepatic insulin resistance leads to increased lipotoxicity through over-stimulated
lipogenesis and initiation of peripheral insulin resistance resulting in development of
periodic hyperglycaemia (Figure 1.8-1B).
38
Figure 1.8-1. A hypothesized model describing the sequence of events leading to development
of Type 2 diabetes. (A) Increased levels of metabolites like lipids (lipotoxicity) and glucose
(glucotoxicity) lead to inflammation in adipose and other insulin sensitive tissue. (B) Selective
hepatic insulin resistance leads to increased lipotoxicity through over-stimulated lipogenesis
and initiation of peripheral insulin resistance resulting in development of periodic
hyperglycaemia. (C) This accelerates insulin secretion from the pancreas causing further
lipotoxicity and inflammation and initiating a cycle of progressive hyperinsulinaemia. (D)
Ultimately, the β-cells fail resulting in frank diabetes.
39
This accelerates insulin secretion from the pancreas causing further lipotoxicity and
inflammation and initiating a cycle of progressive hyperinsulinaemia (Figure 1.8-1C).
Symptoms of metabolic syndrome like hypertension and dyslipidemia start to develop
resulting in secondary pathologies like cardiovascular disease. Eventually the
pancreatic β-cells start to fail leading to development of abnormal glucose tolerance
(glucotoxicity) which with the combination of lipotoxicity and inflammation impairs
β-cell function. Ultimately, this results in postprandial and fasting hyperglycaemia
and frank diabetes (Figure 1.8-1D). In conclusion, selective insulin resistance
observed at the liver leads to impaired regulation of glucose homeostasis and overstimulation of lipogenesis and may be central to the progressive nature of type 2
diabetes. Investigating the selective nature of insulin signalling at the hepatocyte is
critical to understanding this disease.
1.9 Insulin receptor isoforms and their role in selective insulin
signalling
Identifying the regulatory steps at which insulin signalling pathways diverge is needed
for a complete understating of the pathogenesis of type 2 diabetes. Even though major
strides have been made in understanding the mechanism behind the initiation of
insulin resistance, there is still a lack of understanding of how insulin affects its broad
array of signalling. Insulin’s ability to affect different outcomes in specific tissue is
likely due to the many isoforms of key molecules in the insulin signalling network.
Although post-receptor signalling is currently under extensive investigation by the
scientific community the possibility of selective signalling through the IR isoforms
has been neglected. Considering the only aspect of the insulin signalling pathway
unique to insulin action is the IR, a closer look at the two isoforms of the molecule is
warranted.
The IR gene consists of 22 exons and 21 introns located on chromosome 19.129, 130, 357
Alternative splicing of exon 11 generates two different protein isoforms; IR-A
(without exon 11) and IR-B (with exon 11).
358
Exon 11 consists of 36 bp that code
for a 12 amino acid insert (residues 717-728) in the C terminus of the hormone
binding α-subunit358.
40
Alternative splicing of exon 11 is an evolutionary conserved process in mammals and
is likely a major component of insulin-IR mediated metabolic signalling.359 Splicing
of the IR is thought to occur through the inclusion/exclusion of alternative exons
model.360,
361
Regulatory elements that enhance or silence exon inclusion/exclusion
are often present in the alternatively spliced exon or in the flanking introns.362 These
regulatory
elements
interact
with
trans-acting
proteins
that
cause
exon
inclusion/exclusion, with differences in levels and activity of these proteins resulting
in tissue specific alternative splicing.362-364
Recently, advances have been made in understanding the mechanisms regulating
alternative splicing of the IR pre-mRNA. Regulatory sequences were found in intron
10 and exon 11 of the IR and have been shown to mediate both positive and negative
regulation of IR splicing.360, 361 SRp20 and SF2/ASF were shown to bind enhancing
sequences on exon 11, while CUG-BP1 was shown to bind silencing sequences on the
same exon. Consequently, in vitro overexpression of SRp20 and SF2/ASF increased
exon 11 inclusion and IR-B expression, whereas overexpressing CUG-BP1 caused
exon 11 skipping and increased IR-A expression. Although the signalling significance
of exon 11 is unknown, alternative splicing of the IR varies in different tissues and
diseases.
IR is expressed in a wide variety of tissues, with highest concentrations in insulin
target tissue. The two IR isoforms are usually co-expressed at varying levels that
depend on tissue type. In insulin target tissues such as liver, muscle and fat, IR-B
expression is considerably higher than IR-A expression.365 The relative IR-B:A
expression ratio in human liver was previously shown to be 9.8, while adipose and
muscle cells have relative ratios of 3.18 and 1.15 respectively.366 Conversely, IR-A
expression predominates in developmental tissue such as fetal cells367 and in
pathological conditions like cancer.368-370
41
1.9.1 Metabolic vs mitogenic outcomes of IR signalling
The IR isoforms show functional differences with regard to insulin binding,
internalization and signalling. This is not surprising. The predominant expression of
IR-B isoform in metabolically active tissue suggests that insulin signalling through
IR-B is primarily metabolic; whereas predominant expression of IR-A in
developmental and cancerous tissue suggest a strong mitogenic component. Although
the different functions of the IR isoforms are generally accepted, the underlying
molecular mechanisms are still unclear.
Early work on the functional characteristics of the two isoforms showed that IR-A has
a twofold higher affinity for insulin than the IR-B isoform.365, 371, 372 IR-A was also
found to have a faster insulin internalization rate.373, 374 Indeed, data suggests that IRA may be the primary isoform involved in CEACAM-1-mediated internalization of
the insulin-IR complex.375 Co-expression of IR-A and CEACAM-1 in NIH 3T3
fibroblasts significantly increased insulin internalization and degradation in
comparison to cells co-expressing IR-B and CEACAM-1. On the other hand, IR-B
isoform was shown to have stronger activation of the receptor tyrosine kinase
suggesting it is more efficient at propagating the insulin signal throughout the cell.374,
376
Kosaki and colleagues (1993 and 1995) were the first to provide a genuine link
between IR-B isoform and metabolic signalling. In a series of experiments they
showed that dexamethasone-treated Hep G2 cells had a switch in isoform expression
towards IR-B. The increase in IR-B expression was associated with increased insulin
sensitivity for insulin-mediated glucose metabolism and gene expression.377, 378
Mitogenic signalling through IR-A isoform was first evidenced by its high affinity
binding of insulin like growth factor II (IGF-II).367 IGF-II is a growth promoting
hormone that is highly homologous to insulin like growth factor I (IGF-1) and insulin,
all of which can signal through the IR and IGF-1R. IGF-II binding to the IR has been
shown to stimulate cell proliferation during mouse embryonic development.379,
380
Consequent studies investigating IGF-II interaction with the IR isoforms
demonstrated that IGF-II bound IR-A with an affinity close to that of insulin and
stimulated cell growth and proliferation.367, 381, 382
42
These data are consistent with predominant expression of IR-A in developmental
tissue.367, 380 Indeed, IR-A interaction with the GLUT2 channel may be necessary for
glucose uptake in neonatal hepatocytes,383,
384
suggesting IR-A mediates growth
promoting effects in early development.
Recently, it has been shown that proinsulin can also bind IR-A with a high affinity,
thus stimulating the mitogenic pathway through ERK/p70s6K activation. Although
proinsulin showed low metabolic signalling through IR-A, it was almost as equipotent
as insulin in stimulating cell proliferation.385 Given the ability of IGF-II and
proinsulin to stimulate mitogenic pathways through IR-A, it is not surprising that
increased IR-A expression has been observed in several cancers. IR-A expression has
been shown to be increased in breast cancer368, ovarian cancer386, leiomyosarcoma387
and thyroid cancer.388
As tissues age the pattern of IR isoform expression switches from IR-A to IR-B,
particularly in insulin sensitive tissue. Although IR-A is still expressed in adult tissue,
the exact mechanisms behind differential signalling through the two IR isoforms are
yet to be completely elucidated. Nonetheless, several studies have given insight into
the possible roles of IR isoforms in insulin target tissue. In the pancreas, insulin
signalling through the IR-A was found to induce transcription of the insulin gene
through the PI3-K/p70s6K pathway; whereas insulin signalling through IR-B-induced
transcription of β-glucokinase through the PI3-K/AKT pathway.85 The same group
went on to suggest that the molecular basis for the selective signalling may be due to
localization of the two isoforms to different areas of the plasma membrane, based on
the presence or absence of exon 11.85, 389
Recently, Giudice et al. (2011) suggested that differential signalling is dependant on
the internalization and localization of the IR isoforms. Using confocal and structured
illumination microscopy, they showed that internalization of the insulin-IR-A
complex is significantly faster than that of the insulin-IR-B complex. 390 Moreover, it
was observed that IR-A endocytosis was associated with stronger activation of the
ERK proteins and consequent greater stimulation of gene transcription through the
AP-1 complex. Conversely, IR-B internalization was slower and was associated with
greater activation of AKT in comparison to IR-A.
43
Ligand-receptor complexes have been shown to still be able to phosphorylate
downstream mitogenic effectors from endosomes within the cell.391-394 This has led
some to suggest that insulin’s mitogenic signalling may occur through the IR-A
isoform once it has been internalized into the cell.390 Blocking IR internalization
results in inhibition of the Ras/MAPK pathway with no effect observed on IR
autophosphorylation or IRS/AKT activation.395, 396 Taken together, these data suggest
a possible mechanism behind the selective signalling through the IR isoforms: IR-A
has a faster internalization rate thus leading to greater induction of the mitogenic
pathway from within the cell. In contrast, the extended presence of IR-B at the cell
surface leads to stimulation of the PI3-K/AKT pathway and regulation of metabolic
homeostasis.
1.9.2 Hybrid receptors
The regulation of insulin signalling is further complicated by the formation of hybrid
receptors (HRs). Due to the homologous nature of insulin and IGF-1 receptors,129, 358
cells expressing both receptors can form HRs between the IR isoforms and IGF1R.397-399 The formation of HRs is thought to occur in the endoplasmic reticulum and
is governed by a random assembly pattern that is dependent on the relative abundance
of the two receptors.400-403 Although the biological significance of the HRs is yet to be
revealed, several studies have identified ligand-hybrid receptor interactions and their
possible role in disease. Initially, IR/IGF-1R receptors were shown to bind IGF-I and
IGF-II with a high affinity and insulin with a low affinity (Figure 1.9-1).398,
404, 405
Further isoform specific investigations showed that IR-A and IR-B could form HR’s
with equal efficiency.404, 406 IR-A/IGF-1R receptors were able to bind to both IGFs
and insulin, whereas IR-B/IGF-1R receptors had high binding affinity for IGF-I, a low
affinity for IGF-II, and an insignificant interaction with insulin (Figure 1.9-1).406
44
Figure 1.9-1. Schematic diagram of the different insulin receptor hybrids and ligands. IR-A has a faster
internalization rate thus leading to greater induction of the mitogenic pathway from within the cell. In
contrast, the extended presence of IR-B at the cell surface leads to stimulation of the PI3-K/AKT
pathway and regulation of metabolic homeostasis. IR-A homodimers have a high affinity for both
insulin and IGF-II, whereas IR-B homodimers only have a high affinity for insulin. Although IR-A/IRB heterodimer receptors retain their affinity for insulin, they gain an increase in affinity for IGF-II that
is comparable to IR-A homodimers. IR-A/IGF-1R hybrid receptors bind to both IGFs and insulin,
whereas IR-B/IGF-1R hybrid receptors have a high binding affinity for IGF-I, a low affinity for IGF-II,
and an insignificant interaction with insulin (adapted from Jensen and DeMeyts, 2009).
Similarly, IR isoforms have also been shown to form hybrid receptors (i.e. IR-A/IRB).397 In the same way IR/IGF-1R hybrids have high affinity for growth promoting
ligands, so do IR isoform hybrids. Although IR-A/IR-B receptors retain their affinity
for insulin, they gain an increase in affinity for IGF-II that is comparable to IR-A
homodimers (Figure 1.9-1).397 Because hybrid receptor formation increases the
affinity for growth promoting ligands like IGF-II, they have been implicated in
carcinogenesis.400,
405
By the same token, due to the ligand binding properties of
hybrid receptors, their role in insulin resistance and type 2 diabetes has been
investigated, but with inconclusive results.
45
Although some have suggested increased IR/IGF-1R formation in adipose and
skeletal muscle tissue is associated with type 2 diabetes;407, 408 others have shown that
hybrid receptor overexpression is not seen in individuals with insulin resistance.409
Nonetheless, increased formation of hybrid receptors that reduce the number of
insulin binding sites may have a deleterious effect on insulin signalling and cannot be
discounted.
1.9.3 Insulin receptor and its isoforms in diabetes: an incomplete story
The role of IR protein concentration in type 2 diabetes is unclear. It is generally
accepted that chronic hyperinsulinaemia can induce IR downregulation.154-156 Ramos
et al. (2006) demonstrated that Grb10 mediates insulin-induced IR downregulation via
the proteasomal degradation pathway.157 Accordingly, IR content was found to be
decreased in several diseases associated with hyperinsulinaemia like insulinoma410
and myotonic dystrophy type 1 (DM1).411 Obese individuals who underwent bariatric
surgery had an increase in IR protein content concomitant with a decrease in fasting
plasma insulin and an improvement in insulin sensitivity.412 However, whether
insulin-induced IR downregulation is a feedback mechanism to compensate for
hyperinsulinaemia or has a pathological role in insulin resistance is yet to be
elucidated.
Altered expression of the two IR isoforms in key metabolic tissues may also
contribute to the pathogenesis of insulin resistance and type 2 diabetes. Considering
the important role of alternative splicing in regulating biological process, it is not
surprising that it may have a role in disease.413 It has been suggested that 50-60% of
mutations involved in hereditary disease affect the splicing mechanism.414 Although
aberrant IR splicing has been implicated in other diseases,415, 416 the regulation of IR
splicing and its role in insulin resistance and type 2 diabetes is controversial. Several
studies have associated increased IR-B expression in muscle and adipose417, 418 tissues
with insulin resistance and type 2 diabetes.417-422 Conversely, some have noted a
marked increase in relative IR-A expression in muscle tissue of an individual with
type 2 diabetes;423 whereas others have shown there to be no difference in expression
of IR isoforms in muscle and adipose tissue between individuals with and without
type 2 diabetes424, 425 or obesity.426
46
Evidently, there is no clear consensus for the role of IR splicing in type 2 diabetes,
which may be due to variation in methodology and/or variability in disease
progression.
Even so, increased expression of IR-A isoform has been linked with insulin resistance
in individuals with DM1.415 Cells cultured using skeletal muscle from individuals
with DM1 had decreased insulin sensitivity with respect to insulin-induced glucose
uptake and glycogen incorporation. Interestingly, other muscular disorders such as
Duchenne muscular dystrophy (which does not exhibit hyperinsulinaemia or insulin
resistance),427 limb girdle muscular dystrophy and facioscapulohumeral muscular
dystrophy do not show aberrant IR splicing.415 These data suggest a link between
hyperinsulinaemia and increased expression of IR-A isoform. Furthermore, elevated
expression of IR-A in cancer and a rising awareness of the link between
hyperinsulinaemia and carcinogenesis428 provides circumstantial evidence for a
relationship between hyperinsulinaemia and increased IR-A expression. Considering
the different functional characteristics of the two IR receptor isoforms, their roles in
the pathogenesis of type 2 diabetes still remain to be defined. In particular, there is
little data on the regulation of IR splicing at the liver, a key organ in regulating insulin
action.
1.10 Bariatric surgery: A model of type 2 diabetes remission
Type 2 diabetes has long been considered a chronic and incurable disease. However,
almost 50 years ago it was first reported that gastrectomy could improve and in some
cases induce remission of diabetes.429 In 1995, Pories and colleagues observed
remission of type 2 diabetes in 146 morbidly obese individuals within six days of
gastric bypass surgery, which was sustained for up to 14 years.12 It is now generally
accepted that bariatric surgery, a type of gastrointestinal surgery originally designed
to induce weight loss, is an effective treatment for type 2 diabetes.430 So much so that
the IDF has recommended bariatric surgery as an option for treating morbidly obese
individuals with the disease.15
47
Traditionally, bariatric surgery has been divided into three separate categories: 1)
restrictive procedures that cause weight loss by limiting food intake, 2) malabsorptive
procedures that cause weight loss by interference with digestion and absorption and 3)
mixed surgery procedures that are a combination of the restrictive and malabsorptive
procedures.
431, 432
Although there are many different types of bariatric procedures,
three are typically used worldwide. These include, laparoscopic adjustable gastric
banding (LAGB), bilio-pancreatic diversion (BPD), and the Roux-en-Y gastric bypass
(RYGB). Although other procedures like the duodenal-jejunal bypass (DJB), ileal
interposition and sleeve gastrectomy are becoming increasingly popular, the LAGB
and RYGB procedures are the most common.430-432
The LAGB is an example of a restrictive surgical procedure and is relatively simple
when compared to other types of bariatric surgery. In this procedure the superior
portion of the stomach is encircled with a saline-filled band, which restricts entry of
food into the stomach and thus causes early satiety and weight loss (Figure 1.10-1A).
The BPD; however, is complex and involves manipulation of the digestive tract. The
BPD causes food to bypass the small intestine (including the duodenum, jejunum and
a portion of the proximal ileum) in addition to diverting bile and pancreatic secretions
to the terminal segment of the ileum (Figure 1.10-1B). This ensures that bile and food
are only mixed in the last 50-100 cm of the small bowel thus causing drastically
reduced nutrient adsorption.
Figure 1.10-1. Typical bariatric surgical procedures. (A) Laparoscopic adjustable gastric band (LAGB).
(B) Bilio-pancreatic diversion (BPD). (C) Roux-en-Y gastric bypass (RYGB). Adapted from Dixon
(2012).430
48
The RYGB is a mixture of procedures and involves restriction of the stomach volume
and bypass of the excluded stomach and some small bowel (Figure 1.10-1C). A
surgical stapler is used to reduce the stomach volume to a < 30cc gastric pouch. This
pouch is completely separate from the remainder of the stomach and is anastomosed
to the jejunum in a Roux-en-Y fashion. The remainder of the bowel is reconnected to
the alimentary limb via an entero-entero anastomosis. Consequently, the RYGB
procedure causes food to bypass ~95% of the stomach, and a significant portion of the
small bowel (including the duodenum and a short portion of the jejunum).431, 432
1.10.1 Clinical outcomes of bariatric surgery
Traditional bariatric surgeries have been repeatedly observed to cause reduction in
cardiovascular, cancer and diabetes-related mortality.433-438 However, there has been
some controversy in the literature with regards to the quality of the data linking
bariatric surgery and improvement in diabetes. Lack of randomized controlled trials
and differences in methodologies (different types of bariatric surgery performed),
severity of diabetes and measurement of glycaemic parameters have made direct
comparisons of studies problematic.
Nonetheless, several systematic reviews and meta-analysis have shown a strong
association between remission of type 2 diabetes and bariatric surgery. A systematic
review of the Cochrane database concluded that surgery is more effective than
conventional treatment for weight loss and was associated with improvement in type 2
diabetes and dyslipidaemia.439 Another series of systematic reviews and meta-analysis
provided similar conclusions.13, 14 The first meta-analysis was published in 2004 and
involved 134 studies and 22,094 individuals in which remission of diabetes was
defined as persistent normoglycemia without diabetic medications.13 With these
parameters and other biochemical measures, Buchwald and colleagues reported a 77%
overall remission rate of type 2 diabetes after bariatric surgery (Table 1-2).
49
Table 1-2. Metabolic disease state after traditional bariatric surgery. Data adapted
from systematic review and meta-analysis by Buchwald et al. (2004). Data in brackets
is updated remission rates from Buchwald et al. (2009).
LAGB
RYGB
BPD
%EWL
47%
62%
70%
Total
population
61%
Remission of type 2 diabetes
48%
(56%)
84%
(80%)
98%
(95%)
77%
(78%)
Resolution of hypertension
43%
68%
83%
61%
Improvement of hyperlipidemia
59%
97%
99%
79%
Improvement of
hypercholestorolemia
78%
95%
87%
71%
Improvement of
hypertriglyceridemia
77%
91%
100%
82%
LAGB- laparoscopic adjustable gastric banding, BPD- bilio-pancreatic diversion, RYGB-roux-en-Y
gastric bypass, %EWL- % Excess weight loss.
There was also a drastic improvement in nonglycaemic effects with approximately
61% of individuals resolving hypertension while 79% had an improvement in
hyperlipidaemia (Table 1-2). However, there are some caveats to this meta-analysis as
most of the studies included were retrospective and typically had a follow up of 1-3
years. The second meta-analysis by Buchwald et al. (2009), which was an update on
the earlier one, included 621 studies with 135,246 individuals and had a similar
overall diabetes remission rate of 78%, while 62% remained in remission after 2 years
of surgery (Table 1-2).14 The Swedish Obese Subjects (SOS) study is an ongoing (up
to 20 years) case control study that has high quality long term data on individuals who
had bariatric surgery and matched obese controls that underwent regular therapy.436-438
The surgery group had lower incidence rates of diabetes and hypertriglyceridemia at
both the 2 year and 10 year follow up,436 in addition to having sustained weight loss
and decreased mortality.437, 438
Until very recently, only one randomized control trial (RCT) had investigated the
efficacy of bariatric surgery versus conventional treatment for type 2 diabetes.440 This
RCT compared the LAGB to standard therapy in 60 individuals with BMI values of
30-40 kg/m2.
50
Remission of type 2 diabetes (defined by a fasting glucose level < 7 mmol/l and an
HbA1c < 6.2) was observed in 73% of individuals who had LAGB and standard
therapy versus 13% of individuals who received standard therapy alone. In 2012, the
results of two RCT provided further evidence that bariatric surgery can be more
effective than conventional or intensive treatment alone.
In the first RCT, individuals with a history of diabetes (≥ 5 years) or glycated
hemoglobin ≥7.0% were randomly assigned to receive conventional medical therapy
or undergo RYGB or BPD.441 Remission of type 2 diabetes was defined as a fasting
glucose <6.5 mmol/l and a glycated hemoglobin of <6.5% in the absence of diabetic
mediation. At 2 years, 75% of individuals in the RYGB group and 95% of individuals
in the BPD group had remission of diabetes. Conversely, no individuals in the
conventional medical therapy group had remission of diabetes. In the second RCT,
Schauer et al. (2012) investigated the efficacy of RYGB or sleeve gastrectomy versus
intensive medical therapy alone in 150 obese individuals with uncontrolled type 2
diabetes.442 After 1 year, the primary endpoint (glycated hemoglobin ≤ 6%) was
achieved in 12% of individuals in the medical therapy group versus 37% in the sleeve
gastrectomy group and 42% in the RYGB group. The RCT studies by Mingrone et al.
(2012)441 and Schauer et al. (2012)442 present strong evidence for the superiority of
bariatric surgery in treating diabetes and its symptoms over medical therapy alone.
1.10.2 Defining the normalization of glucose homoeostasis after bariatric surgery
The phenomenal outcomes of bariatric surgery have lead to controversy within the
scientific community as to what qualifies as a cure for a chronic disease like diabetes.
Due to the lack of long-term data and level 1 evidence some authors have questioned
the permanency of glycaemic normalization after bariatric surgery.443, 444 Furthermore,
terms like “resolution” or “remission” have been used subjectively and defined by
variable glycaemic measures by different authors.
Because of the exponential increase in articles exploring bariatric surgery as a
treatment for type 2 diabetes, a consensus article by Buse et al. (2009)445 sought to
clarify the definition of treatment outcomes (Table 1-3).
51
Table 1-3. Treatment outcomes for type 2 diabetes as defined by Buse et al. (2009).
Remission
Partial
HbA1c Fasting plasma
glucose
<6.5% 5.6-6.9 mmol/l
Complete
<6.5%
<5.6 mmol/l
Treatment
Duration
No drug treatment or ongoing
procedures
≥ 1 year
No drug treatment or ongoing
procedures
≥ 1 year
The authors recommend the term “remission” be reserved for individuals who
normalize glucose homeostasis for at least 1 year without ongoing drug treatment or
procedures. They also suggested that prolonged complete remission (≥ 5 years) may
be considered a “cure” or “resolution” of type 2 diabetes. However, as the 5 year
period was arbitrarily chosen due to lack of long term data, the authors acknowledged
the risk of relapse in a subset of the population. Consequently, the terms “remission’
and “resolution” will be used in this thesis as defined by Buse et al. (2009).445
1.10.3 Weight -independent anti-diabetic effects of RYGB surgery
Particular bariatric procedures have a higher rate of diabetes remission. This is evident
in Buchwald’s (2009)14 meta-analysis where remission of diabetes after RYGB or
BPD surgery is close to double that observed after LAGB (Table 1-2). Furthermore,
diabetes remission occurs rapidly after RYGB surgery, typically within days to weeks
of surgery.12 Such rapid remission is not observed after restrictive bariatric procedures
(LAGB, VBG), which are thought to improve glucose homeostasis mainly by
inducing weight loss over an extended period of time.436, 440, 446 In a study of 1160
individuals undergoing RYGB surgery, one third of those with diabetes were
normoglycaemic without diabetic medication within 3 days of surgery.447
Similarly, a study by Wickremesekera et al. (2005) observed rapid remission of
diabetes after RYGB surgery which was associated with a dramatic improvement in
insulin sensitivity.448 Increasing evidence suggests that manipulation of the
gastrointestinal tract can cause weight independent anti-diabetic effects. Initial
development of experimental bariatric surgeries like the duodenal-jejunal bypass
(DJB) by Rubino and Marescaux (2004) provided compelling evidence that the
RYGB surgery causes remission of diabetes by mechanisms other than weight loss.449
52
Non-obese rats with type 2 diabetes that underwent DJB showed marked and durable
improvement in glucose homeostasis regardless of caloric intake or weight loss. Other
studies that used the same animal model reported similar findings.450-452
Subsequent observational studies comparing RYGB surgery with other bariatric
surgeries or weight loss treatment provided further evidence for weight independent
mechanisms of diabetic remission. Laferrére et al. (2008) showed that despite
equivalent weight loss, individuals who had RYGB surgery had a greater
improvement in glucose homeostasis than those who achieved weight loss by
dieting.453 Similarly, a prospective study following individuals who had equivalent
weight loss after RYGB or LAGB surgery demonstrated significantly high rates of
diabetic remission in the RYGB group (72% versus 17%).454 Consequently, the
weight independent effects of bariatric surgery have led to the concept of “metabolic”
surgery as a possible treatment for type 2 diabetes in overweight individuals.455
1.10.4 Mechanisms behind remission of diabetes after bariatric surgery
Two major hypotheses have been proposed to explain the early effects of RYGB
surgery on regulation of glucose homeostasis.456, 457 The foregut hypothesis states that
rapid delivery of nutrients to the distal small intestine stimulates L-cells, which
secrete hormones (also called incretins) that augment insulin secretion and/or action
resulting in improved homeostasis. Conversely, the hindgut hypothesis states that
excluding parts of the small intestine reduces or suppresses release of anti-incretin
factors that decrease insulin secretion and/or action thus improving glucose
homeostasis.
Although no evidence suggests the foregut and hindgut hypothesis are mutually
exclusive, several studies have reported altered incretin secretion after gastric bypass
surgery. These studies have reported levels of peptide YY (PYY), glucagon like
peptide-1 (GLP-1), gastric inhibitory peptide (GIP) and ghrelin to be altered after
RYGB surgery.458-463
RYGB surgery has been observed to induce early changes in incretin levels before
any substantial weight loss.463 However, whether these incretins are involved in the
53
rapid remission of diabetes after RYGB surgery remains controversial. A study from
our own group has shown there were no significant changes in GLP-1, GIP or PYY
levels within days of RYGB surgery despite normalization of fasting glucose levels
(manuscript submitted).
The rapid exposure of the foregut to food has also led to formation of two other
hypotheses. The ability of incretins to signal through neuronal networks has led to the
suggestion that a gut-brain-liver axis may have a role in remission of diabetes after
RYGB surgery. Cholecystokinin (CCK) is a peptide hormone released by the gut in
response to nutrients and has been shown to lower glucose production through a
neural network.464 Similarly, central nervous system GLP-1 receptors have been
shown to have a role in enteric sensing of glucose ingestion that modulates peripheral
glucose metabolism, which is impaired in diabetes.465
The role of intestinal nutrient sensing has also been directly implicated in the antidiabetic effect of RYGB surgery. Troy et al. (2008) suggested that increased intestinal
gluconeogenesis after RYGB surgery elevates the portal vein glucose concentration.
This is thought to activate the hepatic glucose portal sensing pathway, which
decreases hepatic glucose output and improves glucose homeostasis.466 However, a
study by Hayes et al. (2011) showed that there were no significant differences
between portal and peripheral glucose concentrations after the first six days of RYGB
surgery.467 This suggests intestinal gluconeogenesis is probably not involved in the
rapid remission of diabetes after RYGB surgery.
1.10.5 Tissue specific changes in insulin action after RYGB surgery
Regardless of the mechanism behind the anti-diabetic effects of RYGB surgery
manipulation of the gastrointestinal tract can significantly improve insulin sensitivity
and β-cell function.468, 469 Improved insulin secretion and restoration of the first phase
insulin response has been associated with normalization of glucose homeostasis after
gastric bypass surgery.470-472 Furthermore, rapid resolution of insulin resistance is
associated with a return to normoglycemia after RYGB surgery (Figure 1.10-2).448
54
Figure 1.10-2. Improvement in fasting plasma glucose concentration and HOMA-IR after
RYGB surgery. HOMA-IR ≤2.4 represent insulin sensitive individuals. Duplicated from
Wickremesekera et al. (2005).
Although others have reported similar findings,454 the mechanisms behind the
improvement in insulin resistance after RYGB surgery remain unclear. Several studies
have suggested that both significant weight loss and decreased energy intake are
responsible for the improvement in insulin sensitivity.257,
473, 474
However, other
studies have shown that even though strict dieting can improve insulin resistance, the
magnitude of improvement is higher in individuals who have undergone RYGB.475
Nevertheless, resolution of insulin resistance occurs rapidly after gastric surgery and
appears to be tissue specific.
Lima et al. (2010) measured both peripheral (euglycemic clamp) and hepatic
(HOMA-IR) insulin resistance in individuals undergoing RYGB surgery. One month
after surgery, hepatic insulin resistance was resolved, whereas peripheral insulin
resistance remained unchanged.476 In the same year, Campos et al. (2010) reported
that rapid normalization of glucose homeostasis was independent of peripheral insulin
resistance, which only resolved when substantial weight loss occurred.477 Similarly, a
study by Foo et al. (2011) showed that normalization of glucose homeostasis within
six days of RYGB surgery was associated with resolution of hepatic insulin resistance
(HOMA-IR), while peripheral insulin resistance (IVITT) increased.475 Taken together,
these data highlight the importance of the liver in remission of type 2 diabetes after
RYGB surgery.
55
1.11 Hypothesis and Aims
Recent advances have increased our understating of insulin resistance and type 2
diabetes. Nonetheless, effective treatments that do not involve surgical manipulation
remain elusive. Bariatric surgery, and especially RYGB surgery, has been shown to
cause rapid remission of type 2 diabetes and insulin resistance. This provides the
scientific community with an unprecedented human model that can be used to
investigate the pathogenesis of type 2 diabetes.
We hypothesize that the liver is critical to remission of diabetes after RYGB surgery
and has a central role in the pathogenesis of type 2 diabetes. There are multiple lines
of evidence supporting this hypothesis:

The liver has significant contributions to metabolic homeostasis via regulation
of glucose and lipid production.

The liver regulates insulin action in other tissues by modulating the peripheral
insulin concentration via hepatic insulin clearance.

Transgenic animal studies have demonstrated that the liver may be more
critical than muscle and fat tissues in regulating glucose homeostasis.

Primary hepatic insulin resistance may cause whole body insulin resistance
that leads to type 2 diabetes.

Resolution of hepatic insulin resistance is thought to be critical to the
remission of type 2 diabetes after RYGB surgery.

Resolution of liver steatosis has been associated with improvement in
regulation of glucose homeostasis.
Consequently, this thesis explores the role that hepatic molecular insulin resistance
has in the pathogenesis of type 2 diabetes.
56
Although critical to insulin action in the periphery, hepatic insulin clearance has been
largely overlooked as a mechanism that may be involved in type 2 diabetes. Likewise,
even though the IR and its isoforms have been known for 30 years, their involvement
in type 2 diabetes is still controversial. The broad aim of this thesis is to use the
RYGB as a human model to further investigate previously implicated as well as
identify novel molecular processes involved in the pathogenesis of type 2 diabetes.
This aim will be achieved by answering the following questions.

Does gene and protein expression of molecules that modulate insulin
signalling (ENPP1) and hepatic insulin clearance (CEACAM-1 and IDE)
change following the improvement in insulin resistance and remission of
diabetes achieved by RYGB?

Do the insulin receptor isoforms A and B have a role in the pathology of type
2 diabetes?

What global changes occur in hepatic gene expression in individuals having
RYGB who experience normalization of insulin sensitivity and remission of
type 2 diabetes.
57
58
CHAPTER TWO: Roux-en Y gastric bypass surgery can
be used to study type 2 diabetes
59
2.1 Introduction: using models to study type 2 diabetes
The use of model organisms to study disease is essential to molecular biology. The
most commonly employed model organisms in diabetes are the mouse and rat.25, 478
Although larger animals such as dogs and pigs have also been used, the move to
transgenic animal models has seen the rise of the rodent as the model of choice for
dissecting the insulin signalling pathway. With these models, we can control for
environmental factors allowing us to characterise single genes.
To advance our understating of type 2 diabetes pathogenesis even further, we need
models that represent the environmental and genetic variability seen in a diverse
human population. Research in humans has been limited to clinical and molecular
experiments using easily accessible samples such as blood and muscle or fat tissues.
However, bariatric surgery not only provides an excellent human model for the study
of type 2 diabetes, but also allows for access to liver tissue during surgery. Further,
some individuals may need added surgery for related or unrelated complications (e.g.
cholecystectomy, incisional hernia, ring removal) and as a result will be available for
repeat biopsy of tissues. For the first time it becomes possible to examine tissue in
individuals with diabetes and the same individual following remission of diabetes by
RYGB. This creates a powerful tool for exploring the pathogenesis of type 2 diabetes
and provides an opportunity exploited in this thesis.
The studies reported in this PhD thesis describe experiments performed on a subgroup
of patients undergoing RYGB at Wakefield Hospital in recent years. All patients
undergoing RYGB surgery since 2000 for severe obesity at Wakefield Hospital were
considered for data and specimen collection. At surgery, various blood and tissue
samples were collected including a liver biopsy. This chapter characterises the study
group used in this thesis to explore the liver molecular mechanisms behind insulin
resistance and type 2 diabetes.
60
2.1.1 Experimental strategy

Individuals selected for the study were demographically profiled and
metabolically characterised using a range of clinical tests that included
measurement of fasting plasma glucose, fasting plasma insulin and HbA1c.

Insulin resistance was estimated using the HOMA-IR algorithm, whereas
glucose tolerance and type 2 diabetes was identified using a oral glucose
tolerance test

Individuals were then categorised into study groups based on their metabolic
characteristics that included: normal glucose tolerance, impaired glucose
tolerance and type 2 diabetes.

The metabolic characteristics of the subsets of individuals who had a second
liver biopsy after RYGB surgery are presented alongside the main study
groups for comparisons.
61
2.2 Research Design and Methods
All patients on whom data and tissue is available before and/or after RYGB surgery
for obesity (>35kg/m2) gave informed consent under ethical approval from the Central
Regional Ethics Committee (Wellington, New Zealand). All surgery was performed
by the same surgeon each time (Prof. Richard Stubbs) and all procedures were
approved by the Central Regional Ethics Committee.
2.2.1 Pre operative data collection and type 2 diabetes diagnosis
Clinical data was collected 2-6 weeks prior to surgery which included, but was not
limited to: weight, height, body mass index, glycated haemoglobin (HbA1c), fasting
plasma glucose, and fasting plasma insulin.
Diagnosis of type 2 diabetes was established in two ways. First, an individual was
diagnosed with type 2 diabetes if there was prior documentation of diagnosis and/or if
the individual was receiving treatment for type 2 diabetes. Alternatively, diabetes was
established by an oral glucose tolerance test (OGTT) which was routinely performed
in all patients without a known history of diabetes. Previously unrecognised diabetes
was established by either:

Fasting glucose≥ 7.0 mmol/L and/or a

OGTT 2 hour glucose ≥ 11.1 mmol/L.
Individuals who had a fasting glucose < 7.00 mmol/l, but an OGTT 2 hour glucose ≥
7.8 ≤11.1 mmol/l were diagnosed with impaired glucose tolerance. Individuals with
type 2 diabetes were further classified as previously unrecognized, diet controlled,
requiring oral hypoglycaemic drugs or insulin taking.
62
2.2.2 Subject selection and grouping
For the studies described in this thesis a subset of 55 patients (from a database of
>1000) was selected based on availability of fresh frozen liver tissue taken at the time
of RYGB surgery. This study cohort included individuals who had: normal glucose
tolerance (NGT group), impaired glucose tolerance (IGT group), and type 2 diabetes
(T2DM group).
Out of those 55, 16 individuals had repeat liver biopsy at the time of subsequent
operations. These individuals were further classified into two re-operation groups as
follows: individuals that had a second liver biopsy and had normal glucose tolerance
(sNGT group) and individuals who had a second liver biopsy and had type 2 diabetes
(sT2DM group). Table 2-1 shows relevant anthropometric data for the five study
groups around the time of RYGB surgery
Table 2-1. Anthropometric data of 55 obese individuals classified by study group.
Variables
NGT
(n=19)
sNGT
(n=8)
IGT
(n=9)
T2DM
(n=27)
sT2DM
(n=8)
Age
44 ± 8
43 ± 9
47 ± 10
54 ± 7
50 ± 7
BMI (kg/m2)
46 ± 6
44 ± 5
49 ± 4
49 ± 10
49 ± 10
Female (%)
16 (84)
6 (75)
6 (66)
15 (55)
5 (63)
Male
3
2
3
12
3
Previously unrecognized (%)
-
-
-
7 (27)
3 (37.5)
Diet Controlled (%)
-
-
-
2 (8)
1 (12.5)
Oral hypoglycaemics (%)
-
-
-
14 (50)
4 (50)
Insulin taking (%)
-
-
-
4 (15)
0
Gender
Diabetes
Values are presented as mean± SD
All individuals were morbidly obese at the time of surgery (BMI>40 kg/m2) and there
were no statistically significant differences in BMI between any of the groups, while
females tended to be overrepresented particularly in the NGT and sNGT group.
63
2.2.3 Tissue sampling and follow up data collection
At the time of surgery a variety of tissue samples were collected including a TruCut® Soft Tissue Biopsy Needle (Cardinal Health) liver biopsy. Follow up data was
taken at three and 12 months after RYGB surgery. At each assessment, data collection
included, weight, body mass index, HbA1c, fasting plasma glucose and fasting plasma
insulin.
A second liver biopsy was taken from a limited number of patients (~16 months after
RYGB surgery) who returned for further surgery which allowed access to the liver.
Almost all of the subsequent surgeries were incisional hernia repairs, except two,
which were ring removal procedures. For those individuals who had a second liver
biopsy, clinical measurements were taken within 2 months of subsequent surgery.
Remission of type 2 diabetes was defined according to the criteria of Buse et al.
(2009),445 which is as follows:

Partial remission was documented if, one year after surgery, the HbA1c% was
<6.5%, fasting plasma glucose was 5.6-6.9 mmol/l and there was no ongoing
treatment.

Complete remission was documented if, one year after surgery, the HbA1c%
was <6.5%, fasting plasma glucose was <5.6 mmol/l and there was no ongoing
treatment.
2.2.4 Biochemical testing
All patients undertook a 12 hour fast prior to blood collection for biochemical tests.
Fasting was either self administered or administered during the hospital stay. Clinical
biochemistry testing was conducted by Aotea Pathology (Wellington, New Zealand)
while testing for insulin was undertaken at Canterbury Health Laboratories
(Christchurch, New Zealand), both of which are accredited by International
Accreditation New Zealand (IANZ).

The particular gastric bypass procedure performed at Wakefield Hospital is the Fobi Pouch, which
includes a silastic ring to define the size of the gastric outlet. In approximately 2% of individuals this is
subsequently removed to improve quality of eating.
64
Glucose was assayed on the Roche Modular P800 Chemistry Analyzer using standard
Roche reagent, whereas HbA1c levels were assayed on the Variant Turbo Ion
Exchange HPLC platform. Insulin was assayed using the Roche Elecsys 2010
automated analyzer and standard Roche reagent. Below is a brief description of the
specimen collection and method principal for each assay.
2.2.4.1 Glucose assay and method principle
Blood was collected in a BD Vacutainer® tube containing K2C2O4 (anticoagulant)
and NaF (antiglycolytic agent). Glucose concentrations were assayed using an
enzymatic method with hexokinase479 (Figure 2.2-1). In this method, hexokinase
catalyzes the phosphorylation of glucose to glucose-6-phosphate. Glucose-6phosphate dehydrogenase then oxidizes glucose-6-phosphate in the presence of
NADP to gluconate-6-phosphate. The rate of NADPH formation is measured
photometrically and is directly proportional to the concentration of glucose in the
sample.
Figure 2.2-1. Glucose method principle. An enzymatic method that uses the rate of NADPH formation
to indirectly measure the concentration of glucose in a sample.
2.2.4.2 HbA1c assay
Blood was collected in a BD Vacutainer® tube containing K2EDTA (anticoagulant).
HbA1c was assayed on the Variant Turbo Ion Exchange High Performance Liquid
Chromatography (HPLC) platform. Briefly, HbA1c is glycated haemoglobin in which
glucose is bound to the N-terminal valine of the haemoglobin β-chain. This changes
its conformation resulting in a molecule with an extra negative charge. Using ion
exchange HPLC, Hb species can be separated based on the net charge differences
between HbA1c and different haemoglobins, thus allowing for its measurement.
65
2.2.4.3 Insulin assay and method principle
Blood was collected in a BD Vacutainer® tube containing K2EDTA (anticoagulant).
Plasma was separated from the red blood cells by centrifuging the tubes at 1300 RCF
for 10 minutes and stored at -80ºC before being sent frozen to Canterbury Health
Laboratories for insulin analysis. Prior to analysis, plasma was treated with an equal
volume of 25% PEG 6000 to precipitate immunoglobulins that may interfere with the
assay.
Following this, the insulin concentration was measured using a two-site noncompetitive sandwich assay that uses two monoclonal antibodies. Briefly, in the first
step the sample is incubated for nine minutes with a reagent containing biotinylated
insulin antibody and a ruthenium labelled insulin antibody, both of which bind to the
free insulin in the sample. Streptavidin-coated paramagnetic microparticles are then
added, which bind to the biotin labelled insulin antibody. The reaction mixture is then
transported to the measuring cell where the immune complexes are magnetically
trapped on the working electrode and the unbound reagent and sample are washed
away. An electrical charge is then applied which excites the bound ruthenium and
stimulates an electrochemiluminescent reaction that produces light. The amount of
light produced is directly proportional to the amount of insulin in the sample.
2.2.5 Estimation of insulin resistance
Homeostasis model assessment (HOMA) was used to estimate insulin resistance using
the following formula;
HOMA-IR = fasting plasma insulin (mU/l) × fasting plasma glucose (mmol/l)
22.5.88
HOMA-IR less than 2.2 has been shown to be representative of a metabolically
normal, insulin sensitive population.90
66
2.2.6 Statistical analysis
Normal distribution of all groups to be compared was tested with D’Agostino and
Pearson omnibus normality test. Non-normally distributed data was log transformed
to comply with parametric test assumptions. Pre-post comparisons were carried out
using a paired t-test. One way analysis of variance (ANOVA) with Bonferroni
correction was used to test for significant differences between means. General linear
model (GLM) with Tukey pairwise comparison was used to test for significant
differences in means between multiple factors. An alpha level of 0.05 was set as the
significance threshold. All analysis was performed using Minitab15. Graphical
visualization was performed using GraphPad Prism 5.
67
2.3 Results
In order to achieve the aims set out in this thesis, the metabolic status of the study
cohort used was characterised by investigating several clinical parameters before and
after RYGB surgery. These parameters included: weight loss (via change in BMI),
insulin resistance (fasting insulin and HOMA-IR), and glucose homeostasis (fasting
glucose and HbA1c).
2.3.1 Weight loss after bariatric surgery
Individuals were followed up with measurements recorded at three and 12 months
after surgery. Out of the 55 individuals included in the study, three had incomplete
follow up BMI data. BMI measurements were logarithmically transformed prior to
statistical analysis using GLM, while Figure 2.3-1 presents the mean BMI values for
the five groups at each of the time points measured. The NGT, IGT and T2DM groups
all had a significant decrease in BMI within 3 months of surgery, which decreased
further at the one year time point.
Figure 2.3-1. Mean BMI ± SD after RYGB surgery in all individuals and the subset that had a second
liver biopsy. *The change in BMI observed 3 and 12 months after RYGB surgery in the NGT, IGT and
T2DM groups was significantly different from pre-RYGB values (GLM).
68
The subset of individuals who had a second liver biopsy had comparable weight loss
to the rest of the study group. Notably, there was no statistically significant difference
in weight loss between the IGT, NGT or T2DM group (p=0.131) and the sNGT and
sT2DM group (p=0.180) at any time point measured.
2.3.2 Improvements in insulin resistance after RYGB surgery
Biochemical parameters were measured at 3 and 12 months after RYGB surgery. Out
of the 55 individuals included in the study, 12 had incomplete follow up biochemical
data. Insulin resistance was estimated by calculating HOMA-IR at each of the time
points. Fasting plasma insulin concentrations and HOMA-IR were logarithmically
transformed prior to statistical analysis using GLM and Tukey’s pairwise comparison.
Figure 2.3-2 presents the untransformed means for fasting insulin and HOMA-IR for
the five groups at each of the time points measured.
Fasting insulin concentrations fell significantly following RYGB surgery in the NGT,
IGT and T2DM group (Figure 2.3-2A and C), which was statistically significant
(P<0.001). The magnitude of reduction in insulin concentrations was similar in all
groups. As expected, the T2DM group had a higher mean insulin concentration than
the NGT group (p=0.02), although this difference normalised after the operation as
there were no statistically significant differences in insulin concentrations between
any of the groups after RYGB surgery. The same fall in fasting insulin was seen in the
subgroup of those having a second liver biopsy (sNGT and sT2DM group).
Insulin resistance improved dramatically after RYGB as evidenced by the statistically
significant (p<0.001) decreases in HOMA-IR values in the NGT, IGT and T2DM
groups (Figure 2.3-2C). Before the surgery, the NGT, IGT and T2DM groups all had
mean HOMA-IR values above 2.4 indicating varying degrees of insulin resistance. As
expected, the T2DM had the highest mean HOMA-IR value, which was significantly
higher that the mean HOMA-IR of the NGT group (p<0.001). Within 3 months of
surgery, the values decreased to 1.46 (± 0.98 SD), 2.3 (± 0.94 SD), and 2.44 (± 1.71
SD) for the NGT, IGT and T2DM group respectively.
69
Figure 2.3-2. Change in mean fasting plasma insulin concentrations and HOMA-IR ± SD in the NGT,
IGT and T2DM groups (A and C) and in the sNGT and sT2DM groups (B and D) up to one year after
RYGB surgery. *Difference in insulin concentration and HOMA IR between the NGT and T2DM was
statistically significant (GLM).** Change in insulin and HOMA-IR was statistically significant 12
months after RYGB in the NGT, IGT and T2DM groups (GLM).
By the one year time point all three groups had a statistically significant (p<0.001)
decrease in HOMA-IR values. The subset of individuals who underwent a second
liver biopsy had similar HOMA-IR profiles to the main study groups after RYGB
surgery (Figure 2.3-2B and D). Furthermore there were no statistically significant
differences in HOMA-IR between any of the groups 3 months after surgery.
70
2.3.3 Remission of diabetes after RYGB surgery
Fasting plasma glucose concentrations were logarithmically transformed prior to
analysis with GLM and Tukey’s pairwise comparison. Figure 2.3-3 shows the
untransformed means of glucose and HbA1c three and 12 months after RYGB
surgery. The T2DM group had the highest mean glucose concentrations, which
normalised within 3 months of RYGB surgery, and remained below the diagnostic
level for type 2 diabetes (fasting plasma glucose of ≥ 7 mmol/l) at 12 months.
Individuals with normal or impaired glucose tolerance; however, had little to no
changes in mean glucose concentrations (Figure 2.3-3A and B).
Figure 2.3-3. Change in mean fasting plasma glucose concentrations and HbA1c ± SD in the NGT, IGT
and T2DM groups (A and C) and in the sNGT and sT2DM groups (B and D) up to one year after
RYGB surgery. *Difference in glucose concentration and HbA1c between the NGT and T2DM was
statistically significant (GLM). **Changes in fasting glucose and HbA1c from Pre-RYGB values were
statistically significant 3 and 12 months after RYGB only in the T2DM group (GLM).
71
As expected, HbA1c levels followed a similar pattern to the changes in fasting plasma
glucose after RYGB surgery. The T2DM group had the highest mean HbA1c levels,
which were diagnostic of type 2 diabetes. Following RYGB surgery, only the T2DM
group had a decrease in mean HbA1c levels whereas the NGT and IGT groups had no
appreciable changes. Within 3 months of surgery, there were no differences in mean
HbA1c levels between any of the groups. Furthermore, 12 months after surgery, the
T2DM group had a mean HbA1c level of 5.85 (± 0.71 SD), which was below the
diagnostic criteria for type 2 diabetes (HbA1c ≥6.5%). The sNGT and sT2DM groups
had similar mean glucose and HbA1c profiles to the main study groups after RYGB
surgery (Figure 2.3-3 B and D).
Remission of diabetes was based of fasting glucose concentrations and HbA1c levels
one year after RYGB surgery. Out of the 27 individuals who had type 2 diabetes, 18
were shown to have complete remission, six had partial remission after surgery, two
had no remission and one was unknown (due to a missing 12 month HbA1c value).
2.3.4 Metabolic status during liver biopsy
Liver tissue from each individual in the five study groups was used throughout this
thesis to investigate the molecular processes affected by insulin resistance or type 2
diabetes. Table 2-2 shows metabolic characteristics of the NGT, IGT and T2DM
groups around the time of initial liver biopsy (pre RYGB surgery). At the time of
initial liver biopsy, the T2DM group had significantly (p<0.001, ANOVA) higher
HbA1c and fasting glucose levels than both the NGT and IGT groups, while all
groups had abnormally high plasma insulin concentrations and some level of insulin
resistance. The NGT group had the lowest insulin resistance while the T2DM group
had the highest as inferred from HOMA-IR. The plasma insulin concentrations were
only significantly different between the T2DM and the NGT group (p<0.001,
ANOVA) as were the HOMA-IR values (p<0.001, ANOVA).
72
Table 2-2. Metabolic status at the time of initial liver biopsy
Variables
NGT group
(n=19)
IGT group
(n=9)
T2DM group
(n=27)
BMI (kg/m2)
46 ± 6.0
49 ± 3
49 ± 10
HbA1c (%)
5.5 ± 0.4
5.6 ± 0.5
7.6 ± 1.1*
HOMA-IR
3.70 ± 3.6
6.07 ± 3.1
10.78 ± 8.6*
Fasting plasma insulin (pmol/L)
109 ± 96
164 ± 80
196 ± 115†
Fasting plasma glucose (mmol/L)
5.1 ± 0.6
5.7 ± 0.6
8.1 ± 2.3*
NGT vs T2DM data are significantly different; *p<0.001, †p<0.05 (ANOVA). Values are presented as
mean± SD.
Individuals in the sNGT and sT2DM groups all had a second liver biopsy at
subsequent operations. Metabolic characteristic for the sNGT and sT2DM groups at
the time of initial and second liver biopsy are listed in Table 2-3 for comparison. Out
of the 16 individuals who had a second liver biopsy, two had incomplete follow up
data. Both the sNGT and sT2DM groups had significant decreases in BMI by the time
of the second biopsy. There were no statistically significant differences in BMI
between the two groups at either operation or in the amount of weight lost.
Table 2-3. Metabolic status of individuals at the time of initial and subsequent liver
biopsy.
Variables
sNGT group
(n=8)
Pre-RYGB
Post-RYGB
sT2DM group
(n=8)
Pre-RYGB
Post-RYGB
BMI (kg/m2)
44 ± 5
28 ± 4*
49 ± 10
30 ± 5*
HbA1c (%)
5.4 ± 0.2
5.0 ± 0.7
7.1 ± 1
5.5 ± 0.2†
HOMA-IR
2.5 ± 0.45
0.78 ± 0.34†
10.3 ± 6.2
1.4 ± 0.8*
FPI (pmol/L)
80 ± 16
27 ± 13*
219 ± 127
48 ± 23†
FPG (mmol/L)
4.9 ± 0.4
4.7 ± 0.4
7.4 ± 1.9
4.7 ± 0.6†
Pre-RYGB vs post-RYGB data are significantly different *p<0.001, †p<0.05 (Paired t-test). Values are
presented as mean± SD.
73
For the sT2DM group, there was a significant (p < 0.05, Paired t-test) decrease in
mean fasting plasma glucose and HbA1c levels, which was representative of complete
remission of type 2 diabetes in all except one individual who had a missing 12 month
HbA1c value and thus could not be categorized.
Hepatic insulin sensitivity (as measured by HOMA-IR) improved dramatically in both
the sNGT and sT2DM group, even though the sNGT group were already relatively
insulin sensitive before RYGB surgery. Similarly, fasting plasma insulin
concentrations fell significantly in both groups. Nonetheless, the sT2DM had the
greatest improvement in insulin sensitivity which was associated with the
normalisation of insulin resistance in those individuals.
74
2.4 Discussion
Bariatric surgery has provided the scientific community with an unprecedented human
model of type 2 diabetes. RYGB surgery in particular induces drastic improvements
in glycaemic control in morbidly obese individuals.14, 430, 439 This chapter characterises
the metabolic status of individuals who contributed tissue for investigation of liver
molecular mechanisms involved in insulin resistance and type 2 diabetes.
All individuals included in this study were morbidly obese (BMI>40 kg/m2) and as
expected all had a great and continuous decrease in BMI up to one year after RYGB
surgery (Figure 2.3-1). Although weight loss undoubtedly has a role in the
improvements seen in metabolic homeostasis, RYGB surgery has been shown to
induce normoglycemia without significant weight loss.448,
454, 475, 480
Further, while
liver insulin resistance as measured by HOMA-IR decreases rapidly post RYGB
surgery, whole body insulin resistance only decreases with weight loss and may
remain for up to six months after surgery.448, 476, 477, 480, 481 This strongly suggests the
liver is central to remission of type 2 diabetes after RYGB surgery.
We grouped the individuals included in the study according to their metabolic
characteristics. As expected mean fasting plasma glucose and HbA1c levels were
normal in the NGT group (Figure 2.3-3A and C), and slightly raised in the IGT group.
However there were no statistically significant differences between the NGT and IGT
group in any of the parameters measured at RYGB surgery. In addition, the IGT
group had a smaller sample size in comparisons to the NGT and T2DM groups and as
a result we excluded the IGT group from the studies described in this thesis.
Unlike the NGT group, the T2DM group had significantly raised mean fasting plasma
glucose and HbA1c values representative of individuals with type 2 diabetes (Figure
2.3-3A and C). Although insulin resistance was present in varying severity in all
groups, the T2DM group were the most insulin resistant as they had the highest mean
fasting plasma insulin concentration (Figure 2.3-2A) and HOMA-IR values (Figure
2.3-2C).
75
The mechanisms governing improvement in metabolic homeostasis after RYGB
surgery are unclear.432 Nonetheless, improvements in insulin sensitivity are critical to
normalisation of glucose homeostasis in individuals with type 2 diabetes. In this
study, RYGB surgery induced an increase in insulin sensitivity regardless of disease
presence. Interestingly, the NGT and T2DM groups both had a significant decrease in
fasting plasma insulin concentrations and HOMA-IR values within a year of surgery
(Figure 2.3-2A and C). Resolution of insulin resistance in the T2DM group coincided
with remission of diabetes in 88% of individuals, which is similar to remission rates
previously reported.14 Although two individuals did not have remission of diabetes,
they did have an improvement in insulin resistance as judged by HOMA-IR, which
was < 2.4 for both individuals.
One of the novel characteristics of this thesis is the use of liver tissue biopsied from
the same individual before and after significant improvement in insulin sensitivity
and/or remission of type 2 diabetes. Figures 2.3-1, 2.3-2 and 2.3-3 show that both the
sNGT and sT2DM group had similar changes in metabolic characteristics as the
groups they stemmed from (NGT and T2DM group). Even though most of the sNGT
group were insulin sensitive at RYGB surgery, they still had a significant increase in
insulin sensitivity at second liver biopsy. The sT2DM group had resolution of insulin
resistance at second liver biopsy as the mean HOMA-IR decreased to from 10.3 (± 6.2
SD) to 1.4 (±0.8 SD), which was well within the insulin sensitive range (HOMA IR
<2.4). Resolution of insulin resistance in the sT2DM group coincided with the
remission of diabetes as shown by normalisation of mean fasting plasma glucose
concentrations and HbA1c levels (Table 2-3). Importantly, there were no differences
in BMI between the sNGT and sT2DM groups at any one time point measured
(Figure 2.3-1) nor at the time the time of the second liver biopsy (Table 2-3). This
removed weight loss as a possible confounding factor.
Collectively, the data presented here shows that individuals used in this study suitably
model insulin resistance and type 2 diabetes. The individuals grouped in the NGT and
sNGT groups had normal glucose homeostasis and were relatively insulin sensitive.
Individuals grouped in the T2DM and sT2DM groups had deranged glycaemic control
and were severely insulin resistant. Further, RYGB surgery induced significant
improvements in insulin sensitivity in all individuals which coincided with remission
76
of diabetes in those individuals that had the disease. We used these study groups to
answer two key questions asked throughout the thesis. First, what are the molecular
differences in liver insulin signalling between individuals with or without type 2
diabetes? Second, how do pathways involved in insulin action in the liver change after
resolution of insulin resistance and what role do they have in remission of type 2
diabetes. The next two chapters use the RYGB model to explore several key
molecules involved in insulin action and whether these molecules have a significant
role in the pathogenesis of insulin resistance and type 2 diabetes.
77
78
CHAPTER THREE: Investigation of molecules involved
in the modulation of insulin signalling strength and
hepatic insulin clearance
79
3.1 Introduction
Insulin action is dependant on the circulating insulin concentration and the consequent
activation of the insulin receptor tyrosine kinase. Investigating the mechanisms that
govern insulin action is critical to our understanding of insulin resistance and type 2
diabetes. This chapter presents data on two key processes involved in insulin action,
namely hepatic insulin clearance and negative regulation of IR activity.
Negative regulation of IR activity through various mechanism has previously been
implicated in the pathogenesis of insulin resistance.158 One of these mechanism is the
inhibition
of
insulin
binding
to
the
IR
α-subunit
by
ectonucleotide
pyrophosphatase/phosphodiesterase 1 (ENPP1 also referred to as plasma cell
membrane glycoprotein 1 or PC-1).166 Increased ENPP1 protein levels have been
found in skeletal muscle,167 adipose tissue168 and skin fibroblasts169, 170 from insulinresistant individuals. Overexpressing ENPP1 in vitro170 and in vivo172 led to impaired
insulin signalling and action, while selectively suppressing the levels of ENPP1 in the
liver of mice improved insulin sensitivity.173 Moreover, there is compelling evidence
suggesting a role for ENPP1 in genetic forms of insulin resistance. Genotypephenotype association studies have identified a gain of function mutation (K121Q),
which increases the inhibitory function of ENPP1 and may have a role in insulin
resistance and type 2 diabetes.174, 175
Hepatic insulin clearance mediates insulin action by regulating the systemic insulin
concentration and has been found to be diminished in obesity and type 2 diabetes.57-60
Although the pathological importance of this process is unclear, a recent study by
Bojsen-Møller et al. (2013)482 has shown that hepatic insulin clearance increases as
early as one week after RYGB surgery. The liver regulates insulin clearance by
modulating the rate at which insulin is internalized and degraded. Two key molecules
involved in this process are carcinoembryonic antigen-related cell adhesion molecule
1 (CEACAM-1) and insulin degrading enzyme (IDE). CEACAM-1 is involved in
mediating the internalization of the insulin-insulin receptor complex.64-66 Transgenic
mice with aberrant CEACAM-1 expression or action have impaired insulin clearance,
secondary insulin resistance, abnormal glucose tolerance and visceral adiposity.
70
64, 65,
On the other hand, IDE is responsible for the degradation of insulin63 and has
80
previously been implicated in the pathogenesis of type 2 diabetes.72-74 Furthermore,
inhibitors of IDE have previously been proposed as a potential anti-diabetic therapy.71
Despite the evident roles of ENPP1, CEACAM-1 and IDE in insulin resistance, they
are yet to be extensively characterised in liver tissue of human individuals with and
without type 2 diabetes. Consequently, this chapter presents the gene expression and
protein abundance of the three aforementioned molecules in individuals with and
without type 2 diabetes before and after RYGB surgery.

Protein abundance was assayed in the NGT, T2DM, sNGT and sT2DM group
using E-PAGE™ 48-well protein electrophoresis and western blotting.

Gene expression was assayed in the NGT, T2DM, sNGT and sT2DM group
using RT-qPCR with TaqMan Gene Expression assays and the Relative
Quantification method.
81
Metabolic characteristics of individuals from whom liver tissue was available are
presented in Chapter II. Liver gene expression and protein abundance of ENPP1,
CEACAM-1 and IDE was assayed in individuals with type 2 diabetes (T2DM group)
or normal glucose tolerance (NGT group). In addition, gene expression and protein
abundance was also assayed after RYGB surgery (and remission of diabetes) in
individuals with normal glucose tolerance (sNGT group) or type 2 diabetes (sT2DM
group). In this way we expected to learn something of the role of these genes in the
development of type 2 diabetes.
3.2.1 RNA extraction
Total RNA was phenol chloroform extracted by a method based on a technique
developed by Chomczynski et al. (1987).483 Briefly, 2-3mg of frozen liver tissue was
pulverized under liquid nitrogen after which 800μl of TRIzol® (Life Technologies), a
monophasic solution containing phenol and guanidinium thiocyanat, was added to the
resultant powder. The mixture was then homogenised with an Omni Tissue
Homogenizer (Omni International) and 200 μl of chloroform was added, vigorously
mixed then centrifuged at 12,000g for 15 minutes at 4-8ºC. Following centrifugation
the mixture separates into a red phenol-chloroform phase, an interphase and a
colourless upper aqueous phase containing RNA.
To precipitate the RNA, the aqueous phase was transferred to an RNAase/DNAase
free Eppendorf tube containing 500μl of isopropyl alcohol and 1.3 μl (conc)
GlycoBlue (Life Technologies). GlycoBlue is a blue dye covalently linked to
glycogen and was used as a co-precipitant to increase RNA pellet mass and aid in
pellet visualization. Following 10 minute incubation, samples were centrifuged at
12,000 g for 10 minutes at 4-8ºC. The resultant RNA pellet was washed with 75%
ethanol and air dried for 10 minutes. Depending on pellet mass, 20-30μl of DEPCtreated water was used to re-suspend the RNA pellet after which it was incubated at
55ºC for 10 minutes. RNA was stored at -80 ºC prior to quality control and use in first
strand cDNA synthesis.
82
3.2.2 RNA quality control
Quality and quantity of RNA was analyzed on the 2100 Agilent Bioanalyzer, using
the Agilent RNA 600 Nano kit and BioSizing software (Agilent Technologies). The
Agilent 2100 Bioanalyzer uses microfluidic capillary electrophoresis to separate RNA
samples according to their molecular weight and then detects them using laserinduced fluorescence detection. The Biosizing software produces an electropherogram
where the amount of fluorescence correlates with the amount of RNA. An RNA
integrity number (RIN) was calculated by an algorithm which calculates the RNA
integrity from the electrophoretic data using a trained artificial neural network. A RIN
number was computed for each RNA profile and classified it into one of 10
predefined categories of integrity. A RIN of 1 represented a completely degraded
sample, whereas a RIN of 10 represented an intact RNA sample. A RIN of 5.5 or
higher is appropriate for RT-qPCR experiments.484,
485
Table 3-1 lists mean RNA
integrity numbers for all the groups that were compared. All RIN numbers were above
the 5.5 threshold for RT-qPCR and as such passed quality control.
Table 3-1. Mean RNA integrity number
for all groups being compared.
Study Group
Mean RIN (SD)
NGT group
7.1 (1.0)
T2DM group
7.8 (0.9)
Pre RYGB sNGT
7.5 (1.6)
Post-RYGB sNGT
7.9 (1.0)
Pre-RYGB sT2DM
8.0 (0.7)
Post-RYGB sT2DM 8.1 (0.8)
3.2.3 First strand cDNA synthesis
Total RNA was reverse transcribed using the SuperScript Vilo cDNA synthesis kit
(Life Technologies). 1 μg of RNA was added to a 20μl reaction containing
SuperScript III reverse transcription enzyme, random primers, RNaseOUT,
Recombinant Ribonuclease Inhibitor and RNase free water.
83
SuperScript III is an engineered version of M-MLV reverse transcription enzyme
which is not inhibited by ribosomal and transfer RNA allowing it to synthesize cDNA
from total RNA. Reactions were incubated at 25ºC for 10min, then at 42 ºC for 60
minutes followed by an 85ºC 5 min incubation to deactivate the reverse transcriptase.
The resulting cDNA was diluted 1:10 and stored at -20ºC prior to use in RT-qPCR
reactions.
3.2.4 Analysis of gene expression with RT-qPCR
All RT-qPCR reactions were done using EXPRESS qPCR SuperMix (Life
Technologies, USA) and TaqMan Gene Expression Assays (Applied Biosystems,
USA). Each 20μl reaction contained 2μl of cDNA, 10μl of EXPRESS qPCR
SuperMix (2X) and 1μl of TaqMan Gene Expression Assay (20X). All RT-qPCR
reactions were analyzed with the ABI 7300 Real-Time PCR System (Applied
Biosystems, USA) as per the thermocycling conditions presented in Table 3-2. Each
sample was run in triplicate and each time a threshold cycle (Ct) was obtained using
the 7300 Sequence Detection Software 1.3.1.
Table 3-2. Thermocycling conditions for RT-qPCR reactions using EXPRESS
qPCR SuperMix
Cycle Step
Initial Denaturation
Temperature (ºC) Time
95
2 min
Denaturation
95
Anneal/Extension and Data Collection 60
Cycles
1
15 sec 45
1 min
3.2.5 EXPRESS qPCR SuperMix
EXPRESS qPCR SuperMix (Life Technologies, USA) contains platinum Taq DNA
polymerase, uracil DNA glycosylase, dNTPs that contain dUTP instead of dTTP and
ROX reference dye. Platinum Taq DNA polymerase is a recombinant Taq DNA
polymerase complexed with antibodies that blocks polymerase activity at ambient
temperature.486 This blocking activity is removed by heat resulting in activation of the
polymerase during the first denaturation step.
84
Essentially an automatic hot start, it increases sensitivity and specificity. dUTP
incorporates uracil into any amplified DNA while uracil DNA glycosylase removes
the uracil residues. This prevents the re-amplification of carryover PCR products.487
ROX dye is composed of a glycine conjugate of 5-carboxy-X-rhodamine,
succinimidyl ester and is used to normalise the fluorescence signal between reactions.
3.2.6 TaqMan gene expression assays
Human 18S rRNA (4319413E), POLR2A (Hs 01108291_m), ENPP1/PC-1 (Hs
01054040_m1), CEACAM-1(Hs 00610438_m) and IDE (Hs 00266109_m) assays
(Applied Biosystems, USA) were used. TaqMan chemistry was chosen over the more
standard PCR chemistries such as SYBR green dye due to its increased specificity and
decreased post-PCR processing. TaqMan Gene Expression Assays are hydrolysis
probes based of work done in 1991 by Holland et al.488 Briefly, one assay consists of
two primers (900nM/primer final concentration) and one MGB probe. A minor
groove binder (MGB) at the 3’ end of the probe increases the melting temperature
(Tm) without increasing probe length thus allowing for the design of shorter probes.489
The probe is labelled with a reporter dye at the 5’ with either 6-FAM (ENNP1,
CEACAM-1, and IDE) or 6-VIC (18S) and is used at a final concentration of 250nM.
Consequently, a fluorescent signal generated during amplification is used to detect a
specific PCR product as it accumulates.
3.2.7 RT-qPCR method principle and analysis of data
Real-time quantitative PCR is based on the exponential nature of PCR where the
amplicon doubles every PCR cycle,490 which can be modelled by the following
equation:
Xn = Xo × (1+Ex) n
Where Xn is the number of target molecules at cycle n, Xo is the initial number of
target molecules, Ex is the efficiency of target amplification and n is the number of
cycles. For real-time amplification using TaqMan Gene Expression Assays, the Xn is
proportional to the reporter fluorescence (R) as modelled by:
85
Rn = Ro × (1+Ex) n
Where Rn is the reported fluorescence at cycle n and the Ro is the initial reporter
fluorescence.
The cycle threshold (Ct) is the cycle at which the fluorescent signal reaches a fixed
threshold and is the main unit for calculating gene expression. The Ct value is
inversely correlated to the amount of starting template meaning the lower the amount
of template the longer it takes to reach threshold.
3.2.8 Relative quantification of gene expression
The Ct of all three replicates for a particular sample was averaged and used in
subsequent data analysis as recommended by Schmittgen and Livak.491 Relative
quantification (RQ) was used to analyze differences and changes in gene expression
between groups. RQ compares the expression of the target gene in a treatment group
(test sample) to that of an untreated or control group (calibrator sample). For
comparisons of gene expression between individuals with and without type 2 diabetes,
the NGT group was used as the calibrator sample and the T2DM group was used as
the test sample. For comparisons of gene expression before and after RYGB surgery,
the Pre-RYGB sNGT or sT2DM group was used as the calibrator sample and the
Post-RYGB sNGT or sT2DM group was used as the test sample respectively.
Changes in gene expression were expressed as relative amount and as fold change,
both of which were calculated using the 2-∆∆Ct formula. The derivation of the original
2-∆∆Ct formula was described in detail in the Applied Biosystems User Bulletin No.2
(P/N 4303859) and by Livak and Schmittgen.490, 492 Using the 2-∆∆Ct method, the Ct for
the target gene was normalised to a reference gene Ct in each sample, which
generated a ΔCt value. To generate the fold change, the ΔCt values of the calibrator
sample were subtracted from the test sample which produced a ΔΔCt value that was
exponentiated to generate the fold change (Figure 3.2-1).
86
Figure 3.2-1. 2-∆∆Ct formula for calculating fold changes in gene expression between a calibrator and
test sample. Exponentiating the ΔCt (2-∆Ct) of the calibrator or test sample allowed for presentation of
gene expression as individual data points.
The ΔCt for the calibrator or test sample was also exponentiated (2-∆Ct) to produce
relative gene expression, which allowed plotting of gene expression as individual data
points. Exponentiation was necessary as the fluorescence data is recorded on a
logarithmic scale (log2) and needs to be converted to linear scale for appropriate data
visualisation.
A main assumption of the 2-∆∆Ct method is that the amplification efficiencies for both
the target and reference gene are approximately equal. However there can be
significant deviations in amplification efficiencies between different gene assays that
may give erroneous results. TaqMan Gene Expression Assays have previously been
reported to have 100±10% efficiency.493 To confirm this, the amplification efficiency
for each TaqMan Gene Expression Assay was tested. The efficiency (E) was
calculated using the Ct slope method and the following equation:
Ex = 10(-1/slope)-1
cDNA transcribed from liver RNA was serially diluted and the Ct values were
determined for each dilution. A plot of Ct versus log cDNA concentration was
constructed. Amplification efficiency was calculated from the slope of the graphs and
is presented in Table 3-3. All of the TaqMan Gene Expression Assays were
approximately equal and were appropriate for use with the 2-∆∆Ct method.
87
Table 3-3. TaqMan gene expression assay amplification
efficiency.
TaqMan gene expression assay
Efficiency (%)
Human 18S rRNA (Reference Gene)
98
POLR2A (Reference Gene)
99
ENPP1/PC-1 (Target Gene)
100
CEACAM-1(Target Gene)
97
IDE (Target Gene)
91
3.2.9 Reference genes
Human 18S rRNA and POLR2A (Hs 01108291_m) were compared with regard to
suitability as reference genes by presenting the PCR data from the biological
replicates as 2-Ct. An ANOVA was performed on the data and both reference genes
were found to be suitable as there was no variation in expression between the groups
being compared (p>0.1). However 18S rRNA was chosen as the reference gene due to
a more normal distribution and low variance as recommend by Mane et al. (2008).494
3.2.10 Protein extraction
Protein for three of the sixteen individuals was not available prior to RYBG (2 from
sNGT group and 1 from sT2DM group). Protein was extracted by pulverizing 3-4mg
of frozen liver under liquid nitrogen. The resultant powder was incubated with buffer
containing Complete Protease Inhibitor Cocktail (Roche), 30 mM TrisCl, 7 M urea, 2
M thiourea and 4% (w/v) CHAPS) for 45 minutes on ice, with vigorous vortexing
every 15 minutes. The mixture was centrifuged at 14000g for 10 minutes to remove
any insoluble material. The supernatant was taken and aliquoted further for storage at
-80 ºC. Protein was quantified using Bradford reagent (Bio Rad) on a Bio-Rad
Benchmark Microplate Reader.495
88
3.2.11 Protein resolution with E-PAGE™
The E-PAGE™ 48-well protein electrophoresis system was used to resolve protein
samples on pre-cast E-PAGE 48 8% gels (20-300kDa separation range) (Life
Technologies). The precast gels contain a gel matrix and electrodes enclosed in a UVtransparent cassette. They were loaded onto a specifically designed electrophoresis
device that combines a base and power supply. The E-PAGE gels contain SDS so
resolution was performed under denaturing conditions for 23 min.
3.2.12 Sample preparation and loading
Prior to sample and marker loading 5µl of deionized water was loaded into all wells.
All samples were denatured in a total volume of 15µl. Protein lysate was made up to
15µg and mixed with E-PAGE loading buffer and β-mercaptoethanol (Sigma) (5%
v/v). Samples were incubated at 70ºC for 10 minutes and loaded onto the gel. To
control for between/within gel variability all groups being compared were resolved in
alternating lanes on the same gel. SeeBlue Plus2 Pre-stained Standard and MagicMark
XP Western Protein Standard (Life technologies) were used as molecular weights to
assess protein migration both during electrophoresis and western blotting respectively.
3.2.13 Electrophoretic transfer to PVDF and western blotting
Immediately after resolution, gels were equilibrated in transfer buffer (39 mM
Glycine, 48 mM Tris, methanol 20% v/v) for 10 minutes. Hybond-P PVDF membrane
(Amersham) was activated with methanol for 10 seconds and washed in deionized
water for 5 minutes after which it was equilibrated in transfer buffer for 10 minutes.
Proteins were transferred from the gel to the Hybond-P PVDF membrane with a BioRad Electrophoretic transfer cell for 2-3hours at 50V. Transfer of proteins was
confirmed with Ponceau S stain. Membranes were blocked with 5% skim milk
powder in TBS-Tween at 4°C overnight and then probed for 2 h at RT with ENPP1
(1:500, Santa Cruz), CEACAM-1 (1:500 Santa Cruz), IDE (1:1000 R&D) and Panactin (1:10000 Milipore).
89
Blots were then incubated with alkaline phosphatise-conjugated anti mouse antibody
(Life Technologies) for 0.5 h. The chemiluminescent signal was developed using
CDP-Star chemiluminescent substrate (Life Technologies) and captured on x-ray film
(Kodak).
3.2.14 Stripping
Due to the scarcity of the samples, PVDF membranes were stripped once and reprobed for a different protein. Membranes were incubated in stripping buffer (100
mM β-mercaptoethanol, 2% (w/v) sodium dodecyl sulphate, 62.5 mM Tris-HCL pH
6.7) at 60°C for 30 min with gentle agitation every 10 min. After rinsing with water,
membranes were washed 3 times in TBS-T for 15 min at room temperature using
larger volumes of wash buffer. Membranes were re-blocked overnight and re-probed
the following day.
3.2.15 Relative quantification of protein abundance
The protein bands on the x-ray film were digitized by the Chemidoc XRS system (Bio
Rad) and then analyzed using Quantity One software (Bio Rad). Bands were
quantified and expressed as volume quantity; the sum of the intensities of the pixels
within a defined volume boundary × pixel area (intensity units × mm2). Local
background subtraction was used to facilitate background correction. This involves
calculating separate background intensity for each volume.
Local background is calculated by taking the combined intensity of the pixels in a 1pixel border around the volume area and dividing by the number of border pixels.
This average background intensity is subtracted from each pixel contained inside the
corresponding volume boundary. Any pixels inside the volume that have the same
intensity as the background pixels are reduced to zero.
496
Protein relative abundance
was calculated by normalizing the intensity of the ENPP1, CEACAM-1 or IDE band
to the intensity of the Actin band (loading control).
90
Statistical analysis was performed on 2-ΔCt RT-qPCR data between each study group.
Consequently gene expression data was expressed as both 2-ΔCt in graphical form,
whereas the fold change for significant differences was reported in text. Statistical
analysis of protein abundance was performed on the value generated from
normalisation of band intensity to the loading control.
Normal distribution of all groups to be compared was tested with D’Agostino-Pearson
test. Non-normally distributed data was log transformed to comply with parametric
test assumptions. Pre-post comparisons were carried out using a paired t-test.
Student’s t-test was used to test significance between two groups. The Pearson
product-moment correlation coefficient was used to test the strength and significance
of association between variables. An alpha level of 0.05 was set as the significance
threshold. All analysis was performed using Minitab15. Graphical visualization was
performed using GraphPad Prism 5.
91
3.3 Results
3.3.1 Liver ENPP1 levels in diabetes
ENNP1 has been shown to negatively modulate IR activity and has been implicated in
insulin resistance. Consequently, levels of ENPP1 mRNA expression and protein
abundance were assayed in individuals with and without type 2 diabetes and again
after resolution of insulin resistance and remission of type 2 diabetes.
Figure 3.3-1A presents mean gene expression of ENPP1 in the NGT and T2DM
groups. The mean hepatic gene expression of ENPP1 was lower in the T2DM group
when compared to the NGT group (Figure 3.3-1A), which is statistically significant
(p=0.01, t-test). The T2DM group had a mean -1.5 (95% CI –6.2 to 2.8) fold lower
ENPP1 mRNA expression than the NGT group at the time of RYGB surgery.
Figure 3.3-1C displays the western blot for ENPP1 protein abundance in NGT and
T2DM groups. Loading was alternated based on disease state to avoid gel bias. The
molecular weight of ENPP1 in SDS-PAGE is reported to be 110-130 kd depending on
tissue type. Some part of ENPP1 reportedly exists as dimmer and the 220kd band
visible in Figure 3.3-1C is most likely the dimeric form of ENPP1.497
ENPP1 protein abundance data were log transformed prior to analysis with t-test in
order to meet parametric test assumptions. Figure 3.3-1B shows the untransformed
mean protein abundance of ENPP1. Mean ENPP1 protein abundance was lower in the
T2DM group in comparison to the NGT group (Figure 3.3-1B), which was also
statistically significant (p=0.03, t-test). In this study, individuals with type 2 diabetes
had a mean -2 fold lower (95% CI, -3.6 to -1.3) abundance of ENPP1 than individuals
with normal glucose tolerance at the time of RYGB surgery.
92
Figure 3.3-1. Mean ENPP1 levels in liver tissue of morbidly obese individuals. NGT; Normal glucose
tolerance (n=19), T2DM; Type 2 diabetes (n=27). (A) Mean + SE ENPP1 mRNA expression presented
as 2-∆Ct normalised to 18S. (B) Mean + SE ENPP1 protein abundance relative to Actin. Protein was not
available from two individuals in the NGT and one individual in the T2DM group. (C) Corresponding
western blot for ENPP1 on all individuals who did not have a second liver biopsy. To avoid gel
artefacts all samples to be compared were loaded onto the same E-PAGE™ gel, while spaces delineate
different gels. Samples with an * were not included in analysis. For complete western blot film see
Appendix.
3.3.2 Liver ENPP1 levels before and after RYGB surgery
ENPP1 mRNA expression and protein abundance was analysed in individuals who
had normal glucose tolerance (sNGT group) or type 2 diabetes mellitus (sT2DM
group) before and after RYGB surgery. ENPP1 mRNA expression data (2-∆Ct) was log
transformed prior to analysis with paired t-test in order to meet parametric test
assumptions. Figure 3.3-2A shows the untransformed mean ENPP1 gene expression.
Liver ENPP1 mRNA expression remained unchanged after RYGB surgery in the
sNGT and the sT2DM group.
93
Figure 3.3-2B shows the untransformed mean abundance of liver ENPP1 protein
before and after RYGB surgery, while Figure 3.3-2C is the corresponding western
blot of ENPP1 protein for all the individuals who had a second liver biopsy after
RYGB surgery. There were six pairs of liver samples from individuals with normal
glucose tolerance (sNGT) while there were seven pairs of samples from individuals
with type 2 diabetes (sT2DM). Pre and post RYGB samples were loaded adjacent to
each other and sNGT and sT2DM pairs were alternated. ENPP1 protein abundance
data was log transformed in order to meet parametric test assumptions. The sT2DM
Pre-RYGB ENPP1 protein abundance tended to be lower than that of the sNGT PreRYGB, although this was not statistically significant (t-test). This is likely because of
the small sample size in the subset of individuals who had a second liver biopsy and
consequent decrease in statistical power when using the student’s t-test.
Figure 3.3-2. Liver ENPP1 gene expression and protein abundance in liver tissue of morbidly obese
individuals before and after RYGB surgery (● sNGT, n=8; ○ sT2DM, n=8). (A) Mean ENPP1 mRNA
expression + SE presented as 2-∆Ct normalised to 18S. (B) Mean ENPP1+ SE protein abundance relative
to Actin. Pre RYGB protein was not available from two individuals in the sNGT and one individual in
the sT2DM group. The sT2DM group had a significant increase in the sT2DM after RYGB surgery
(paired t-test). (C) Corresponding western blot for ENPP1 for individuals that had a second liver
biopsy. To avoid gel artefacts all samples to be compared were loaded onto the same E-PAGE™ gel,
while spaces delineate different gels. For complete western blot film see Appendix.
94
However, although the sNGT group had no appreciable change in ENPP1 protein
levels after RYGB surgery, the sT2DM group had a statistically significant (p=0.01,
paired t-test) increase in ENPP1 protein abundance (Figure 3.3-2B). The individuals
who had resolution of insulin resistance and remission of type 2 diabetes following
RYGB had a mean 2.7 (95% CI, 1.8 to 4.0) fold increase in ENPP1 protein
abundance.
ENPP1 protein data from the sNGT and sT2DM groups were combined and log
transformed prior to calculation of a Pearson product-moment correlation in order to
explore any relationship with change in fasting plasma insulin concentrations and
HOMA-IR. ENPP1 protein abundance negatively correlated with fasting plasma
insulin concentration (Figure 3.3-3A) and HOMA-IR index (Figure 3.3-3B).
Consequently, improvement in hepatic insulin sensitivity after RYGB surgery as
measured by HOMA-IR was associated with increasing levels of ENPP1 protein.
Figure 3.3-3. (A) Pearson product-moment correlation of liver ENPP1 protein relative abundance (log)
vs fasting plasma insulin concentration (log). (B) Pearson product-moment correlation of liver ENPP1
protein relative abundance (log) vs HOMA-IR (log).
95
3.3.3 Liver CEACAM-1 and IDE levels in diabetes
CECAM-1 and IDE are both involved in hepatic insulin clearance through their
influence on insulin internalization and degradation respectively. Expression levels of
CEACAM-1 and IDE mRNA were analysed in individuals with normal glucose
tolerance (NGT group) and individuals with type 2 diabetes mellitus (T2DM group) at
the time of RYGB surgery. CEACAM-1 and IDE mRNA expression data (2-∆Ct) was
log transformed prior to analysis with t-test in order to meet parametric test
assumptions. There were no statistically significant differences between the NGT and
T2DM group in CEACAM-1(Figure 3.3-4A) or IDE mRNA expression (Figure 3.34B) at the time of RYGB surgery.
Figure 3.3-4 Mean + SE CAECAM-1 mRNA expression (A) and IDE mRNA expression (B) presented
as 2-∆Ct normalised to 18S. NGT; Normal glucose tolerance (n=19), T2DM; Type 2 diabetes (n=27).
Protein abundance of CEACAM-1 and IDE was also assayed at the time of RYGB
surgery (Figure 3.3-5). The molecular weight of CEACAM-1 is 160-180 kd, whereas
the molecular weight of IDE is reportedly 118 kd. Relative protein abundance data
was log transformed prior to analysis with t-test. Although CEACAM-1 protein
abundance tended to be lower in the T2DM group, this difference was not statistically
significant. Similarly, there was no statistically significant difference in IDE protein
abundance at the time of RYGB surgery between the NGT and T2DM group.
96
Figure 3.3-5. CEACAM-1 and IDE protein abundance in liver tissue of morbidly obese individuals.
NGT; Normal glucose tolerance (n=17), T2DM; Type 2 diabetes (n=26). Mean + SE CEACAM-1 (A)
and IDE (B) protein abundance relative to Actin. (C) Corresponding western blot for CEACAM-1 and
IDE on all individuals who did not have a second liver biopsy. To avoid gel artefacts all samples to be
compared were loaded onto the same E-PAGE™ gel, while spaces delineate different gels. Samples
with an * were not included in analysis. For complete western blot films see Appendix.
3.3.4 Liver CEACAM-1 and IDE levels before and after RYGB surgery
Levels of CEACAM-1 and IDE mRNA were also analysed before and after RYGB
surgery in individuals who had normal glucose tolerance (sNGT group) and
individuals who had type 2 diabetes mellitus (sT2DM group). CEACAM-1 mRNA
expression did not change after RYGB surgery in the sNGT group or the sT2DM
group (Figure 3.3-6A). Likewise, there was no statistically significant change in IDE
mRNA expression in the sNGT group after RYGB surgery. However, for the sT2DM
group there was a significant decrease in IDE mRNA expression after RYGB surgery
(p=0.04 paired t-test) (Figure 3.3-6B). Consequently, individuals who had remission
of type 2 diabetes had a mean fold change of -1.56 (95% CI -4.5 to 1.4) in IDE
mRNA expression.
97
CEACAM-1 and IDE protein abundance was also analysed in the same individuals
before and after RYGB surgery as described previously (Figure 3.3-6C and D). There
were no statistically significant changes in CEACAM-1 or IDE protein abundance in
the sNGT or sT2DM group.
Figure 3.3-6. Liver CEACAM-1 and IDE gene expression and protein abundance in liver tissue of
morbidly obese individuals before and after RYGB surgery (● sNGT, n= 8; ○ sT2DM, n=8). (A) Mean
± SE CEACAM-1 expression presented as 2-∆Ct normalised to 18S. (B) Mean ± SE IDE mRNA
expression was significantly decreased in the sT2DM group after RYGB surgery (paired t-test). (C)
Mean ± SE CEACAM-1 and (D) IDE protein abundance relative to Actin. Protein was not available
from two individuals in the NGT and one individual in the T2DM group (E) Corresponding western
blot for IDE and CEACAM-1 for individuals that had a second liver biopsy. For complete western blot
film see Appendix.
98
3.4 Discussion
The purpose of this study was to investigate mRNA expression and protein abundance
of molecules responsible for regulating insulin action in type 2 diabetes. To achieve
this, levels of ENPP1, CEACAM-1, and IDE were assayed in individuals with normal
glucose tolerance or type 2 diabetes before and after RYGB surgery. ENPP1 mediates
IR activity, whereas CEACAM-1 and IDE are responsible for insulin internalization
and degradation respectively. Any changes in levels of these molecules are likely to
affect insulin action which may lead to or contribute to insulin resistance.
All individuals included in this study were morbidly obese (BMI>40 kg/m2) at the
time of RYGB surgery. As expected fasting plasma glucose and HbA1c % levels were
normal in individuals with normal glucose tolerance. However, individuals with type
2 diabetes had significantly higher fasting plasma glucose and HbA1c % values,
which returned to normal after RYGB surgery (Figure 2.3-3). Although insulin
resistance was present in the majority of individuals included in the study, those with
type 2 diabetes had significantly higher fasting plasma insulin concentrations and
HOMA-IR values, than those with NGT. RYGB surgery was associated with
improved insulin sensitivity in both groups of individuals, with normalisation of
fasting plasma insulin concentrations and HOMA-IR values being associated with
remission of type 2 diabetes (Figure 2.3-2 and 2.3-3).
ENPP1, CEACAM-1, and IDE expression was compared between individuals
grouped according to their glucose tolerance status, and hence insulin resistance.
Individuals with normal glucose tolerance (NGT and sNGT groups) were used as
controls.
ENPP1 inhibits insulin signalling by directly binding the α-subunit of the insulin
receptor.166 Evidence of elevated ENPP1 protein levels in muscle, adipose and skin of
insulin-resistant individuals has led to the hypothesis that this molecule may be
integral to the development of insulin resistance.167-170 If ENPP1 contributes to the
development of insulin resistance then it should be overexpressed in individuals with
advanced insulin resistance (T2DM group) and decrease with resolution of insulin
resistance (sT2DM group).
99
The principal role of CEACAM-1 in insulin action is to mediate insulin
internalization,
64-66
including insulin.63,
whereas IDE is responsible for degrading a number of substrates
498
Insulin internalization and degradation are both required for
effective insulin clearance at the liver, which others have shown to increase after
RYGB surgery in individuals who have a remission of diabetes.482 Changes in mRNA
expression and protein content of these genes after the remission of type 2 diabetes
(sT2DM group) may shed light on the mechanisms involved in hepatic insulin
clearance at the liver.
3.4.1 ENPP1
Although ENPP1 may contribute to the pathology of insulin resistance, conflicting
evidence has been reported in the literature. Overexpression of ENPP1 protein in the
muscle and liver of animal models leads to development of insulin resistance.499
However, in an animal model of diabetes there was no difference in ENPP1 muscle
and liver content in comparison to control animals.497 Even though some
experimentally produced type 2 diabetes animal models resulted in increased ENPP1
content500 others did not.501, 502 Furthermore, there was no change in muscle ENPP1
content after bariatric surgery where there had been a measured increase in insulin
sensitivity.412
There have been no studies reporting ENPP1 levels in the liver of human individuals
with or without type 2 diabetes. Considering the implications involved in pathological
overexpression of ENPP1, this study sought to address this mechanism. Although we
found ENPP1 mRNA expression to be lower in individuals with type 2 diabetes
(Figure 3.3-1A), there was no change in expression after remission of diabetes. This
may be due to variable protein turnover as studies have shown that in liver cells
insulin can increase ENPP1 protein abundance without changing its gene
expression.503 Nonetheless, there was a clear association between ENPP1 protein
abundance and type 2 diabetes. One of the key findings was that liver ENPP1 protein
abundance was 2 fold lower in individuals with type 2 diabetes in comparison to
individuals with normal glucose tolerance (Figure 3.3-1B). Furthermore, individuals
who had remission of type 2 diabetes had a 2.7 fold increase in ENPP1 protein
expression (Figure 3.3-2).
100
This data differs from previous studies which showed that ENPP1 levels in muscle
and adipose tissue were inversely correlated with insulin sensitivity.
167, 168, 504
We
suspect that the disagreeing evidence reported in the literature and our own data
presented here is due to the varied roles muscle, fat, and liver tissues have in
regulating insulin action.
The liver has a crucial role in regulating the systemic insulin concentration by hepatic
insulin clearance.56 It is the first organ to receive large amounts of insulin released
from the pancreas during the first phase response (1000-5000 pmol/l) and removes
~50% of it from the blood stream.62,
63
Considering the liver extracts and recycles
supraphysiological levels of insulin during first phase insulin secretion, preventing
overstimulation of metabolic pathways may be necessary for normal glucose
homeostasis. Evidence for a possible insulin self-desensitisation mechanism was first
provided by Menzaghi et al. (2003) who found that insulin induces a rapid increase in
ENPP1 at the cell membrane in liver cells.503 Additionally, individuals with insulin
receptor or IRS-1 mutations resulting in defunct insulin signalling had a significant
reduction in ENPP1 activity and content. 505
In this study, ENPP1 protein abundance was increased with an improvement in insulin
sensitivity in individuals who had a remission of type 2 diabetes and was inversely
correlated with both fasting plasma insulin concentration and HOMA-IR (Figure 3.33). The increase in ENPP1 levels after an improvement in insulin sensitivity is
consistent with the hypothesis that ENPP1 acts as a part of a hormone selfdesensitization mechanism. However the fact that it occurs only in individuals who
have a remission of type 2 diabetes is interesting and could be explained by the nature
of insulin secretion and its abrogation in type 2 diabetes.
The first-phase insulin response is triggered by an increase in glucose concentration
and markedly increases the insulin pulse mass.40 One of the earliest defects in
individuals who develop type 2 diabetes is a decreased first phase insulin pulse mass.
40, 48, 52
Restoration of the first-phase insulin response after gastric bypass surgery in
individuals who have a normalisation of insulin sensitivity and remission of type 2
diabetes has been documented.471 Consequently, the increase in ENPP1 levels in
individuals who have remission of type 2 diabetes after gastric surgery could be a
response to the return of the first phase insulin response.
101
If this is the explanation then ENPP1 is unlikely to be contributing to the development
of insulin resistance at the liver in morbidly obese individuals but instead may be a
compensatory response for increased insulin sensitivity after RYGB surgery.
However as this study does not address the K121Q polymorphism it can not be
concluded that ENPP1 does not have a role to play in liver insulin resistance. Recently
it has been shown that the relatively rare Q121 variant has a greater inhibitory effect
on insulin receptor than the more common K121 variant in insulin target cells506 and
may interact with adiposity to modulate glucose homeostasis
507
. But the question of
whether there is a direct link with type 2 diabetes is yet to be answered. Although the
K121Q polymorphism has been shown to be associated with type 2 diabetes in metaanalysis
508, 509
there is some conflicting evidence
510, 511
making it difficult to
conclusively link it with type 2 diabetes.
3.4.2 CEACAM-1 and IDE
Transgenic mice with a functional mutation in liver CEACAM-1 develop
hyperinsulinaemia as a result of impaired insulin clearance.65 This leads to secondary
insulin resistance, abnormal glucose tolerance, increased fatty acids, hepatic steatosis
and visceral adiposity.65, 70Consequently the ability of CEACAM-1 to regulate insulin
internalisation and hence clearance has been implicated in the pathogenesis of insulin
In this study CEACAM-1 mRNA and protein abundance did not change in varying
states of insulin resistance nor did it change after an increase in insulin sensitivity
following RYGB. Although CEACAM-1 protein content has been shown to be
decreased in liver of individuals with severe obesity or liver steatosis, it did not differ
between individuals with or without type 2 diabetes.512 This is in accordance with the
data presented here, suggesting that CEACAM-1 may not be directly involved in the
pathogenesis of type 2 diabetes.
IDE has also been implicated in the pathogenesis of insulin resistance. In human
studies, two IDE polymorphisms were associated with an increased risk of type 2
diabetes,74 whereas mice lacking IDE developed hyperinsulinaemia leading to
102
progressive insulin resistance and ultimately glucose intolerance.73 In this study there
were no significant differences in IDE gene expression or protein content between
individuals with normal glucose tolerance or type 2 diabetes. However, there was a
significant decrease in IDE mRNA expression in individuals who had a remission of
type 2 diabetes (Figure 3.3-6B). A study by Pivovarova et al. (2009) showed that
treating liver cells with high glucose concentrations can cause an increase in IDE
mRNA expression,513 which was associated with a loss of insulin-induced changes in
IDE activity. This provides a possible explanation for our findings.
Although the data presented here does not address IDE activity, reduced IDE mRNA
expression was only seen in those individuals that had a significant decrease in fasting
plasma glucose levels (Table 2-3). Taken together these findings suggest that
hyperglycaemia may provoke a disturbance in IDE activity in type 2 diabetes which
may lead to inefficient hepatic insulin clearance causing an increase in systemic
insulin concentration. This presents a potential molecular mechanism responsible for
driving the progressive insulin resistance seen in type 2 diabetes. However, in our
studies, IDE mRNA and protein did not have a statistically significant correlation with
fasting plasma glucose concentrations, suggesting the mechanisms by which high
glucose concentrations affect IDE-mediated degradation of insulin are complex.
Considering IDE also has a major role in degrading amyloid-β protein it is reasonable
to assume its activity may be regulated by components of both pathways.
In conclusion, the results from this study demonstrate a reverse ENPP1 pattern to that
previously described in muscle and adipose tissue of insulin-resistant individuals.
They suggest that ENPP1 acts as a desensitizer under conditions of insulin sensitivity
rather than being an important contributor to insulin resistance in the liver.
Furthermore, although individuals who undergo RYGB surgery have an increase in
hepatic insulin clearance, data presented here suggest that this is not associated with
changes in expression of CEACAM-1 or IDE.
103
104
CHAPTER FOUR: Liver insulin receptor isoforms and
their role in the pathogenesis of type 2 diabetes
105
4.1 Introduction
Insulin signalling occurs primarily through the insulin receptor.129, 130 In the 1980s the
discovery of the IR isoforms358 led to extensive research into the molecular
mechanism regulating insulin action. Since then, a massive body of evidence has been
gathered regarding post receptor events of the insulin signalling pathway. Despite this,
the regulation of these outcomes by the IR isoforms in insulin target tissue, such as
the liver, is still unclear.
Alternative splicing of exon 11 in the INSR gene generates two different protein
isoforms, IR-A (without exon 11) and IR-B (with exon 11).358 Because of its
mitogenic characteristics IR-A predominates in developmental tissue such as fetal
cells and in pathological conditions such as cancer.367,
368, 382
Conversely, IR-B is
expressed in differentiated adult cells and is predominant in insulin sensitive tissues
which regulate glucose homeostasis.365,
366
The high affinity for growth promoting
ligands367, 514 and a faster internalisation rate of the IR-insulin complex373, 374, 390 are
molecular mechanisms that are thought to account for the mitogenic properties of IRA activation. On the other hand, the slower internalisation rate390 and stronger
activation of the IR tyrosine kinase376-378 accounts for the metabolic function of IR-B
activation by insulin. Signalling through IR receptor isoforms is further complicated
by the formation of hybrid receptors. IR-A, IR-B and IGF-IR proreceptors generate
heterodimeric hybrids that have been shown to reduce the number of insulin binding
sites and may contribute to the pathogenesis of insulin resistance.404, 407
Multiple lines of evidence point to the liver being the vital organ in the development
of type 2 diabetes. First, unregulated hepatic gluconeogenesis is thought to be a
critical component of the fasting hyperglycaemia in late stage type 2 diabetes.19, 21, 23,
515
Second, tissue specific IR knockout studies in mice suggest that of the three insulin
target tissues involved in glucose homeostasis, the liver may be of primary
importance.111, 116, 123, 516 Neither adipose nor muscle insulin receptor knockout mice
have perturbations in glucose and insulin levels.112,
114, 517
But liver insulin receptor
knockout mice develop both hyperglycaemia and hyperinsulinemia which ultimately
leads to diabetes.116, 123 The liver also appears to be a vital organ in the remission of
type 2 diabetes after RYGB surgery. Liver insulin resistance as measured by HOMA-
106
IR has been shown to rapidly resolve after RYGB surgery, in the same timeframe as
remission of type 2 diabetes, whereas peripheral insulin resistance seems to only
improve sometime later and after substantial weight loss.448, 476, 477, 481
Although there is conflicting evidence in the literature,418, 419, 421, 423, 424, 426, 518 altered
relative expression of the two IR isoforms has previously been implicated in the
pathogenesis of type 2 diabetes. Most studies analysing these isoforms have involved
skeletal muscle and adipose tissue but relatively little work has been done in human
liver. Because there is reason to believe the liver may be the key organ in the
development of type 2 diabetes, we have chosen to investigate whether the insulin
receptor isoforms change in liver tissue after remission of type 2 diabetes.
Additionally, we explore whether any changes in the ratio of the two isoforms has
implications for liver insulin signalling.

IR protein content and isoform gene expression was assayed using western
blot and RT-qPCR in the NGT, T2DM, sNGT and sT2DM groups.

IR protein content and isoform gene expression in Hep G2 cells treated with
supraphysiological levels of insulin was assayed using western blot and RTqPCR.

Attempted siRNA-mediated knockdown of IR-A or IR-B to investigate their
functional characteristics regarding insulin signalling.

Overexpression of IR-A or IR-B in Hep G2 cells and assaying activation of
AKT (Ser473 phosphorylation) with western blot and levels of PCK1 mRNA
transcription using RT-qPCR after insulin treatment.
107
The cohort described in Chapter 2 was used in this study. This cohort included
individuals with normal glucose tolerance (NGT group, n= 19), and type 2 diabetes
(T2DM group, n=27). In addition we made use of liver tissue obtained from a subset
of individuals who were re-operated on subsequent to the RYBG, for unrelated
reasons, who had either normal glucose tolerance (sNGT, n=8) or type 2 diabetes
(sT2DM, n=8).The following methods pertain to analyses of IR-A and IR-B isoform
mRNA expression and IR protein content in these individuals
4.2.1 Analysis of IR-A and IR-B mRNA expression using RT-qPCR
For RNA extraction/quality control, first strand cDNA synthesis and thermocycling
conditions refer Chapter 3 Research Design and Methods. Briefly, all RT-qPCR
reactions were done using EXPRESS qPCR SuperMix (Life Technologies) and
TaqMan Gene Expression Assays (Applied Biosystems, USA). Single stranded cDNA
was analyzed with the ABI 7300 Real-Time PCR System (Applied Biosystems,
USA). Each sample was run in triplicate and each time a threshold cycle (Ct) was
obtained using the 7300 Sequence Detection Software 1.3.1. The average of all three
replicates for a particular sample was calculated and used in subsequent data analysis.
Custom designed IR-A (A127515) and IR-B (A1Q7CX) assays were used (Applied
Biosystems, USA). IR-A and IR-B differ by the presence or absence of a 36 base pair
sequence (Exon 11) making it difficult to assay using RT-qPCR with conventional
chemistries such as SYBR green. Because of their increased specificity, Custom
TaqMan Gene Expression Assays were designed and produced specifically for this
experiment. For both assays the forward and reverse primers were identical. Forward
Primer:
5’-
GATTACCTGCACAACGTGGTT-3’;
Reverse
Primer:
3’-
GCCAAGGGACCTGCGTTT- 5’.
108
Figure 4.2-1. IR-A and IR-B amplified region including MGM probe binding area. IR-A 6-FAM dye
labelled MGM binds across the exon 10/12 boundary. IR-B 6-FAM dye labelled MGM probe binds
across the exon 11/12 boundary.
The MGM probe for the TaqMan assay for IR-A crossed the exon 10/12 boundary
while the MGM probe for IR-B crossed the exon 11/12 boundary (Figure 4.2-1).
Amplification efficiency of IR-A and IR-B probes were assayed using the Ct slope
method and Ex = 10(-1/slope)-1 equation as described in Chapter 3 Research Design and
Methods. Amplification efficiency for IR-A was 100% while the efficiency for IR-B
was 98%. Changes in gene expression were expressed as relative amount normalised
to a reference gene and as fold change, both of which were calculated using the RQ
method (2-∆∆Ct). Eukaryotic 18S rRNA (4319413E, Applied Biosystems) was used as
the reference gene.
4.2.2 Analysis of IR protein abundance using E-PAGE electrophoresis and
western blotting
Insulin receptor protein abundance was analysed in human liver tissue as described in
detail in Chapter 3 Research Design and Methods. Briefly, 15 µg of protein per
sample was resolved by E-PAGE electrophoresis (Life Technologies) and transferred
to Hybond-P PVDF membrane (Amersham). Membranes were blocked with 5% skim
milk powder in TBT-T at 4°C overnight and then probed for 2 h at room temperature
with Anti-Insulin Receptor β-subunit antibody (Clone CT-3, 1:500, Millipore) Blots
were then incubated with alkaline phosphatase-conjugated anti mouse antibody (Life
Technologies) for 0.5 h. The chemiluminescent signal was developed using CDP-Star
chemiluminescent substrate (Life Technologies) and captured on x-ray film (Kodak).
The protein bands on the x-ray film were digitized by the Chemidoc XRS system (Bio
109
Rad) and then analyzed using Quantity One software (Bio Rad). Bands were
quantified and expressed as volume quantity; the sum of the intensities of the pixels
within a defined volume boundary × pixel area (intensity units × mm2). Local
background subtraction was used to facilitate background correction. Insulin Receptor
protein relative abundance was calculated by normalizing the intensity of the IR to the
intensity of the Actin band.
4.2.3 Functional studies of insulin receptor and its isoforms in cultured liver cells
The Hep G2 cell line was purchased from the American Type Culture Collection
(ATCC number: HB-8065, Manassas, VA, USA). This cell line was derived from the
liver tissue of a 15 year old male with differentiated hepatocellular carcinoma. Hep
G2 cells are adherent epithelial-like cells that grow as a monolayer and in small
aggregates.
4.2.3.1 Propagation and storage
Hep G2 cells were maintained in a 500cm2 Square TC-Treated Culture Dish
(Corning) in Dulbecco’s modified eagle medium (DMEM) (Life Technologies),
supplemented with 10% fetal calf serum (FCS) (Life Technologies, NZ) and
antibiotic/antimycotic (penicillin 100units/ml, streptomycin 100µg/ml and fungizone
0.25µg/ml - Life Technologies) at 37ºC and 5% CO2. From this point forward the
acronym
DMEM
denotes
DMEM
supplemented
with
10%
FCS
and
antibiotic/antimycotic.
At 70-80% confluence, cells were detached with Trypsin/EDTA solution (TE) (Life
Technologies). The trypsin was inactivated with an equal volume of DMEM and
centrifuged at 1100g for 5 minutes. The cell pellet was washed with DMEM to
remove residual TE solution and re-centrifuged at 1100g for 5 minutes. The washed
pellet was re-suspended in 6mL of DMEM and then combined with a further 6mL of
DMEM with 10% (v/v) DMSO, which was added dropwise to the cell suspension for
a final volume of 12mL and a 5% (v/v) DMSO concentration.
110
1mL of this cell suspension was aliquoted into 1.5mL cryovials (Nalgene) and cooled
to -80ºC at a rate of 1ºC/min in a Cryo 1ºC freezing container (Nalgene) containing
250mL isopropanol. Cells were transferred to a liquid nitrogen dewar for long term
storage in liquid nitrogen.
4.2.3.2 In Vitro experiments
Depending on the amount of cells need, 1-2 cryovials were brought out of storage and
defrosted until the cell mixture reached an ice slush state. At this point the cell slurry
was transferred to either a 25cm2 (Corning) or a 75cm2 (Labcon) TC-Treated angled
neck flask containing DMEM warmed to 37ºC. Cells were left to attach for 6-12 hours
after which media was refreshed. The following day, cells were trypsinized, counted
in a Neubauer Counting Chamber and used in subsequent experiments.
4.2.3.3 RNA extraction from cultured cells
For cultured cell monolayers, media was removed and cells were washed in 1mL
phosphate-buffered saline (PBS). 800µL of TRIzole® (Life Technologies) was added
per well, homogenised and incubated at room temperature for 5 min before transfer to
RNAase/DNAase free eppendorf tubes. RNA was extracted as described in Chapter 3
Research Design and Methods.
4.2.3.4 RNA quantification
RNA quantity was measured using a NanoDrop ND-1000 Spectrophotometer. The
absorbance of the RNA sample was measured between 260 and 280nm. An A260
reading of 1.0 is equivalent to 40µg/ml of RNA. The concentration of nucleic acid
was determined using the Beer-Lambert law which predicts a liner change in
absorbance with concentration. The absorbance maximum of RNA is at 260nm and
the ratio of the absorbance at 260 and 280nm gives an indication of RNA purity. Only
RNA with OD260/OD280 ratio greater than 1.8 was used.
111
4.2.3.5 RT-qPCR of IR-A/IR-B and PCK1 in Hep G2 cells
The principle and most of the methodology was identical to the methods described in
Chapter 3 Research Design and Methods unless otherwise stated.
Total RNA was reverse transcribed using the iScript Advanced cDNA Synthesis kit
for RT-qPCR (Bio Rad). 1 μg of RNA was added to a 20μl reaction containing iScript
MMLV reverse transcription enzyme, random primers, dNTPs, oligo(dT), buffer and
RNase inhibitor. Reactions were incubated at 42ºC for 30min, then at 85ºC for 5
minutes. The resultant cDNA was diluted 1:10 and stored at -20ºC prior to use in RTqPCR reactions.
All RT-qPCR reactions were done using TaqMan Gene Expression Master Mix and
TaqMan Gene Expression Assays (Applied Biosystems, USA). Custom designed IRA (A127515)/IR-B (A1Q7CX) assays and PCK1 TaqMan Gene Expression Assay on
Demand (Hs00159918_m1) were used (Applied Biosystems). Single stranded cDNA
was analyzed with the ABI 7300 Real-Time PCR System (Applied Biosystems USA).
Each sample was run in duplicate and each time a threshold cycle (Ct) was obtained
using the 7300 Sequence Detection Software 1.3.1. The average of replicates for a
particular sample was calculated and used in subsequent data analysis.
4.2.3.6 Protein extraction
For cultured cell monolayers, media was removed and cells were washed in 1mL
PBS. Cells were lysed in the well by adding 100μl Laemmli buffer (62.5mM Tris-Hcl,
2% SDS, 25% glycerol, 5% β-mercaptoethanol, 0.01% bromophenol blue) containing
Complete Protease Inhibitor Cocktail (Roche) and Phosphatase Inhibitor Cocktail C
(Santa Cruz). The cells were homogenised and incubated for 5 minutes. The lysate
was transferred to an eppendorf tube and incubated at 100ºC for 5 minutes.
Centrifugation was performed at 14,000 g for 10 minutes at 4ºC to remove insoluble
debris. The supernatant was aliquoted and stored at -80ºC prior to SDS-PAGE.
112
4.2.3.7 SDS-PAGE
SDS-PAGE gels were cast using the Mini-Protean Tetra Casting Module (Bio Rad).
Each gel had 10 wells with a 4% stacking layer (4% Acrylamide/Bis, 0.125M Tris, pH
6.8, 0.1% SDS, 0.1% TEMED, 0.05% APS) and a 10% resolving layer (10%
Acrylamide/Bis, 0.375M Tris, pH 8.8, 0.1% SDS, 0.1% TEMED, 0.05% APS).
Samples were resolved using the Mini-PROTEAN 3 cell (Bio Rad). 20µL of lysate
was added to each well and run at 100V until the dye front reached the resolving layer
at which point the voltage was increased to 160V until the dye front reached the end
of the gel.
4.2.3.8 Electrophoretic transfer to PVDF and western blotting
Immediately after resolution gels were equilibrated in transfer buffer (39 mM
Glycine, 48 mM Tris, methanol 20% v/v) for 10 minutes. Hybond-P PVDF membrane
(Amersham) was activated with methanol for 10 seconds and washed in deionised
water for 5 minutes after which it was equilibrated in transfer buffer for 10 minutes.
Proteins were transferred from the gel to the Hybond-P PVDF membrane with a BioRad Mini Trans-Blot Cell for 2-3hours at 50V. Transfer of proteins was confirmed
with Ponceau S stain. The membrane was cut according to molecular weight which
allowed for probing of different proteins simultaneously. Membranes were blocked
with 5% skim milk powder in TBS-Tween at 4°C overnight and then probed for 2 h at
RT with primary antibody. Blots were then incubated with alkaline phosphataseconjugated anti mouse or rabbit antibody (Life Technologies) for 0.5 h. The
chemiluminescent signal was developed using CDP-Star chemiluminescent substrate
(Life Technologies) and captured on x-ray film (Kodak).
4.2.3.9 Stripping
PVDF membranes were only stripped once and re-probed for a different protein.
Membranes were incubated in stripping buffer (100 mM β-mercaptoethanol, 2% (w/v)
sodium dodecyl sulphate, 62.5 mM Tris-HCL pH 6.7) at 60°C for 30 min with gentle
agitation every 10 min. After rinsing with water, membranes were washed 3 times in
TBS-T for 15 min at room temperature using larger volumes of wash buffer.
Membranes were re-blocked overnight and re-probed the following day.
113
4.2.4 Transient Knockdown of IR-A and IR-B isoform expression using siRNA
Small interfering RNAs (siRNA) ~ 21 to 23 nucleotides in length induce sequencespecific gene silencing. The efficiency of gene silencing is dependant on the
susceptibility of target transcripts to siRNA-mediated RNA-induced silencing
complex (RISC) activity.519 Typically two or more siRNA sequences with high
prediction scores against the target sequence are chosen for knockdown of the target
gene. 519, 520 However due to the short sequence of exon 11 (36 base pairs) the design
of completely different siRNA sequences was not possible. The GeneAssist Custom
siRNA Builder (Applied Biosystems) was used to design Silencer Select siRNA for
IR-A and IR-B isoform by providing the target sequence. Pre designed and validated
Silencer Select siRNA (Applied Biosystems) for Insulin Receptor (s7477 and s7479)
were used for comparison.
Table 4-1. IR-A and IR-B siRNA sequence
Target Sequence Input
SiRNA Sequence
(5’->3’, sense)
SiRNA Sequence
(5’->3’, antisense)
IR-B
siRNA
(s239121)
AAAAACCUCUUCAGGCACUGGU
GCCGAGGACCCUAG ( insr Eeon
11)
AAACCUCUUCAG
GCACUGGtt
CCAGUGCCUGAA
GAGGUUUtt
IR-B
siRNA
(s239120)
AAAAACCUCUUCAGGCACUGGU
GCCGAGGACCCUAG ( insr exon 11)
AACCUCUUCAGG
CACUGGUtt
ACCAGUGCCUGA
AGAGGUUtt
IR-A
siRNA
(s244179)
AACGUGGUUUUCGUCCCCAGGG
AGUCCUCGUUUAGGAAAGA(insr
exon 10/12)
AGGCCAUCUCGG
AAACGCAtt
UGCGUUUCCGgU
GGCCUgg
4.2.4.1 siRNA transfection
Hep G2 cells were transfected using RNAiMAX (Life Technologies), a cationic lipid
transfection agent. There are two different protocols for transfecting cells in culture,
forward and reverse transfection. Traditionally forward transfection has been more
common. In forward transfection cells are first plated in multi-well plates and left to
adhere. Transfection complexes are prepared and added the following day. With the
advent of siRNA and high throughput screening, reverse transfection has been
developed.
114
In this protocol transfection complexes are prepared and added to multi-well plates
after which cells and media are added. Reverse transfection is quicker than forward
transfection and has been found to increase transfection efficiency in Hep G2 cells.
Commercial companies such as Life Technologies who provide transfection agents
such as Lipofectamine and RNAiMAX state that Hep G2 cells must be reverse
transfected with RNAiMAX (Life Technologies) to achieve adequate transfection
efficiency.521 This likely due to the increased cell surface area exposed to the
transfection complex in suspension in comparison to monolayer. For this reason,
reverse transfection with RNAiMAX was used to transfect siRNA into Hep G2 cells.
4.2.4.2 Reverse transfection of siRNA
Transfection complexes were prepared in sterile RNAase/DNAase eppendorf tubes in
master mix format for each siRNA and then added to a 24-well TC-Treated plate
(Labcon). The volumes used in the following description are in per well format. The
transfection complex volume for each well was 100µL. Opti-MEM (Life
Technologies), a modified DMEM media with reduced serum content, was used to
dilute both the siRNA and RNAiMAX (Life Technologies). siRNA (1.5µL of 20mM
stock, 50nM final well concentration) was diluted in 50µL of Opti-MEM and mixed
gently. RNAiMAX was diluted in 50µL of Opti-MEM and combined with the siRNA
dilution. The transfection complex was incubated at room temperature for 20 minutes.
During this incubation, Hep G2 cells were trypsinized and centrifuged at 1100 g for 5
minutes. The cell pellet was re-suspended in DMEM without antibiotic/antimycotic
and counted in a Neubauer Counting Chamber.
The cell suspension was diluted to 100 cells/µL. Transfection complex was added to
the bottom of each well after which 500µL of cell suspension was added to each well
for a final concentration of 4 x 104 cells/well. Cells were incubated at 37ºC and 5%
CO2 for 24 hours after which media was refreshed. Gene and protein knockdown was
assayed 48 and 72 hours after transfection respectively using RT-qPCR and western
blotting as described previously. All knockdown experiments were done in duplicate.
Experimental controls were run in parallel which included negative control (scramble
siRNA, Applied Biosystems) and mock transfection.
115
Silencer Select GAPDH (Applied Biosystems) was used to optimize transfection
conditions and as a positive control.
4.2.5 IR-A and IR-B Isoform overexpression vector
pcDNA3.1+ vector containing IR-B cDNA (a kind gift from Ronald Kahn Lab, Joslin
Diabetes Centre, Boston, USA) was used to over express IR-B isoform in Hep G2
cells. The vector was received dried on Whatman #1 filter paper from Prof. Conan
Fee at the Biomolecular Interaction Centre, Christchurch, New Zealand. To retrieve
the vector the marked circular area that contained it was cut and inserted into a 1.5mL
RNAase/DNAase free eppendorf tube. 100µL of TE buffer (10mM Tris, 1 mM
EDTA, pH 8.0) was added after which the tube was vortexed and incubated at room
temperature for 5 minutes. After the incubation it was vortexed again and centrifuged
at 1000g for 1 minute. The resulting supernatant was used to transform E. coli for
propagation of the vector.
4.2.5.1 Transformation
Prior to thawing cells, 10µL of vector was added to an eppendorf tube and chilled on
ice. Subcloning Efficiency DH5α (E. coli) Competent cells (Life Technologies) were
thawed on ice and gently mixed with pre-chilled pipette tips. 50µL was aspirated and
added to either the tube containing the vector or to an empty pre-chilled 1.5mL
eppendorf tube for storage at -80ºC. The cells were gently mixed and incubated on ice
for 30 minutes. The cells were then heat shocked in a 42ºC water bath for exactly 20
seconds and immediately cooled on ice for 2 minutes.
Following the 2 minute cooling period, 950µL of pre-warmed (37ºC) SOC medium
(tryptone 10g/L, yeast extract 5g/L, NaCl 5g/l, 20mM glucose, Sigma Aldrich) was
added and incubated at 37ºC for 1 hour at 225 rpm. After the 1 hour incubation with
SOC medium, 50 and 100µL of bacterial suspension was spread onto two different
LB agar (Life Technologies) plates containing 50µg/mL ampicilin (Sigma Aldrich)
and incubated at 37ºC overnight. The following day bacteria were evenly spread
across the plate in isolated colonies.
116
4.2.5.2 Vector propagation and extraction
Several colonies were picked with a sterile pipette tip and placed in separate 50mL
tubes (BD Falcon) containing 5mL LB medium (Sigma Aldrich) with 50µg/mL
ampicilin and incubated at 37ºC overnight at 225rpm. The following day, 500µL of
cell suspension was diluted with 500µL of LB medium containing 50% glycerol (final
concentration 25%) and stored at -80ºC as stock.
The remaining bacterial suspension was centrifuged at 1500g for 5 minutes and the
plasmid was extracted from the resulting pellet using the EZNA Plasmid Miniprep Kit
I (Omega Bio-Tek). This kit is based on the DNA extraction method established by
Birnboim et al. (1979)522 in which alkaline conditions are used to selectively
precipitate and remove chromosomal bacterial DNA whilst retaining plasmid DNA.
Additionally HiBind columns that can reversibly bind DNA are used for efficient
purification. Plasmid DNA that was bound to the HiBind column was eluted with TE
buffer.
Plasmid
DNA
was
quantified
using
the
NanoDrop
ND-1000
Spectrophotometer and confirmed using both enzyme digestion and PCR.
4.2.5.3 Plasmid enzyme digestion
The IR-B cDNA size was briefly confirmed by enzyme digestion. pcDNA3.1 IR-B
plasmid was digested with XbaI and HindIII (Fermentas) according to the following
protocol; 13.75µL RNAase/DNAase free H2O, 2µL HindIII, 1µL XbaI, 2µL Tango
buffer (10x stock) and 1.25ul DNA (800ng/µL) was added to a tube and incubated at
37ºC for 2.5 hours. The digest was separated on a 1% TAE Agarose gel and resolved
using a Mini SUB Gel GT electrophoresis chamber (Bio Rad) at 5V/cm.
HyperLadder I (Bioline) was used as a DNA marker. Bands were visualised after
staining with Ethidium Bromide (Sigma Aldrich) under UV light and images were
recorded using the Chemidoc XRS system (Bio Rad).
117
Figure 4.2-2. Map of pcDNA3.1+ vector containing human IR-B clone.
4.2.5.4 PCR of linear plasmid
PCR was performed on liner plasmid DNA. The primer pairs for PCR of the INSR
gene were designed with the Gene Link Oligo Explorer software using the consensus
coding sequence (CCDS) obtained from the NCBI reference sequence database. The
primers were designed to target the 5’ and 3’ end of the sequence. Further nucleotides
were added to create restriction sites for HindIII and XbaI with a view to cloning the
IR-A isoform into the pcDNA 3.1 vector. The forward primer sequence was further
manipulated to add a Kozak sequence for increased translational efficiency.523,
524
Primer sequences are shown in Figure 4.2-3.
Figure 4.2-3. Forward and reverse primer sequences for confirmation of IR-B cDNA clone and further
cloning of IR-A cDNA designed using the insulin receptor consensus coding sequence (INSR CCDS).
Italicized nucleotides are enzyme restriction sites.
118
PCR was performed using BioMix Red PCR master mix (Bio Line). Due to the high
G/C content in the first 50bp of the INSR gene (88%) and the forward primer (66%)
DMSO was used as an additive at a final concentration of 5%. To prevent spurious
priming PCR cycling conditions were optimised using the Touchdown technique as
reported by Don et al. (1991).525 The nearest-neighbour theory of DNA duplex
stability was used to calculate melting temperature (Tm).526 Briefly, overall duplex
stability or otherwise melting temperature of an oligonucleotide can be predicted from
the primary sequence using the stability and temperature dependant behaviour of
every dinucleotide pair in the oligo. Using Gene Link Oligo Explorer software, the Tm
was estimated to be 81ºC and 75ºC for the forward and reverse primer respectively.
Each 50µL PCR reaction contained 18µL RNAase/DNAase free H2O, 2µL DNA,
1.25µL forward and reverse primer (20µM stock, 0.5µM final concentration), 2.5µL
DMSO and 25µL BioMix Red master mix. Thermocycling parameters are shown in
Table 4-2. The PCR product was separated on a 1% TAE agarose gel and resolved
using a Mini SUB Gel GT electrophoresis chamber (Bio Rad) at 5V/cm. HyperLadder
I (Bioline) was used as a DNA marker. Bands were visualised after staining with
Ethidium Bromide (Sigma Aldrich) under UV light and images were recorded using
the Chemidoc XRS system (Bio Rad).
Table 4-2. Touchdown PCR thermocycling parameters for insulin receptor.
Touchdown Phase
(decrease annealing temperature by 1ºC every
cycle for 17 cycles)
Cycle Step
Initial
denaturation
Temperature
95ºC
Time
5 min
Cycles
1
Denaturation
Annealing
Extension
95ºC
85ºC
72ºC
30 sec
1 min
4.30
min
17
Denaturation
Annealing
Extension
95ºC
69ºC
72ºC
30 sec
1 min
4.30
min
17
Final extension
72ºC
7 min
1
Hold
4ºC
∞
1
119
4.2.5.5 Attempted cloning of IR-A
Hep G2 cells were grown in a 25cm2 TC-treated tissue culture flask (Corning) in
DMEM (Life Technologies), supplemented with 10% foetal calf serum (FCS) (Life
technologies, NZ) at 37ºC and 5% CO2. RNA was extracted using TRIzole® as
described previously. First strand synthesis was performed using SuperScript III FirstStrand Synthesis System for RT-qPCR (Life Technologies). The reaction contained
5µg RNA, 2.5µM Oligo(dT)20, 10mM dNTP mix, 5mM MgCl2, 10mM DTT, 40U
RNase OUT, and 200U SuperScript III. Reactions were incubated at 50ºC for 50 min
and terminated at 85ºC for 5 min. cDNA synthesised from Hep G2 RNA contains a
mix of IR-A and IR-B cDNA, both of which would be amplified during PCR.
Using Hep G2 cDNA as template, insulin receptor PCR was executed using the
Finnzyme Phusion High-Fidelity PCR kit (Thermo Fischer) and thermocycling
conditions presented in Table 4-3. Phusion DNA polymerase is a Pyrococcus-like
enzyme which has been shown to have 50 times more fidelity than the more
traditional Thermus aquaticus DNA polymerase. As such it was reasoned it would be
a superior choice for cloning. Each 50 reaction contained 5µL cDNA, 10µL of 5x
High Fidelity Phusion buffer, 1µL of 10mM dNTP (200µM final concentration),
2.5µL of 20µM forward and reverse primer (0.5µM final concentration), 2.5µL
DMSO (5%) and 0.5µL of Phusion DNA polymerase (0.02U/µL).
Table 4-3. Phusion High Fidelity PCR thermocycling
parameters for insulin receptor
Cycle Step
Temperature (ºC) Time
Cycles
Initial denaturation
95
1
30 sec
Denaturation
95
Annealing/Extension 72
30 sec
30
2.30 min
Final extension
72
10 min
1
Hold
4
∞
1
The optimal annealing temperature for Phusion Hot Start DNA polymerase differs
significantly from Taq-based polymerases. Consequently, the Tm for the primer set
described in Figure 4.2-3 was recalculated using the Tm calculator available on the
Finnzyme’s website (www.finnzyme.com)
120
The Tm was estimated to be 88ºC and 78ºC for the forward and reverse primer
respectively. Because the annealing temperature for both the primers exceeded 72ºC,
a two step protocol where the annealing and extension steps are combined was used.
4.2.5.6 OriGene IR-A expression vector
pCMV6-XL4 vector containing IR-A cDNA clone was used to overexpress IR-A
isoform (OriGene Technologies) (Figure 4.2-4). The pCMV6-XL4 IR-A clone was
derived from a single E. coli colony and arrived as a transfection ready plasmid. It
was received as 10µg of lyophilized plasmid in a 2-D Matrix tube. Plasmid was
reconstituted by adding 100µL of sterile ultra pure H2O giving a final concentration of
100ng/µL. Subcloning Efficiency DH5α (E. coli) were transformed with 5µL of IR-A
plasmid. It was then propagated and extracted as described previously.
Figure 4.2-4. Map of pCMV6-XL4 vector containing human IR-A clone.
4.2.5.7 Plasmid transfection
IR-A and IR-B plasmids were transfected into Hep G2 cells using branched
Polyethylenimine (PEI) (Sigma Aldrich). PEI is a stable cationic polymer with
ethylenimine motifs giving it a positively charged backbone.527, 528 This allows it to
bind negatively charged DNA which forms the PEI-DNA complex. This complex can
bind to the cell surface and is internalised via endosomal vesicles into the cell
cytoplasm.529
121
Stock solutions of PEI were prepared by diluting 1mg of branched PEI in 1mL of
sterile RNAase/DNAase free H2O. This solution was then neutralized with HCl and
sterilised by filtration through a 0.22µM filter.527 Aliquotes were stored at -80ºC.
Transfection complexes were prepared in sterile RNAase/DNAase eppendorf tubes in
master mix format for each plasmid and then added to a 24-well TC-Treated plate
(Labcon). The volumes used in the following method description are in per well
format. The transfection complex volume for each well was 100µL. Opti-MEM (Life
Technologies) was used to dilute both the plasmid and PEI (Sigma Aldrich). IR-A or
IR-B (800µg) was diluted in 50µL of Opti-MEM and mixed gently. 2µL of PEI was
diluted in 50µL of Opti-MEM and combined with the plasmid dilution. The
transfection complex was incubated at room temperature for 20 minutes. During this
incubation, Hep G2 cells were trypsinized and centrifuged at 1100 g for 5 minutes.
The cell pellet was re-suspended in DMEM without antibiotic/antimycotic and
counted in a Neubauer Counting Chamber. The cell suspension was diluted to 100
cells/µL. Transfection complex was added to the bottom of each well after which
500µL of cell suspension was added to each well for a final concentration of 15 x 104
cells/well. Cells were incubated at 37ºC and 5% CO2 for 24 hours after which media
was refreshed. Experimental manipulation was conducted 48 hours after transfection.
4.2.6 Insulin treatment of Hep G2 cells
Cells were seeded at 5 x 104/well in a 24-well TC-Treated plate (Labcon) and allowed
to attach for 24 hours. Following the 24 hours incubation, insulin (Life Technologies)
was added to the medium of four wells for a final concentration of 100nM. The
insulin was replenished every 48 hours for 6 days at which point both total RNA and
protein was extracted. A further four wells without insulin treatments were used as
controls. Total RNA was converted to cDNA as described previously. RT-qPCR was
performed using Custom Design TaqMan assays for IR-A and IR-B isoform.
Eukaryotic 18s was used as the reference gene. Insulin Receptor protein abundance
was assayed by western blot using Anti-Insulin Receptor β-subunit antibody (Clone
CT-3 1:500, Millipore). Pan-actin (Clone C4 1:10000 Millipore) was used as a
loading control. Relative quantification of protein abundance was done as described
previously.
122
4.2.7 Insulin-induced AKT phosphorylation in cells overexpressing IR-A or IR-B
Experiments were done in triplicate on 3 different days using Hep G2 cells
overexpressing IR-A or IR-B isoform and cells transfected with empty pcDNA3.1
vector (control). Cells were serum starved 24 hours prior to experimental
manipulation. Cells were stimulated with 100nM insulin for 5 minutes and lysed
directly in the well as described previously. Cells not treated with insulin (baseline)
were run in parallel. Following 10% SDS-PAGE resolution and transfer to
nitrocellulose, membranes were blocked with 5% skim milk powder in TBS-T at 4°C
overnight.
Blots were probed with primary antibody diluted in TBS-T for 2 hours at room
temperature. Phosphorylated AKT (p-AKT) was assayed using p-Akt1 (Clone 11E6,
1:800, Santa Cruz), a monoclonal antibody corresponding to the phosphorylated
Serine 473 residue of human AKT. Anti-Insulin Receptor β-subunit antibody (Clone
CT-3, 1:500, Millipore) and Pan-actin (Clone C4, 1:10000, Millipore) were used to
confirm overexpression and as a loading control respectively. Blots were stripped as
described previously and re-probed for total AKT (t- AKT) using a polyclonal AKT
antibody
(Clone
H-136,
1:2000,
Santa
Cruz).
Relative
quantification
of
phosphorylated AKT protein was calculated by normalizing the intensity of the pAKT band to the intensity of the t- AKT band.
4.2.8 Insulin-induced inhibition of PCK1 transcription in cells overexpressing
IR-A or IR-B
Experiments were done in triplicate on two different days using Hep G2 cells
overexpressing IR-A or IR-B isoform and cells transfected with empty pcDNA3.1
vector (control). Cells were serum starved 24 hours prior to stimulation with 1µM
insulin for 12 hours. Cells not treated with insulin (baseline) were run in parallel.
RNA extraction, cDNA synthesis and RT-qPCR were performed as described
previously. No-reverse transcriptase controls were run to control for genomic DNA
contamination.
123
PCK1 TaqMan Gene Expression Assay on Demand (Hs00159918_m1, Applied
Biosystems) was used to assay for PCK1 mRNA levels. Eukaryotic 18s was used as a
reference gene. Gene expression was presented as fold change according to the 2-∆∆Ct
method. Gene expression is also presented as 2-∆Ct which allowed for visualisation of
relative amounts of gene expression as individual data points.
Statistical analysis was performed on 2-ΔCt RT-qPCR data. Consequently gene
expression data was expressed as both 2-ΔCt in graphical form, whereas the fold change
for significant differences was reported in text. Statistical analysis of protein
abundance was performed on the value generated from normalisation of band
intensity to the loading control.
Normal distribution of all groups to be compared was tested with D’Agostino-Pearson
test. Non-normally distributed data was log transformed to comply with parametric
test assumptions. Pre-post comparisons were carried out using a paired t-test.
Student’s t-test was used to test significance between two groups. One way ANOVA
was used to test significance between multiple groups. The Pearson product-moment
correlation coefficient was used to test the strength and significance of association
between variables. An alpha level of 0.05 was set as the significance threshold. All
analysis was performed using Minitab15. Graphical visualization was performed
using GraphPad Prism 5.
124
4.3 Results
4.3.1 Liver insulin receptor protein abundance in individuals with type 2
diabetes
Liver IR protein was analysed in individuals who had normal glucose tolerance (NGT
group) and type 2 diabetes mellitus (T2DM group) as described previously. IR
relative abundance was log transformed prior to analysis with the student’s t-test. The
T2DM group had a lower mean IR protein abundance in comparison to the NGT
group (Figure 4.3-1A), which was statistically significant (p=0.02). Consequently, in
this study, individuals with type 2 diabetes had a -1.3 (95% CI, -1.6 to -1.1) fold lower
abundance of IR at the time of RYGB surgery than individuals with normal glucose
tolerance. Notably, at the time of RYGB surgery, the T2DM group had mean fasting
plasma insulin of 196 pmol/l, whereas the NGT group had mean insulin of 109
pmol/l.
Figure 4.3-1. (A) Mean ± SE IR protein abundance in liver tissue of morbidly obese individuals at the
time of RYGB surgery. NGT; Normal glucose tolerance (n=17), T2DM; Type 2 diabetes (n=26). (B)
Corresponding western blot for individuals that did not have a second liver biopsy. To avoid gel
artefacts all samples to be compared were loaded onto the same E-PAGE™ gel, while spaces delineate
different gels. Samples with an * were not included in analysis. For complete western blot film see
Appendix.
125
Consequently, IR protein abundance was associated with circulating insulin and
HOMA-IR (Figure 4.3-2). Data from the NGT and T2DM groups was combined and
log transformed prior to calculation of a Pearson product-moment correlation. Insulin
Receptor abundance negatively correlated with fasting plasma insulin concentration
(Figure 4.3-2A) and HOMA-IR index (Figure 4.3-2B).
Figure 4.3-2. (A) Pearson product-moment correlation of liver IR protein relative abundance (log) vs
fasting plasma insulin concentration (log). (B) Pearson product-moment correlation of liver IR protein
relative abundance (log) vs HOMA-IR (log).
4.3.2 Liver insulin receptor protein abundance after RYGB surgery
Figure 4.3-3A presents the mean relative IR protein abundance at the time of RYGB
surgery and approximately 16 months after. Both the sNGT and sT2DM group tend to
have an increase in IR protein after RYGB surgery, although the difference does not
reach statistical significance for the sNGT (p=0.183, paired t-test) or the sT2DM
group (p=0.08, paired t-test). However, regrouping the data into pre and post RYGB
groups produces a statistically significant (p=0.01, paired t-test) difference, most
likely due to the increased sample size. Consequently, there was a significant 1.4 fold
(95% CI, 1.1 to 1.8) increase in IR content after RYGB surgery. Increased IR content
was associated with a decrease in mean circulating insulin concentrations in both the
sNGT (80 to 27 pmol/l) and sT2DM (219 to 48 pmol/l) groups.
126
Figure 4.3-3. (A) Mean ± SE IR protein abundance in liver tissue of morbidly obese individuals before
and after RYGB surgery (● sNGT, n = 6; ○ sT2DM, n = 7). Mean IR abundance increased significantly
after RYGB surgery (paired t-test). (B) Corresponding western blot for IR for individuals that had a
second liver biopsy. To avoid gel artefacts all samples to be compared were loaded onto the same EPAGE™ gel, while spaces delineate different gels. For complete western blot film see Appendix.
4.3.3 IR-B:A mRNA ratio in individuals with type 2 diabetes
Liver IR-A and IR-B mRNA expression was assayed at the time of RYGB surgery.
Relative expression levels of IR-A and IR-B mRNA are presented as 2-∆Ct in Figure
4.3-4B and C. Neither IR-A or IR-B transcript showed significant difference among
NGT and T2DM groups, although IR-A expression tended to be higher in the T2DM
group (Figure 4.3-4B). The ratio of IR-B:A was calculated by dividing the IR-B 2-∆Ct
values by the IR-A 2-∆Ct values. The mean IR-B:A ratio in the T2DM group (5.2) was
significantly lower than the mean IR-B:A ratio in the NGT group (6.6), (p = 0.04, ttest).
127
Figure 4.3-4. Insulin receptor A and B isoform mRNA expression levels in liver tissue of morbidly
obese individuals. NGT; Normal glucose tolerance (n=19), T2DM; Type 2 diabetes (n=27). (A)
Mean + SE IR-B:A mRNA expression ratio calculated by IR-B 2-∆Ct /IR-A 2-∆Ct. The T2DM group
had a lower mean IR-B:A ratio than the NGT group, which was statistically significant (t-test). (B)
Mean + SE IR-A mRNA expression and (C) IR-B mRNA expression presented as 2-∆Ct normalised to
18S.
4.3.4 IR-B:A mRNA ratio before and after RYGB surgery
Expression levels of IR-A and B were also analysed in individuals who had normal
glucose tolerance (sNGT) or type 2 diabetes (sT2DM) before and after RYGB
surgery. The fold change was calculated using the 2-∆∆Ct method by using the preRYGB sample as baseline, while the IR-B:A ratio was calculated by dividing the IRB 2-∆Ct values by the IR-A 2-∆Ct values. Liver IR-B:A isoform ratio increased after
remission of insulin resistance and type 2 diabetes as shown in Figure 4.3-5A.
128
The mean IR-B:A ratio in the sT2DM group increased significantly (p=0.02, paired ttest) from 5.4 (95% CI, 4.0 to 6.7) to 8.6 (95% CI, 7.1 to 10.1) after RYGB surgery,
returning to levels seen in the sNGT group. No significant change was seen in the
mean IR-B:A ratio after RYGB surgery in the sNGT group.
The change in IR-B:A ratio in individuals who experienced remission of type 2
diabetes was due to a statistically significant (p=0.01, paired t-test) decrease in IR-A
mRNA expression following surgery/remission, rather than an increase in IR-B
mRNA expression (Figure 4.3-5B and C). IR-A mRNA expression decreased -1.5
fold (95% CI, -2.2 to -1.1) in the sT2DM group after RYGB surgery, whereas this was
not seen in the sNGT group.
Figure 4.3-5. IR isoform relative expression in liver tissue of morbidly obese individuals before and
after RYGB surgery (● sNGT, n = 8; ○ sT2DM, n = 8). (A) Mean + SE IR-B:A mRNA expression
ratio calculated by IR-B 2-∆Ct /IR-A 2-∆Ct. The mean IR-B:A ratio had a statistically significant increase
after RYGB surgery in the sT2DM group (paired t-test). (B) Mean IR-B mRNA expression and (C)
mean IR-A mRNA expression + SE presented as 2-∆Ct normalised to 18S. Mean IR-A mRNA
expression had a statistically significant decrease after RYGB surgery in the sT2DM group (paired ttest).
129
Consequently, the reduced IR-A mRNA expression only occurred in individuals who
had remission of type 2 diabetes (general liner model, p=0.04) with expression
retuning to levels seen in individuals with normal glucose tolerance after RYGB
surgery. There was no statistically significant change in IR-B mRNA expression after
RYGB surgery in either group. Because IR can form hybrid heterodimers with IGF1R, the expression level of IGF-1R was also assayed. There was no statistically
significant changes in IGF-1R expression levels after RYGB in either group with the
sNGT group having a mean 1.3 fold change (95% CI, -1.1 to 1.7) and sT2DM group
having a mean 1.0 fold change (95% CI, -1.4 to 1.4).
4.3.5 IR abundance and relative isoform expression in Hep G2 cells treated with
supraphysiological levels of insulin
The purpose of this experiment was to investigate how chronic exposure of liver cells
to high concentrations of insulin affects insulin receptor isoform expression. Hep G2
cells were treated with 100nM insulin over 6 days. Cells not treated with insulin were
grown simultaneously and used as controls. IR protein abundance and isoform
expression is shown in Figure 4.3-6.
Figure 4.3-6. Mean + SE Insulin Receptor protein abundance (A) and isoform expression presented
as 2-∆Ct normalised to 18S (B) in Hep G2 cells treated with insulin (n=4).
130
Cells treated with 100nM insulin had a significantly lower abundance of IR in
comparison to control cells (Figure 4.3-6A) (p=0.04). However, there was no
difference in insulin IR-A or IR-B mRNA expression after insulin treatment (Figure
4.3-6B). Consequently, insulin treatment did not change the IR-B:A ratio. Notably,
unlike normal human liver cells, Hep G2 cells have 2 fold higher expression of IR-A
mRNA.
4.3.6 Functional studies of IR isoform specific insulin signalling
Because of the change in IR-B:A isoform expression after remission of diabetes, we
wanted to evaluate the functional consequences of signalling through the IR-A
isoform in a human liver cell line. Investigation of isoform specific insulin signalling
was first attempted with siRNA-mediated knockdown of IR-A or IR-B expression.
Hep G2 cells were either transfected with siRNA that targeted the exon 10/12
boundary (IR-A) or siRNA that targeted exon 11 (IR-B). None of the custom design
siRNA targeting specific IR isoforms caused efficient or specific knockdown.
Although IR-A siRNA s244179 caused complete knockdown of insulin receptor
protein (Figure 4.3-7A), it was not specific to the IR-A isoform as it also decreased
IR-B expression (Figure 4.3-7B). Neither of the IR-B siRNA (s239120 and s239121)
caused significant knockdown of insulin receptor protein (Figure 4.3-7A).
Figure 4.3-7. siRNA-mediated knockdown of insulin receptor A and B isoform. (A) Insulin receptor
relative abundance and (B) IR-B and IR-A mRNA % knockdown in Hep G2 cells transfected with
siRNA (n=3).
131
Although IR-B siRNA s239121 did not downregulate IR-B mRNA expression, IR-B
siRNA s239120 did achieve inefficient non-specific knockdown of both isoforms
(Figure 4.3-7B). siRNA targeting a homologues region of the IR isoforms (s7477 and
s7479) were added for comparison.
4.3.7 Overexpression of IR-B and IR-A in Hep G2 cells
Because siRNA-mediated knockdown of IR-A or IR-B isoform was not feasible,
mammalian vectors were used to over express the IR isoforms. The pcDNA 3.1
plasmid was used to over express IR-B isoform. Initially, cloning of IR-A transcript
into the pcDNA 3.1 plasmid was going to be used for overexpressing IR-A. However
there were some technical difficulties.
Amplification of non complex templates such as plasmid DNA was only possible
after extensive optimization and use of both the Touchdown technique and DMSO as
an additive. Figure 4.3-8A shows a DNA agarose gel of Touchdown-PCR product
using IR specific primers and linear pcDNA3.1 IR-B plasmid as template. There is a
species migrating at ~4100bp representing the IR and confirming the presence of IRB transcript.
Figure 4.3-8. (A) Insulin receptor PCR using linear pcDNA3.1 IR-B showing a prominent species at
~4100bp and a non specific product at ~150bp. (B) HindIII and XbaI digestion of pcDNA3.1 IR-B. (C)
DMSO and MgCl2 gradient of insulin receptor PCR using Taq man enzymes and Hep G2 cDNA. (D)
Successful insulin receptor PCR using Phusion enzyme and Hep G2 cDNA.
132
Furthermore, cutting the plasmid with HindIII and XbaI released the insert which
migrated at the size (~4100bp) expected of the IR (Figure 4.3-8B). However, there is
also a smaller species at approximately 150bp that was amplified during IR PCR. This
was most likely non-specific priming due to the high GC content of the forward
primer (first 50bp of the IR sequence has >80% GC content) and a sequence of five G
nucleotides present in 3’-end of the IR transcript. This mis-priming produced a 150
base pair size fragment, which was visible in all PCR reactions using the IR specific
primers.
Because of the complex and GC rich nature of the transcript, amplifying the IR
transcript with cDNA transcribed from Hep G2 RNA using traditional Taq enzyme
was not possible. Varying reaction components such as MgCl2 and DMSO
concentration did not aid in amplifying the IR transcript from complex templates such
as cDNA (Figure 4.3-8C). Amplification of IR from cDNA was only possible using
proof-reading enzymes such as the Phusion DNA polymerase (Figure 4.3-8D).
However, non specific priming was still interfering with the PCR reaction leading to a
decreased amount of IR transcript yield. Digestion and ligation reactions were
attempted but because of the relatively low yield of insulin receptor PCR product
further optimization was necessary.
4.3.8 Confirmation of isoform specific over expression
Due to time and budgetary constraints relating to laboratory access, a transfection
ready plasmid carrying the IR-A cDNA was purchased from OriGene. This plasmid
was used to overexpress IR-A in Hep G2 cells.
Figure 4.3-9 is a representative gel of IR PCR fragments amplified from cDNA
reverse transcribed from Hep G2 cells transfected with the pCMV6-XL4 IR-A (Hep
G2 IR-A) or pcDNA3.1 IR-B vector (Hep G2 IR-B). To confirm overexpression, PCR
was only allowed to proceed for 20 cycles. As expected no species are visible in the
Hep G2 Empty Vector (control) lane. However, in the Hep G2 IR-A lane, a fast
migrating species at ~60bp is present and is representative of the IR-A fragment as it
lacks exon 11. The slower migrating species visible in the Hep G2 IR-B lane at ~96bp
is representative of IR-B as it contains exon 11 (36bp).
133
Figure 4.3-9. Insulin receptor exon 10-12 fragment PCR in Hep G2 cells overexpressing IR-A or IR-B
isoform. Eukaryotic 18S was used as a loading control. PCR was run for 20 cycles only.
4.3.9 AKT phosphorylation in Hep G2 cells overexpressing IR-A or IR-B
Insulin-induced metabolic signalling was studied in Hep G2 cells transfected with
either IR-A (Hep G2 IR-A) or IR-B (Hep G2 IR-B) cDNA containing plasmid. Cells
transfected with empty vector were used as control (Hep G2 EV).
Figure 4.3-10A is a representative western blot of nine that were used to quantify
activation of AKT and confirm receptor overexpression. All cells were starved
overnight prior to insulin stimulation; empty vector was used as control (Hep G2 EV).
Insulin-induced metabolic signalling was investigated by stimulating cells with
100nM insulin for 5 minutes and assaying levels of Akt-Ser473 phosphorylation. Total
AKT abundance was used to normalise p-AKT abundance. Overexpression of insulin
receptor is evident in Hep G2 IR-A and Hep G2 IR-B cells. Actin was used as a
loading control.
134
As shown in Figure 4.3-10A, overexpressing IR-B, but not IR-A, significantly
increased AKT phosphorylation compared to cells transfected with empty vector
(p=0.01, ANOVA).
Figure 4.3-10. AKT phosphorylation in cells overexpressing IR isoform A
or B. (A) Mean + SE AKT phosphorylation of 9 biological replicates over
three days. (B) Representative western blot of AKT phosphorylation in
HepG2 cells overexpressing IR-A (Hep G2 IR-A) or IR-B (Hep G2 IR-B)
after treatment with insulin. For complete western blot film see Appendix.
135
4.3.10 Insulin-induced PCK1 mRNA expression in Hep G2 cells overexpressing
IR-A or IR-B
Insulin regulates glucose homeostasis partially through inhibiting transcription of
PCK1 mRNA that codes for PEPCK, a key enzyme in the gluconeogenesis pathway.
Consequently, the ability of the individual IR isoforms to regulate PCK1 expression
was investigated. Figure 4.3-11 shows PCK1 mRNA expression after insulin
treatment in cells overexpressing each isoform. All cells were starved overnight prior
to insulin stimulation; empty vector was used as control. Cells were treated with 1µM
insulin for 12 hours after which PCK1 mRNA expression was estimated by RTqPCR.
Figure 4.3-11. Mean + SE PCK1 mRNA expression presented as 2-∆Ct normalised to 18S
in HepG2 cells overexpressing IR-A or IR-B after treatment with insulin (n=6).
Cells transfected with empty vector had a mean -1.7 fold (95% CI, -2.7 to -1.1)
reduction in PCK1 mRNA expression after insulin treatment. Similarly, Hep G2 cells
overexpressing IR-B isoform had a mean -2.6 fold (95% CI, -4.6 to -1.5) reduction in
PCK1 mRNA expression after treatment with insulin. However, Hep G2 cells
overexpressing IR-A tended to have a slight increase in PCK1 mRNA expression (1.1 fold, 95% CI -1.4 to 1.6) which was not statistically significant (p=0.76, students
t-test). Collectively these data suggest that high levels of IR-A expression may have a
detrimental impact on insulin-induced inhibition of gluconeogenesis.
136
4.4 Discussion
The purpose of this study was to investigate whether IR isoform ratios change in
individuals who lose insulin resistance through RYGB surgery. IR protein abundance
and isoform mRNA expression was assayed in liver of individuals with normal
glucose tolerance or type 2 diabetes. Regardless of the status of glucose tolerance, all
participants studied before and after RYGB surgery had an improvement in insulin
sensitivity and a decrease in fasting plasma insulin (Table 2-3). Individuals with type
2 diabetes had a normalisation of fasting plasma glucose and HbA1c levels indicating
a return to normoglycemia.
At the time of RYGB surgery, individuals with type 2 diabetes had lower liver IR
protein abundance in comparison to individuals with normal glucose tolerance (Figure
4.3-1A). Downregulation of IR by insulin has previously been demonstrated in
vitro.153, 154, 156, 157 Considering that individuals with type 2 diabetes, on average, had
significantly higher fasting plasma insulin levels, it is likely that the higher
concentrations of insulin induced degradation of the IR.155,
530
Not surprisingly, IR
abundance negatively correlated with fasting plasma insulin concentration (r = -0.51,
P<0.001). Furthermore, liver IR protein abundance increased after RYGB surgery
(Figure 4.3-3A). This increase was associated with a decrease in fasting plasma
insulin concentration and has previously been documented in skeletal muscle of
individuals who have undergone gastric bypass surgery.412 The molecular
mechanisms that regulate insulin-stimulated degradation of the IR and its consequent
influence on insulin signalling are still relatively unknown. However, evidence
suggests the adaptor protein Grb10 mediates entry of the IR into the proteasomal
degradation pathway.157
One of the key findings in this study was the change in IR-B:A isoform ratio after
RYGB surgery (Figure 4.3-5). In normal human liver, IR-B expression
predominates.366, 424 The ratio of IR-B to IR-A has previously been reported to be 9.8
in normal lean individuals and 6.9 in obese individuals with and without type 2
diabetes.366 However in this study group, obese individuals with type 2 diabetes had a
significantly lower IR-B:A ratio (5.2) than obese individuals without type 2 diabetes
(6.6). Furthermore, the IR-B:A ratio increased from 5.4 to 8.6 only in those
137
individuals who had a remission of type 2 diabetes, returning to levels seen in the
normal glucose tolerant group before RYGB (IR-B:A ratio 8.5).
The change in IR-B:A ratio after RYGB surgery was primarily driven by reduced
expression of IR-A rather than increased expression of IR-B isoform. IR-A mRNA
expression had a 50% decrease in individuals who had normalisation of glucose
homeostasis and a dramatic improvement in insulin sensitivity (Figure 4.3-5), whereas
there was no statistically significant changes in IR-B or IGF-IR expression.
In this study, IR-A expression was associated with severe hyperinsulinemia and
fasting hyperglycaemia. Increased IR-A expression has previously been documented
in skeletal muscle of a markedly insulin-resistant individual with type 2 diabetes.423
Similarly, individuals with myotonic dystrophy type 1, a disease characterized by
severe insulin resistance, also have increased expression of IR-A mRNA isoform.415
The aberrant insulin receptor ratio is not seen in other myopathies suggesting it may
be specific to disorders that are associated with hyperinsulinemia and impaired
regulation of glucose homeostasis.415 Furthermore, IR-A has been established as a
predominantly mitogenic receptor and has been found to be overexpressed in certain
cancers. An increasing awareness of the link between hyperinsulinemia and
carcinogenesis428 provides circumstantial evidence for a relationship between
hyperinsulinemia and increased IR-A expression.
Although it appears insulin has a role in regulating IR isoform splicing, current
evidence in the literature is inconclusive.
423, 531, 532
Treating Hep G2 cells with high
concentrations of insulin downregulated IR abundance, but there was no significant
change in the isoform expression ratio (Figure 4.3-6). Furthermore, because IR-A has
been found to have a role in hepatocyte glucose uptake by forming IR-A/GLUT2
complexes, it is possible hyperglycaemic conditions may induce increased IR-A
expression.383,
384
Even though the splicing molecules that regulate alternative IR
splicing have been established,361, 533 the factors that drive their expression are still
unclear. It appears that alternative IR splicing involves both hormonal and metabolic
factors.
138
Current literature and data presented here suggests there may be a relationship
between conditions characterized with severe hyperinsulinemia and increased IR-A
mRNA expression, which may have consequences for insulin signalling.
4.4.1 Changes in IR isoform expression affect insulin signalling in Hep G2 cells
IR-B and A isoform have previously been identified as regulators of metabolic and
mitogenic signalling respectively.534 Because of this and the observed change in IR
isoform ratio after improvement in insulin sensitivity and remission of diabetes, we
wanted to characterise the functional consequences of signalling through the IR-A.
We used Hep G2 cells as a human liver cell model to investigate what roles individual
IR isoforms have on insulin signalling relating to glucose homeostasis. Although
siRNA were initially used to facilitate this, there were some technical difficulties. Due
to the highly homologous nature of the IR isoforms, the only targetable region of the
mRNA sequence that differed between the two isoforms was the 36 bp exon 11. This
significantly reduced the siRNA sequence permutations and resulted in non-specific
knockdown of individual isoforms (Figure 4.3-7). Consequently, plasmids containing
either the IR-A or IR-B transcript were used to overexpress each isoform in Hep G2
cells after which insulin-mediated metabolic signalling was investigated.
Insulin transmits metabolic signalling through activation of the IR tyrosine kinase. In
turn the activated kinase phosphorylates IRS proteins that interact with PI 3-kinase
resulting in activation of AKT. This is critical for the metabolic effects of insulin
which includes regulation of glycogen synthesis and gluconeogenesis.127 Hep G2 cells
overexpressing IR-B had 50% more AKT phosphorylation than control cells after
insulin treatment. However, Hep G2 cells overexpressing IR-A did not have a similar
level of increase in AKT phosphorylation (Figure 4.3-10A). This further substantiates
the role of IR-B as a metabolic signalling isoform.
Uncontrolled
hepatic
gluconeogenesis
and
impaired
insulin
secretion
22
characteristics thought to be critical for development of hyperglycaemia.
are
Insulin
exerts its inhibitory action on gluconeogenesis by inhibiting transcription of PCK1
mRNA which codes for an enzyme (PEPCK) that catalyzes one of the first steps in
gluconeogenesis.189, 190 Evidence from animal studies suggests that PCK1 expression
139
has a fundamental role in regulating glucose homeostasis and insulin sensitivity.535, 536
Data presented in this study suggests that insulin-induced downregulation of PCK1
transcription, and therefore its regulation of gluconeogenesis, is IR isoform specific.
In Hep G2 cells overexpressing IR-A, insulin did not downregulate PCK1 expression,
while cells overexpressing IR-B had insulin-induced downregulation of PCK1 mRNA
expression. Furthermore, cells overexpressing IR-B tended to have a greater
downregulation of PCK1 mRNA than control cells (Figure 4.3-11). Taken together,
data presented in this study further confirms the role of IR-B as the more
metabolically active isoform and suggest that increased expression of IR-A may have
detrimental outcomes on insulin-mediated metabolic signalling.
This is in agreement with previous studies that have suggested that increased IR-A
expression may contribute to the insulin-resistant state. Skeletal muscle cells from
individuals with myotonic dystrophy type 1 had decreased insulin sensitivity and
increased IR-A expression.415 In the data presented here, there is elevated IR-A
expression in individuals with extreme insulin resistance and type 2 diabetes, which is
normalised after RYGB surgery. Increasing IR-A expression not only skews the IRB:A ratio towards the less metabolically active isoform, but also may increase the
formation of IR-A/IR-B heterodimers which may detract from the metabolic actions
of insulin by decreasing the amount of IR-B homodimers.397 Whether this alteration in
ratio is the cause of insulin resistance or the result of markedly elevated circulating
insulin concentration and its associated metabolic milieu is yet to be determined. Even
so, the relevance of the two insulin receptor isoforms in regulating insulin signalling,
in particular that which leads to metabolic outcomes, is clearly pointed to by this data.
Regardless of the underlying mechanisms governing remission of type 2 diabetes
remission after RYGB surgery, there is an associated increase in liver insulin
sensitivity.448, 476 In this study, this was associated with normalization of the IR-B:A
ratio in the liver of individuals who have had a remission of type 2 diabetes. Data
presented here suggests that IR-B rather than IR-A is central to the regulation of
glucose metabolism at the liver. Skewing the ratio towards IR-A expression may have
a detrimental affect on insulin signalling and presents a potential molecular
mechanism which may contribute to the pathogenesis of type 2 diabetes.
140
CHAPTER FIVE: Changes in global hepatic gene
expression after RYGB Surgery
141
5.1 Introduction
In the previous chapters of this thesis there was a focus on candidate genes involved
in the primary binding and turnover of insulin and how they may contribute to the
pathogenesis of type 2 diabetes. Type 2 diabetes is a polygenic disease involving a
complex interaction between environmental and genetic factors which often presents
as a heterogeneous mix of phenotypes.537 Because of the heterogeneous nature of both
the disease and the human population it is unlikely there will be one critical gene that
will be responsible for the onset of type 2 diabetes. Consequently, this chapter takes a
more systemic approach in investigating the changes to the transcription profile before
and after RYGB surgery in the liver, a key metabolic organ.
Microarray gene expression profiling is an excellent tool that allows biologists to
observe global changes in gene expression that occur before and after a given
stimulus. In conjunction with gene annotation databases, microarray gene analysis
provides an opportunity to identify critical pathways involved in disease progression
which can be the subject of further study.538, 539 This approach has been used to great
affect in both animal,345-348, 540-543 and human544-548 studies. Recently, there has been a
focus on identifying the changes in gene expression during the progression from
insulin resistance to type 2 diabetes. Several animal model studies suggest that genes
in pathways regulating inflammation and hepatic lipid metabolism may have a causal
role in the development of diabetes.346-348, 541, 543
Genes regulating inflammation have been shown to be decreased in studies using
human muscle and liver following improvements in metabolic parameters after
RYGB surgery.544,
545
Furthermore, genes that regulate lipid metabolism are also
differentially expressed after RYGB surgery.545 However, one key distinction these
studies did not address was the possible difference between individuals with and
without type 2 diabetes. As shown in Chapter 2, RYGB surgery can induce durable
remission of type 2 diabetes that is associated with resolution of insulin resistance. In
this chapter we examined the differential global hepatic gene expression before and
after RYGB surgery (and remission of diabetes) and between individuals with and
without type 2 diabetes.
142

RNA extracted from liver tissue biopsied from the sNGT and sT2DM group
was sent to the Research Centre for Genetic Medicine, Children’s Research
Hospital for gene microarray assay on the Illumina platform.

Raw data was received and subjected to microarray expression analysis using
BRB ArrayTools software.

Select genes that were differentially regulated after RYGB surgery or between
individuals with and without type 2 diabetes were further confirmed with
TLDA RT-qPCR.

Analysis of the confirmed genes in the sNGT and sT2DM groups allowed us
to control for variable such as weight loss in order to identify genes most
likely affected by improvements in insulin resistance or remission of type 2
diabetes.
143
The following methods pertain to microarray analysis and confirmatory PCR of liver
gene expression in individuals who had normal glucose tolerance (sNGT group) or
type 2 diabetes (sT2DM groups) before and after RYGB surgery or remission of
diabetes. For detailed description of study cohort refer Chapter 2. Microarray data on
one sample from the sT2DM group was not available.
5.2.1 RNAstable transport to Children’s Research Hospital
Previously extracted RNA was sent to Research Centre for Genetic Medicine,
Children’s Research Hospital, Washington DC. Microarray gene analysis was
conducted by staff at the Core facility using Illumina Bead Microarray analysis. Total
RNA was extracted as described in Chapter 2 Research Design and Methods and was
sent using RNAstable technology (Biomatrica). RNA was preserved using the
principle of anhydrobiosis which enables long term RNA storage. Total RNA (5µg) in
DEPC-treated water was aliquoted into a 96-well RNAstable plate. The plate was left
to dry overnight in a laminar flow hood. After drying, the plate was sealed with an
aluminium foil seal and placed in a moisture barrier foil bag. To recover RNA, 10µL
of DEPC-treated water was added and incubated for 15 minutes at room temperature.
RNA was then ready for downstream applications.
5.2.2 cRNA synthesis
RNA amplification has become the standard method for preparing RNA for
microarray analysis as it facilitates gene expression analysis with minimal sample
amount.549, 550 Amplification of RNA prior to microarray analysis increases both the
sensitivity and quality of the data. The Illumina® TotalPrep™ -96 RNA Amplification
Kit (Ambion) was used to amplify and synthesize biotinylated cRNA for gene
expression analysis on an Illumina platform. This method is based on the RNA
amplification protocol as described by Van Gelder et al. (1990).551 The reaction
consists of an oligo (dT) primer bearing a T7 promoter using ArrayScript™, a reverse
transcription enzyme specifically design to produce higher yields of first strand cDNA
144
than more traditional enzymes. The cDNA then undergoes second strand synthesis
after which it is purified and used as a template for in vitro transcription with the
MEGAscript ® T7 RNA polymerase and biotin-UTP. This generates thousands of
biotinylated antisense RNA copies of mRNA in a sample.
Aliquots of 150ng of high-quality total RNA from each sample was used for cRNA
synthesis as per the manufacturer instructions. The concentration of each RNA sample
was
determined
by
NanoDrop®
Spectrophotometer
ND-1000
(NanoDrop
Technologies) and the quality of RNA samples was assessed with Agilent 2100
Bioanalyzer (Agilent Technologies Inc). Briefly, RNA was converted to cDNA with
Reverse Transcription Master Mix for 2 hours at 42ºC. Following this Second Strand
Master Mix was added to each sample and incubated at 16ºC for 2 hours. The double
stranded cDNA was purified via a cDNA Filter Cartridge and the eluted cDNA was
used in an in vitro transcription reaction along with IVT Master Mix to synthesize
biotinylated cRNA. The cRNA was then purified via a cRNA Filter Cartridge and the
resulting product was used for microarray analysis.
5.2.3 Illumina microarray
RNA expression profiling was performed using Illumina® Gene Expression BeadChip
Array technology and the Direct Hybridization Assay system (Illumina Inc., San
Diego, CA).
Illumina microarray technology is based on arrays of randomly assembled glass silica
beads that are 3 microns in diameter and spaced ~5.7 microns apart.552 Each bead has
~1 million identical probes (50-mer) covalently attached to the surface through an
amine linkage. Different bead types are pooled together to form assay panels while
every panel has an average of 30 identical bead types producing a high level of bead
redundancy. Because the beads are randomly distributed across the substrate surface
they must be decoded to identify the bead type and location.
Decoding is achieved by a highly effective hybridization process (error rate < 1 x 10-4
per bead) from a 29-mer address sequence attached to each bead.553 This process also
validates the hybridization performance of every bead on every BeadChip ensuring
the arrays are of a high quality.
145
In this experiment, assays were conducted using the Illumina Expression BeadChip
format. A BeadChip is a planar silica substrate slide that has uniform wells created by
plasma etching. The arrays are arranged in a multiple sample format which reduces
sample-to-sample variability. 750ng of biotinylated cRNA was detected by
hybridization to the 50-mer probes on the HumanHT-12_v4_BeadChip (Illumina Inc).
The HumanHT-12 Expression BeadChip has genome-wide transcriptional coverage of
well characterized genes, candidate genes, and splice variants. Each array has more
than 47,000 probes, which were designed to cover content from the NCBI RefSeq
Release 38 (November 7, 2009) as well as legacy UniGene content (Table 5-1).
Table 5-1. Illumina HumanHT-12_v4_BeadChip expression content
Probes
NM
Description
Coding transcript, well established annotation (28,688)
XM
Coding transcript, provisional annotation (11,121)
NR
Non-Coding Transcript, well established annotation (1,752)
XR
Non-Coding Transcript, provisional annotation (2,209)
Source
RefSeq source release (Human RefSeq Release 38)
UniGene Experimentally confirmed mRNA sequences that align to ETS clusters
cRNA hybridization and microarray analysis was done according to the WholeGenome Gene Expression Direct Hybridization protocol (Illumina Inc). After
normalizing, the cRNA was dispensed to the BeadChips which were placed in Hyb
Chamber inserts and then into Hyb Chambers. The chambers were incubated in an
Illumina Hybridization Oven for 16 hours at 58ºC. After incubation the BeadChips
were removed, washed with Wash E1BC buffer and blocked with Block E1 buffer. To
detect the signal the BeadChips were treated with Cy3-streptavidin for 10 minutes.
After a final wash with Wash E1BC buffer the BeadChips were dried via
centrifugation and scanned using the HiScan™SQ Reader. Obtained decoded images
were analyzed by GenomeStudio™ Gene Expression Module (Illumina Inc).
146
5.2.4 GeneomeStudio™ Gene Expression Module data output
Data files were received from the Research Centre for Genetic Medicine, Children’s
Research Hospital in Microsoft Excel format and included; Probe ID, Gene Symbol,
Average Signal, Detection P-value, Search Key, Illumina Gene key and Chromosome
Location and Gene Definition. The Average Signal is the average intensity of the bead
type/target in the group while the Detection P-value is a statistical calculation that
gives the probability the signal from a given bead type is greater than the average
signal from the negative controls. The minimum statistically significant detectable
fold change for any given gene is 1.35 fold while a detection p-value less than 0.05
signifies that there is 95% chance the gene expression was accurately detected.
The Excel files were received as: Raw Data.xls, Background Corrected.xls,
Background Corrected and Normalized.xls and Normalized not background
Corrected.xls. According to Dunning et al. (2008), Illumina local background
subtraction is beneficial for detecting differential gene expression.554 However,
background normalization may introduce substantial variability, particularly at low
intensities and may also increase the amount of false positives. Differential gene
expression analysis was initially carried out using the Background Corrected.xls file;
however, because in this instance negative values post background correction caused
missing values, the Raw Data.xls file was used for gene expression analysis instead.
5.2.5 Analysis of microarray data with BRB ArrayTools
Gene expression microarray analysis is becoming a standard tool in most Molecular
Biology laboratories for hypothesis forming or screening for molecules involved in
disease pathology. Valid analysis of gene expression microarray experiments requires
substantial knowledge of statistical methods and is often done by experts in the
Bioinformatics field. However, recent development of analysis software allows
biomedical scientists to use valid and powerful statistical methods to achieve their
experimental objectives without the need to learn programming language. Biometric
Research Branch (BRB) ArrayTools version 4.3.1 software, developed by Dr. Richard
Simon and the BRB-ArrayTools Development team (BRB, National Cancer Institute,
147
US), was used to filter and execute statistical analysis of the microarray data. BRB
ArrayTools contains analytic and visualization tools that are integrated into excel via
add-ins and use the R version 3.0.1 (R Core Development Core Team, 2013)
environment to facilitate analysis. Although an understanding of the processes behind
microarray gene analysis is still required, BRB-Array Tools simplifies the complex
and computationally intensive processes that generate significant differential gene
expression.
5.2.5.1 Processing, collating and filtering the data
The Raw Data.xls file was converted to a tab delimited file (Raw Data.txt) and
imported into BRB ArrayTools using the General Format Importer. The data was
filtered using the Re-filter, Normalise and Subset the Data function. The Spot Flag
Filter function was used to exclude any Average Probe Signal that did not have a
significant detection p-value (<0.05) thus allowing only accurately detected genes to
be analysed. In addition, only those genes that were present in > 50% of arrays and
had expression that differed ≥ 1.35 fold from the median in at least 20% of the arrays
were used for analysis.
5.2.5.2 Data normalisation
Microarray data was normalised to remove any non-biological variation introduced by
differences in sample preparation and the production or processing of arrays. Data
was normalised using quantile normalisation (Figure 5.2-1).555 This is a complete data
normalisation algorithm which combines data from all arrays to form the
normalisation relation. Quantile normalisation was chosen over other methods for two
reasons. Firstly, as it uses the complete data set to execute normalisation, it is more
representative of the actual experiment. And secondly it has been shown to be
superior when making pairwise comparisons.555
148
Figure 5.2-1. Box and whisker plots of expression data from all 30 arrays before and after Quantile
Normalisation. Boxes show the 25th and 75th percentile of the distribution of log transformed
intensities. The horizontal line in the middle of the box represents the median. The dotted lines
represent the minimum and maximum values. The circles represent any outliers that lie outside 1.5
times the interquartile range.
149
5.2.6 Gene annotation
Genes that passed filtering were annotated using the SOURCE database which was
designed particularly for microarray gene analysis.556 SOURCE combines many
publicly available databases (OMIM, SwisProt, UniGene, GenBank and others) and
unifies information on gene function, gene ontology and gene expression data.556-560
SOURCE contains gene reports for both characterised and uncharacterised genes.
Characterised genes are titled with Human Gene Nomenclature Committee approved
conventions,561 while uncharacterised genes are named using their UniGene titles.
5.2.7 Class comparison
In this study, differentially expressed genes were identified by using the multivariate
permutation test (MPT) in BRB ArrayTools. The MPT is a type of class comparison
statistical test which uses supervised methods that stratify specimens by class.
Differential gene expression after RYGB surgery was identified by comparing the
pre-RYGB (sNGT and sT2DM combined) vs post-RYGB classes (sNGT and sT2DM
combined), pre sNGT vs post sNGT classes and pre sT2DM vs post sT2DM classes.
Differential gene expression between individuals with and without type 2 diabetes
was identified by comparing sNGT vs sT2DM classes (pre and post data combined),
pre sNGT vs pre sT2DM classes, and post sNGT vs post sT2DM classes. All intensity
data was log2 transformed prior to statistical analysis.
5.2.8 Multivariate Permutation Test and the Random Variance Model
Genes that were differentially expressed were identified using a multivariate
permutation test (MPT) with a random variance model (RVM) t-test.562-564 The MPT
is a statistical test designed to control the number or proportion of false discoveries. It
is especially effective for experiments with a small sample size where the assumption
of a Gaussian distribution may be false.563, 565
150
False discoveries are a major issue in statistical analysis of microarray data due to the
multiple comparison problem created by testing for statistical significance between
thousands of different genes. Although several methods for controlling the number of
false discoveries have been developed there are some problems with these
approaches. The Bonferroni adjustment to p-values is a popular, simple, and effective
technique that controls the number of false positives.566, 567 However, it is often too
conservative for application in microarray studies and may lead to false negative
results because of the inherent correlation present in gene expression.563, 566
Benjamini and Hochberg developed a less conservative technique that controls the
proportion of false positives otherwise termed the false discovery rate (FDR).567
Although this technique is viable and still widely used it also operates under the
assumption of independence, which may not be the case in genes that are coregulated. Furthermore, these techniques are based of p-values calculated using
parametric t/F tests that may not be accurate in the extreme tails of the normal
distribution for a study with a small sample size.
The small sample size in this study was partly addressed by use of a random variance
model.564 Standard t/F tests are based on the assumption that the within class variance
is different for each gene. Consequently the variance for each gene is estimated
separately. However, when the sample size is small this estimation is inaccurate and
the statistical power of the t/F test is poor.563, 565 The RVM t-test permits the sharing
of information among probe sets without assuming that all the probe sets have the
same variation. Specifically, the RVM t-test assumes that the genes have different
variances, but it regards these variances as independent samples from the same
distribution. The variances are drawn from a two parameters (α and β) inverse gamma
distribution whose parameters are estimated from the complete set of expression
data.564, 565 In studies where the sample size is small the RVM increases the degrees of
freedom, which improves estimation of variances and statistical power for detecting
differentially expressed genes.564, 565
Using the RVM t-test p-values in a MPT avoids the assumption that the errors have a
Gaussian distribution and as such is more accurate.564 During each permutation in the
MPT, the class labels are randomly reassigned and the p-value is recalculated for each
probe set.562, 565
151
The generated p-value for each gene is a measure of the extent it appears differentially
expressed between the random classes generated by the permutations (a highly
significant p-value represents a false positive).565 The genes are then ordered by their
permutation p-values and a p-value threshold is set. A large number of permutations
are made and for each potential p-value threshold the program records the number of
genes on the list. The distribution of the genes that have a smaller p-value than the
threshold (false positives) can be calculated. The algorithm selects a threshold p-value
so that the proportion of false discoveries is no greater than C% of the time (C denotes
the confidence interval).
In this study, a total of 1000 permutations were completed to identify the list of
probes sets containing < 5% false positives at a confidence of 90%. Differential
expression was considered significant at p ≤ 0.001.
5.2.9 TaqMan® Low Density Array confirmatory RT-qPCR
Confirmatory PCR was carried out on genes that had a fold change of 1.35 fold or
higher in any direction and had highly significant p-value (p =/<0.001) from class
comparison with multivariate permutation test. Genes that were differentially
regulated in the sT2DM group after RYGB surgery or were differentially expressed
between individuals with and without type 2 diabetes were given priority for
confirmation. Twelve TaqMan® Low Density Arrays (TLDA) were purchased from
Applied Biosystems. Each TLDA consisted of 384 wells preloaded with 48 TaqMan®
Gene Expression Assays, 3 of which were reference genes (Figure 5.2-2). Each TLDA
came in a TaqMan® Array Micro Fluidic Card format which was capable of assaying
8 samples simultaneously.
152
Figure 5.2-2. TaqMan® Array Micro Fluidic Card. (A) Schematic representation of a TLDA micro
fluidic card. (B) The list of TaqMan® Gene Expression Assays populating the 48 reaction wells in each
sample channel.
5.2.9.1 TaqMan® Array Micro Fluidic Card loading, sealing and analysis
First strand cDNA synthesis was carried out as described in Chapter 3 Research
Design and Methods. cDNA was divided into 11µL aliquots and stored at -20ºC.
Samples were transported frozen to the Cancer Genetics laboratory, Department of
Biochemistry, University of Otago, Dunedin for analysis on the ABI Prism® 7900HT
Sequence Detection System.
Frozen cDNA was thawed on ice and 10µL was added to a microcentrifuge tube
containing 40µL nuclease free water and 50µL of 2x TaqMan® Universal PCR Master
Mix (Applied Biosystems). The tubes were mixed by gentle vortexing and briefly
spun in a mini-centrifuge. The TaqMan® Array Micro Fluidic Card was allowed to
reach room temperature (~15min incubation) after which 100µL of the sample-PCR
master mix was loaded into the TLDA card reservoirs. The TLDA cards were
centrifuged twice in a Heraeus centrifuge using the Heraeus Custom buckets and card
holders at 331g for 1 minute at an up/down ramp rate of 9. After centrifugation the
TaqMan® probes and primers within the reaction wells were re-suspended.
153
The wells were sealed using a TaqMan® Array Micro Fluidic Card Sealer which
isolates the wells after the sample-PCR master mix is loaded. The TLDA cards were
run on an ABI Prism ® 7900HT Sequence Detection System (Applied Biosystems).
Sequence Detection Software version 2.1 (Applied Biosystems) and Relative
Quantitation (RQ) Manager Software were used to calculate a threshold cycle (Ct).
Each sample was run in triplicate using 12 TLDA plates.
5.2.9.2 Statistical analysis of TLDA RT-qPCR data
DataAssist™ software was used to compile the Ct data generated by RQ manager.
Technical replicates were averaged and a Grubb’s test was used to filter outliers.568
Relative Expression Software Tool (REST, 2009) developed by Corbett Research Pty
Ltd and Dr Michael W. Pfaffl569 was used to estimate and confirm up and down
regulation of genes that were found to be significant with microarray analysis.
REST 2009 allows for use of multiple reference genes for normalisation and uses
integrated randomization (a similar concept is used in the MPT) to test for statistical
significance. The hypothesis test represents the probability that the difference between
the sample and control groups is due only to chance.
It randomly permutes class labels up to 10, 000 times and counts the number of times
the relative expression of the randomly assigned groups is greater than the sample
data. In addition, REST 2009 applies bootstrapping techniques that provide 95%
confidence intervals for expression ratios, without normality or symmetrical
distribution assumptions. Using REST 2009, Ct data was normalised using a
normalization factor that was generated by taking the geometric mean of Ct values of
the reference genes (18S, ACTB and PPIA). TLDA RT-qPCR data was presented as
fold change along with a confidence interval and p-value. Gene expression data was
also presented as absolute values relative to the 18S (2-∆Ct).
154
5.2.10 Analysis of LGALS4 protein abundance using E-PAGE electrophoresis
and western blotting
LGALS4 protein abundance was analysed in human liver tissue to confirm the
microarray and RT-qPCR results. Briefly, 15 µg of protein per sample was resolved
by E-PAGE electrophoresis (Life Technologies) and transferred to Hybond-P PVDF
membrane (Amersham). Membranes were blocked with 5% skim milk powder in
TBT-T at 4°C overnight and then probed for 2 h at room temperature with AntiLGALS4 antibody (1:2000, Santa Cruz) Blots were then incubated with alkaline
phosphatise-conjugated anti mouse antibody (Life Technologies) for 0.5 h. The
chemiluminescent signal was developed using CDP-Star chemiluminesent substrate
(Life Technologies) and captured on x-ray film (Kodak). The protein bands on the xray film were digitized by the Chemidoc XRS system (Bio Rad) and then analyzed
using Quantity One software (Bio Rad). Bands were quantified and expressed as
volume quantity; the sum of the intensities of the pixels within a defined volume
boundary × pixel area (intensity units × mm2). Local background subtraction was
used to facilitate background correction. LGALS4 protein relative abundance was
calculated by normalizing the intensity of the LGLAS4 band to the intensity of the
Actin band.
Statistically significant differences in LGALS4 expression were tested with student’s
t-test. Normal distribution of all groups to be compared was tested with D’AgostinoPearson test. Non-normally distributed data was log transformed to comply with
parametric test assumptions. An alpha level of 0.05 was set as the significance
threshold.
155
5.3 Results
5.3.1 Differential gene expression after RYGB surgery
Differential gene expression was analyzed by comparing pre-RYGB gene expression
to post-RYGB gene expression in 15 individuals that had second liver biopsies. Three
different comparisons were made including: gene expression changes after RYGB
surgery in all 15 individuals, gene expression changes in the sNGT group only, and
gene expression changes in the sT2DM group only.
Out of the 575 genes that passed filtering criteria, class comparison analysis revealed
71 genes that were significantly differentially expressed (> 1.35 fold) after RYGB
surgery. When controlled for disease presence, the sNGT group had only one
differentially expressed gene, whereas the sT2DM had 50, 13 of which were only
differentially regulated after remission of type 2 diabetes. Table 5-2 presents 84 genes
that had a statistically significant change in gene expression organized according to
Gene Ontology Biological Process into one of 11 functional categories; glucose
metabolism, lipid metabolism, amino acid metabolism, inflammation and immunity,
blood coagulation, regulation of transcription, cell growth and differentiation,
xenobiotic metabolism, cell structure, transport, and miscellaneous. Only a third of
differentially expressed genes were upregulated while the rest were downregulated
after RYGB surgery.
5.3.1.1 Glucose metabolism
Eight of the differentially expressed genes were involved in either glucose metabolism
or insulin action and most were downregulated. Inhibin, beta E (INHBE), is involved
in insulin secretion and was decreased after RYGB surgery. Notably, INHBE, which
is a secreted protein, was decreased 2.33 fold in individuals who had a remission of
type 2 diabetes. Genes involved in glucose metabolism such as enolase 3 (ENO3),
serine dehydratase (SDS) and serine/threonine-protein kinase SIK1 (SIK1) were all
decreased after RYGB surgery. Conversely, phosphoenolpyruvate carboxykinase
(PCK1) expression was increased after RYGB surgery only in individuals who had a
remission of type 2 diabetes.
156
Table 5-2. Illumina microarray differentially expressed (≥ 1.35 fold) liver genes ~16 months after RYGB surgery (n=15).
Illumina
ID
(ILMN_)
Gene (Symbol)
GO Biological Process
1678904
1811114
1717639
Enolase 3 (beta, muscle) (ENO3)
Serine dehydratase (SDS)
Serine/threonine-protein kinase SIK1
(SIK1)
Inhibin, beta E (INHBE)
Phosphoenolpyruvate carboxykinase
(PCK1)
Glycolysis, skeletal muscle tissue regeneration
Gluconeogenesis, L-serine metabolism
regulation of gluconeogenesis/ triglyceride biosynthetic
process/mitotic cell cycle
Growth, insulin secretion
Response to insulin stimulus, gluconeogenesis
-1.67 (-1.69)*
-1.64 (1.89)#
-1.59 (-1.69)*
1.00E-07
1.15E-04
2.46E-04
-1.89 (2.33)*
1.56
4.70E-06
4.39E-05
Lipid transport, receptor-mediated endocytosis
-1.72
8.21E-04
1784871
2138765
Low density lipoprotein receptor
(LDLR)
Fatty acid synthase (FASN)
Perilipin-2 (PLIN2)
-1.69
-1.52 (-1.69)*
7.37E-05
8.15E-05
2041293
1670134
Squalene monooxygenase (SQLE)
Fatty acid desaturase 1 (FADS1)
-1.43
-1.37 (1.56)*
9.80E-04
4.48E-04
1683598
-1.47
5.46E-04
2234956
Long-chain-fatty-acid--CoA ligase 4
(ACSL4)
Leptin receptor (LEPR)
Fatty acid biosynthesis
Cellular lipid metabolic process, lipid storage, long-chain
fatty acid transport, development of adipose tissue
Cholesterol biosynthetic process,
Unsaturated fatty acid biosynthetic process, response to
insulin stimulus
Lipid biosynthetic process
1.35
9.00E-07
2346987
Apolipoprotein(a) (LPA)
Cholesterol metabolic process, energy reserve metabolic
process, cytokine-mediated signalling pathway
Lipid transport, lipid metabolic process, blood circulation
1.46*
1.53E-05
1652037
1757807
Uridine phosphorylase 2 (UPP2)
Ethanolamine-phosphate phospholyase (AGXT2L1)
Pyrimidine nucleoside catabolic process
Cellular amino acid metabolic process
-1.61
-1.45
9.37E-04
6.33E-05
1811767
1731948
2053415
Fold Change
Post/Pre
P-value
Functional
categories
Glucose
metabolism
Lipid metabolism
Amino acid
metabolism
157
Illumina
ID
(ILMN_)
2056975
Gene (Symbol)
Fold Change
Post/Pre
P-value
Hypoxanthine-guanine
phosphoribosyltransferase (HPRT1)
1.40*
8.00E-07
1657862
Adenosylhomocysteinase (AHCY)
1.45*
6.40E-06
2395451
Argininosuccinate synthase (ASS1)
Guanine salvage, hypoxanthine salvage, purine
ribonucleoside salvage, positive regulation of dopamine
metabolic process
Sulfur amino acid metabolic process, circadian sleep wake
cycle
Arginine biosynthetic process, inirectly involved in control
of blood pressure,
1.37*
4.98E-05
1780575
C-reactive protein (CRP)
-2.56 (-3.33)*
1.19E-05
1774874
Interleukin-1 receptor antagonist
protein (IL1RN)
Interferon alpha-inducible protein 6
(IFI6)
Interleukin-21 (IL32)
Interleukin-1 receptor accessory
protein (IL1RAP)
Complement activation, classical pathway, negative
regulation of lipid storage
Negative regulation of interleukin-1-mediated signalling
pathway, insulin secretion, immune response
Cytokine-mediated signalling pathway
-1.54 (-1.59)*
1.40E-06
-1.47*
2.05E-05
Immune response
Immune response, inflammatory response
-1.35
1.63 (1.91)*
2.64E-04
2.00E-06
Negative regulation of plasminogen activation*
-2.44 (-3.1)*
3.70E-06
Blood coagulation
-1.47*
2.38E-05
1745607
Serpine peptidase inhibitor, clade E,
member 1 (SERPINE1)
Hermansky-Pudlak syndrome 5
protein (HPS5)
Alpha-2-macroglobulin (A2M)
Negative regulation of complement activation/ blood
coagulation
1.42*
3.73E-05
1806023
Jun proto-oncogene (JUN)
Negative regulation of DNA binding and transcription, liver
development, aging, circadian rhythm, immune response,
Regulation of transcription, DNA dependant, multicellular
organismal development
-2.38
6.34E-05
1687384
1778010
2357062
1744381
2411731
1751607
FBJ murine osteosarcoma viral
oncogene homolog B (FOSB)
-2.04
3.89E-04
Functional
categories
Inflammation and
immunity
Blood
coagulation
Regulation of
transcription
158
Illumina
ID
(ILMN_)
2374865
Gene (Symbol)
Activating transcription factor 3
(ATF3)
Basic helix-loop-helix family,
member e40 (BHLHB2)
Regulation of transcription, DNA dependant, positive
regulation of cell proliferation, gluconeogenesis
Negative regulation of sequence-specific DNA binding
transcription factor activity, circadian regulation of gene
expression
-1.89
4.62E-04
-1.72
4.18E-05
1735014
Kruppel-like factor 6 (KLF6)
-1.67 (-1.75)*
1.13E-04
1810214
Transcription factor jun-D (JUND)
-1.67
5.00E-05
1703123
Cysteine/serine-rich nuclear protein
1 (AXUD1/CSRNP1)
Tribbles homolog 1 (TRIB1)
Positive regulation of transcription, DNA dependant,
cytokine-mediated signalling pathway
Regulation of transcription from RNA polymerase II
promoter, aging, circadian rhythm, response to light
stimulus
Positive regulation of transcription from RNA polymerase
II promoter, apoptotic process
Negative regulation of sequence-specific DNA binding
transcription factor activity, regulation of MAPK activity,
JNK cascade
II promoter, dopamine biosynthetic process
Regulation of transcription from RNA polymerase II
promoter, negative regulation of apoptotic process
Calcium-mediated signalling
Transcription from RNA polymerase II promoter, rhythmic
process
II promoter
II, circadian regulation of gene expression
Negative regulation of transcription from RNA polymerase
II promoter, negative regulation of apoptotic process
-1.64 (-1.75)*
2.80E-06
-1.56
1.61E-04
-1.54 (-1.69)*
5.70E-05
-1.47 (-1.69)*
3.50E-06
-1.47 (-1.49)*
-1.47
2.00E-07
2.67E-05
-1.41*
1.10E-06
1.35
3.05E-03
1.56
7.14E-05
1735014
1803811
1782305
2367239
1689378
Nuclear receptor subfamily 4 group
A member 2 (NR4A2)
Cyclic AMP-dependent transcription
factor ATF-5 (ATF5)
Calcipressin-1 (RCAN1)
Nocturnin (CCRN4L)
1751206
Protein AF-10 (MLLT10)
2405305
Aryl hydrocarbon receptor nuclear
translocator-like protein 1 (ARNTL)
Zinc finger protein SNAI2 (SNAI2)
1669113
2082585
Fold Change
Post/Pre
P-value
Functional
categories
159
Illumina
ID
(ILMN_)
1661599
1781285
1714108
1761084
1718977
1784602
1798360
1788895
1708496
1750324
1735367
1659316
Gene (Symbol)
DNA-damage-inducible transcript 4
(DDIT4)
Dual specificity phosphatase 1
(DUSP1)
Tumor protein p53 inducible nuclear
protein 1 (TP53INP1)
Negative regulation of glycolysis, response to hypoxia, cell
proliferation
Positive regulation of apoptotic process, response to
oxidative stress, response to light stimulus
Apoptosis, negative regulation of cell proliferation, positive
regulation of autophagy,
-1.96
4.50E-06
-1.72
1.24E-04
-1.67 (-1.72) *
6.00E-07
Fibronectin type III domain
containing 5 (FNDC5)
Growth arrest and DNA damageinducible protein GADD45 beta
(GADD45B)
Cyclin-dependent kinase inhibitor 1
(CDKN1A)
Positive regulation of brown fat cell differentiation
-1.67 (-1.72)*
< 1e-07
Cell differentiation, apoptotic process, activation of
MAPKKK activity
-1.43
4.70E-05
DNA damage response, signal transduction by p53 class
mediator resulting cell cycle arrest, negative regulation of
cell growth and proliferation, stress-induced premature
senescence
Negative regulation of apoptotic process
-1.41 (-1.42)*
2.23E-05
-1.43*
2.30E-05
Cell differentiation, steroid metabolic process, androgen
biosynthetic process
Intrinsic apoptotic signalling pathway in response to
endoplasmic reticulum stress
Regulation of cell growth, glucose homoeostasis, negative
regulation of insulin-like growth factor receptor signalling
pathway,
Androgen catabolic process, steroid biosynthetic process
-1.39
1.34E-05
-1.35*
2.00E-07
1.35
1.13E-04
1.36 (1.51)*
3.00E-07
Cell cycle arrest, regulation of growth
1.47*
8.65E-05
C-X-C chemokine receptor type 7
(CXCR7)
3-oxo-5-alpha-steroid 4dehydrogenase 2 (SRD5A2)
BRSK2 (BRSK2)
Insulin-like growth factor-binding
protein 5 (IGFBP5)
Estradiol 17-beta-dehydrogenase 11
(HSD17B11)
Hepatocyte cell adhesion molecule
(HEPACAM)
Fold Change
Post/Pre
P-value
Functional
categories
Cell growth and
differentiation
160
Illumina
ID
(ILMN_)
1694240
Gene (Symbol)
Fold Change
Post/Pre
P-value
Dual specificity mitogen-activated
protein kinase kinase 1 (MAP2K1)
1.48*
8.14E-05
2326197
1725193
Fibrinogen-like protein 1 (FGL1)
Insulin-like growth factor-binding
protein 2 (IGFBP2)
Negative regulation of cell proliferation, cell cycle arrest,
positive regulation of cell differentiation, insulin receptor
signalling pathway
Hepatocyte mitogenic activity
Regulation of cell growth, aging, positive regulation of
activated T-cell proliferation
1.6
2.33 (2.82)
6.60E-04
1.00E-07
2124802
Metallothionein 1H (MT1H)
-2.04
2.21E-05
1718766
Metallothionein 1F (MT1F)
-1.85
2.36E-05
1657435
1686664
Metallothionein 1M (MT1M)
Metallothionein-2 (MT2A)
-1.72
-1.45
9.16E-05
3.37E-05
1715401
Metallothionein-1G (MT1G)
-1.41
2.88E-04
1691156
Metallothionein-1A (MT1A)
-1.37
3.43E-04
2136089
1683607
Metallothionein (MTE)
Cytochrome P450 1A2 (CYP1A2)
Cellular response to cadmium ion and zinc/ negative
regulation of growth
Cellular response to cadmium ion and zinc/ negative
regulation of growth
Cellular response to zinc ion/negative regulation of growth
Cellular response to drug/cellular response to zinc
ion/negative regulation of growth
Cellular response to cadmium/zinc and copper ion/negative
regulation of cell growth
Cellular response to cadmium and zinc ion/negative
regulation of cell growth
Metal ion binding
Xenobiotic metabolism, steroid catabolic process
-1.41
1.48
1.67E-04
1.08E-02
1667796
3305938
Hemoglobin, alpha 2 (HBA2)
Serine/threonine-protein kinase Sgk1
(SGK1)
Solute carrier family 2, facilitated
glucose transporter member 3
(SLC2A3)
Oxygen transporter activity
Positive regulation of transporter activity, renal sodium ion
absorption
L-ascorbic acid metabolic process, glucose transmembrane
transporter activity
-1.75 (-2.17)*
-1.72
6.42E-04
2.60E-06
-1.56
2.74E-04
1775708
Functional
categories
Xenobiotic
metabolism
Transport
161
Illumina
ID
(ILMN_)
1791728
2380938
1802205
1801377
1661335
1711015
1754247
1789096
1768278
1706579
1690223
1800787
1759513
1776188
1808041
Gene (Symbol)
Fold Change
Post/Pre
P-value
Calcium-binding mitochondrial
carrier protein SCaMC-2
(SLC25A25)
Synaptotagmin-7 (SYT7)
Rho-related GTP-binding protein
RhoB (RHOB)
Equilibrative nucleoside transporter
4 (SLC29A4)
Spectrin beta chain, non-erythrocytic
1 (SPTBN1)
Alpha-crystallin A chain (CRYAA)
Neutral and basic amino acid
transport protein rBAT (SLC3A1)
OSTalpha protein (OSTalpha)
Transmembrane transport
-1.54
9.17E-05
Transport
Endosome to lysosome transport, positive regulation of
apoptotic process
Transmembrane transport
-1.49
-1.45
6.19E-05
7.59E-04
-1.45*
8.00E-07
Golgi to plasma membrane protein transport, membrane
assembly
Negative regulation of intracellular transport
Cellular amino acid metabolic process
-1.41
2.91E-04
-1.35 (-1.53)*
1.39
6.92E-05
8.32E-04
1.50
1.90E-06
Solute carrier family 22 member 10
(SLC22A10)
Sex hormone-binding globulin
(SHBG)
Transmembrane transporter activity
1.55 (1.88)*
4.30E-06
Hormone transport
1.55 (1.72)*
3.54E-05
Putative uncharacterized protein
CNTNAP2 (CNTNAP2)
RFTN1 protein (RFTN1)
Rho-related GTP-binding protein
RhoE (RND3)
Microtubule-associated proteins
1A/1B light chain 3A (MAP1LC3A)
60S ribosomal protein L10
(RPL10A)
Cell adhesion
-1.41
1.40E-06
Cell adhesion
-1.41
-1.35
2.00E-07
7.49E-05
Organelle membrane, autophagic vacuole
-1.37*
1.00E-06
Anatomical structure morphogenesis
1.14
2.77E-02
Functional
categories
Cell structure
162
Illumina
ID
(ILMN_)
1693334
1771120
2088124
2316955
2262288
Gene (Symbol)
Prolyl 4-hydroxylase subunit alpha-1
(P4HA1)
Transmembrane protein 45B
(TMEM45B)
Transmembrane protein 154
(TMEM154)
Serpin A11 (SERPINA11)
Collagen fibril organization
Serine-type endopeptidase inhibitor activity, regulation of
proteolysis
Protein biosynthesis
Fold Change
Post/Pre
P-value
1.48
1.85E-04
-1.52
4.30E-06
-1.52
3.61E-05
1.37
2.30E-06
Functional
categories
Miscellaneous
Eukaryotic translation elongation
1.39
7.39E-05
factor 1 gamma (EEF1G)
2148527
H19, imprinted maternally expressed
1.66
3.30E-06
transcript (non-protein coding) (H19)
*Genes that were differentially expressed when analysis was restricted to those individuals who had a remission of type 2 diabetes (Pre sT2DM versus Post
sT2DM, n=7). # Genes that were differentially expressed when analysis was restricted to those individuals who had normal glucose tolerance after RYGB
surgery (Pre sNGT versus post sNGT, n=8).
163
5.3.1.2 Lipid metabolism and amino acid metabolism
Eight genes involved in lipid metabolism were differentially regulated after RYGB
surgery. Genes involved in the lipid biosynthetic process such as fatty acid synthase
(FASN), fatty acid desaturase 1 (FADS1), squalene monooxygenase (SQLE) and longchain-fatty-acid--CoA ligase 4 (ACSL4) were all downregulated. Genes involved in
lipid transport such as LDL receptor (LDLR) and perilipin-2 (PLIN2) were also
downregulated.
Conversely,
expression
of
leptin
receptor
(LEPR)
and
apolipoprotein(a) (LPA) was increased after surgery.
Five genes involved in amino acid metabolism were differentially regulated after
RYGB surgery. Adenosylhomocysteinase (AHCY), a gene involved in the circadian
rhythm, and argininosuccinate synthase (ASS1), a gene indirectly involved in control
of blood pressure, were both increased only in the sT2DM group. Expression of
hypoxanthine-guanine phosphoribosyltransferase (HPRT1), a gene involved in the
dopamine metabolic process was also increased after RYGB, whereas the remainder
of the genes involved in amino acid metabolism were downregulated.
5.3.1.3 Inflammation, immunity and blood circulation
Five genes involved in inflammation and immunity were differentially regulated after
RYGB surgery. Expression of C- reactive protein (CRP) and interleukin-21 (IL32),
which are both involved in the immune response, were decreased after RYGB.
Notably, CRP expression had a 3.33 fold reduction in the sT2DM group after surgery.
There were also significant changes in the IL-1 signalling pathway as expression of
interleukin-1 receptor antagonist protein (IL1RN) decreased after RYGB surgery,
while expression of interleukin-1 receptor accessory protein (IL1RAP) was increased
and was 1.91 fold higher in the sT2DM group after remission of diabetes. Three genes
involved in blood coagulation were differentially expressed after RYGB surgery.
Expression of serpine peptidase inhibitor, clade E, member 1 (SERPINE1) was
decreased after surgery and had a 3.1 fold reduced expression in the sT2DM group.
Expression of Hermansky-Pudlak syndrome 5 protein (HPS5) was only decreased in
individuals who had a remission of type 2 diabetes while expression of alpha-2macroglobulin (A2M) was increased.
164
5.3.1.4 Transcriptional regulation
14 genes involved in regulation of transcription were differentially regulated after
RYGB surgery. Jun proto-oncogene (JUN) and jun-D (JUND), both of which were
downregulated after RYGB surgery, regulate transcription of genes involved in a wide
range of process including, the immunity response, aging, circadian rhythm and organ
development. A number of transcription factors that were differentially regulated
were involved in cell growth. Activating transcription factor 3 (ATF3), which is
involved in positive regulation of cell proliferation and gluconeogenesis was
decreased after RYGB surgery. Similarly, FBJ murine osteosarcoma viral oncogene
homolog B (FOSB) and tribbles homolog 1 (TRIB1), which are involved in
multicellular organismal development and regulation of MAPK activity respectively,
were also decreased after RYGB surgery. Genes involved in the apoptotic process,
such as cysteine/serine-rich nuclear protein 1 (CSRNP1) and cyclic AMP-dependent
transcription factor ATF-5 (ATF5) were downregulated. Conversely, zinc finger
protein SNAI2 (SNAI2), a gene involved in negative regulation of apoptotic process is
upregulated in the sT2DM group only.
Multiple genes that are involved in circadian regulation of transcription were
differentially regulated after RYGB surgery. Genes such as basic helix-loop-helix
family member e40 (BHLHB2) and nocturnin (CCRN4L) were both downregulated
after RYGB surgery. Conversely, aryl hydrocarbon receptor nuclear translocator-like
protein 1 (ARNTL) was increased after RYGB surgery.
5.3.1.5 Cell growth and differentiation
13 genes involved in regulating cell growth and differentiation were differentially
regulated after RYGB surgery. Several of these genes that are involved in negative
regulation of cell growth and proliferation as a response to stress were downregulated
after RYGB surgery. These included; dual specificity phosphatase 1 (DUSP1), tumour
protein p53 inducible nuclear protein 1 (TP53INP1), DNA damage-inducible protein
GADD45 beta (GADD45B) and cyclin-dependent kinase inhibitor 1 (CDKN1A).
165
Expression of fibronectin type III domain containing 5 (FNDC5), a gene that
regulates differentiation of white fat cells to brown fat cells was also decreased after
RYGB surgery. Conversely, expression of dual specificity mitogen-activated protein
kinase kinase 1 (MAP2K1), a component of the insulin receptor signalling pathway
involved in cell proliferation and differentiation, was increased only in the sT2DM
group. Expression of insulin-like growth factor-binding protein 5 (IGFBP5) and
insulin-like growth factor-binding protein 2 (IGFBP2) was also increased after RYGB
surgery. Both IGFB5 and IGFBP2 are involved in cell growth and regulation of
glucose homeostasis through the IGF signalling pathway. Notably, IGFBP2
expression increased 2.82 fold in individuals who had a remission of type 2 diabetes
and was the most highly upregulated gene.
5.3.1.6 Xenobiotic metabolism
Eight genes involved in xenobiotic metabolism were differentially regulated after
RYGB surgery. Expression of seven of these genes, which are all members of the
Metallothionein family, was decreased after RYGB surgery. On the other hand,
expression of Cytochrome P450 1A2 (CYP1A2), a gene involved in xenobiotic
metabolism and the steroid catabolic process was increased after RYGB surgery.
5.3.1.7 Cellular transport and structure
10 genes involved in regulating cellular transport were differentially regulated after
RYGB surgery. Solute carrier family 2 facilitated glucose transporter member 3
(SLC2A3), which is involved in glucose transport, was decreased after RYGB surgery.
Similarly, genes involved in intracellular trafficking such as rho-related GTP-binding
protein RhoB (RHOB), spectrin beta chain, non-erythrocytic 1 (SPTBN1) and alphacrystallin A chain (CRYAA) were also downregulated. Conversely, neutral and basic
amino acid transport protein rBAT (SLC3A1), a gene involved in regulating amino
acid metabolism was increased after RYGB surgery.
Six genes involved in cell structure were differently regulated after RYGB surgery.
Genes involved in cell adhesion such as putative uncharacterized protein CNTNAP2
(CNTNAP2) and rho-related GTP-binding protein RhoE (RND3) were downregulated
after RYGB surgery.
166
Microtubule-associated proteins 1A/1B light chain 3A (MAP1LC3A), a gene involved
in autophagic vacuole formation was also downregulated, whereas prolyl 4hydroxylase subunit alpha-1 (P4HA1), a gene involved in collagen fibril formation,
was upregulated. Some genes that were differentially regulated had unknown or
miscellaneous functions. Expression of Transmembrane protein 45B (TMEM45B) and
Transmembrane protein 154 (TMEM154) was decreased after RYGB surgery.
Conversely, H19 and Serpin A11 (SERPINA11) both had an increase in expression
after RYGB surgery.
5.3.2 Differential gene expression between individuals with normal glucose
tolerance or type 2 diabetes
Differential gene expression was analyzed by comparing gene expression in
individuals who had normal glucose tolerance to gene expression in individuals who
had type 2 diabetes. Three different comparisons were made including, differential
gene expression between the sNGT and sT2DM groups (pre and post data combined),
differential gene expression between the sNGT and sT2DM group at the time of
RYGB surgery (Pre RYGB state), and gene expression between the sNGT and
sT2DM group at the time of subsequent surgery (Post RYGB state).
Out of the 575 genes that passed filtering criteria, class comparison analysis revealed
10 genes that were significantly differentially expressed between individuals with and
without type diabetes. When controlled for operation time point, two genes were
differentially expressed between individuals with and without type 2 diabetes at the
initial liver biopsy (Pre-RYGB state) and four genes were differentially expressed at
the second liver biopsy (Post-RYGB). Table 5-3 presents the differentially expressed
genes organized according to Gene Ontology Biological Process
Expression of genes involved in the immune response such as immunoglobulin
lambda-like polypeptide 1 (IGLL1), serum amyloid A-1 protein (SAA1), and
immunoglobulin J chain (IGJ) was elevated in individuals who had type 2 diabetes
regardless of whether they had RYGB surgery. Similarly, expression of galectin-4
(LGALS4), a gene involved in cell adhesion and carbohydrate binding, and pyridine
nucleotide-disulfide oxidoreductase domain-containing protein 2 (PYROXD2), was
elevated in individuals who had type 2 diabetes before and after RYGB surgery.
167
Table 5-3. Illumina microarray differentially expressed genes between the sNGT (n=8) and sT2DM (n=7) groups.
IlluminaID Gene (Symbol)
(ILMN_)
2393765
Immunoglobulin lambda-like polypeptide 1
(IGLL1)
2304512
Serum amyloid A-1 protein (SAA1)
1694034
Galectin-4 (LGALS4)
1684497
Pyridine nucleotide-disulfide oxidoreductase
domain-containing protein 2
(C10orf33/PYROXD2)
Immunoglobulin J chain (IGJ)
2105441
2188966
immune response
Negative regulation of inflammatory response,
platelet activation, positive regulation of
interleukin-1 secretion
Cell adhesion, carbohydrate binding
Fold Change
p-value
sT2DM/sNGT
1.92
5.08E05
1.89
2.23E04
1.82 (2)*
Oxidoreductase activity
1.66*#
Immune response
1.41
-1.48
3308138
Insulin-like growth factor-binding protein complex Signal transduction
acid labile subunit (IGFALS)
Dopamine beta-hydroxylase (DBH)
Synaptic transmission, positive regulation of
vasoconstriction
RNU4-2
spliceosome
3309453
RNU4-1
spliceosome
-1.5 (-1.74)#
3236653
RNU1-5
spliceosome
-1.54#
1729191
Cytochrome P450 2C19 (CYP2C19)
Xenobiotic metabolic process, steroid metabolic
process
-1.74
1746220
-1.4
-1.49 (1.65)#
1.00E07
< 1e07
3.53E03
9.36E04
3.69E03
5.50E06
1.93E05
2.11E05
1.13E04
*Genes that were differentially expressed when analysis was restricted to the Pre-RYGB state (Pre sNGT versus Pre sT2DM). # Genes that were
differentially expressed when analysis was restricted to the Post-RYGB state (Post sNGT versus Post sT2DM).
168
Conversely, expression of insulin-like growth factor-binding protein complex acid
labile subunit (IGFALS), a gene involved in signal transduction, and dopamine betahydroxylase (DBH), a gene involved in synaptic transmission and vasoconstriction
was reduced in individuals who had type 2 diabetes compared to individuals who had
normal glucose tolerance. Furthermore, the expression of two genes (RNU4-2, RNU41) coding for small nuclear RNA’s that are involved in the splicesome was lower in
individuals who had type 2 diabetes. Expression of RNU1-5 was lower in the sT2DM
group only at the second operation.
5.4 Validation of microarray profiles with TLDA RT-qPCR
Of the 95 genes that were found to be significantly differentially regulated by
microarray analysis, 45 were chosen for confirmation by Taq Man Low Density Array
(TLDA) RT-qPCR. 36 of these genes were differentially regulated after RYGB
surgery, whereas nine were differentially regulated between individuals with or
without type 2 diabetes. To test the stringency of the microarray analysis three of the
45 genes (RPL10A, CYP1A2, and DBH) that had a MPT p-value between 0.001 and
0.05 in addition to one control gene (HAMP) that had a borderline significant p-value
of 0.07 were chosen for further validation with RT-qPCR.
Of the 45 Taq Man probes on the TLDA card, two probes (JUND and IGLL1) failed
to produce a detectable signal. The signals from the remaining 43 Taq Man probes
were normalized to 3 reference genes (18S, ACTB and PPIA) prior to data analysis.
Figure 5.4-1 shows the relative average differential gene expression as determined by
microarray or TLDA for 43 genes. For all the confirmed genes there was directional
concordance (increased or decreased) between the expression values generated by
microarray analysis and TLDA analysis. Microarray gene expression and TLDA RTqPCR gene expression was highly correlated (Spearman rank, r = 0.93, p<0.0001). In
almost all cases the TLDA fold change/difference expression values were greater in
magnitude than the fold change determined by Illumina microarray (Figure 5.4-1).
Furthermore, differential expression of HAMP, which was not significant with
microarray analysis (p=0.07) was confirmed to not be significant with TLDA analysis
(p=0.06).
169
Figure 5.4-1. Comparison of Illumina Microarray and TLDA RT-qPCR fold changes of 43 genes that
were found to be differentially expressed. (A) Differentially regulated genes between the sNGT (n=8)
and sT2DM (n=7) groups. (B) Differentially expressed genes after RYGB surgery in both the sNGT
and sT2DM groups (n=15).
170
5.4.1 Differentially expressed genes after RYGB surgery that were confirmed by
TLDA RT-qPCR
TLDA RT-qPCR gene expression analysis was restricted to the sNGT or sT2DM
groups so as to control for the affect of weight loss versus the gene changes induced
by improved insulin sensitivity or remission of diabetes. Genes that were significantly
altered in only the sT2DM group, and tended to have a greater fold change in the
sT2DM than the sNGT group are highlighted in Table 5-4. In this way we can
highlight the genes most likely to be altered by the change in diabetic status as
opposed to weight loss and other factors present in both groups.
Of the two confirmed genes that belonged to the glucose metabolism category, ENO3
had similar changes in gene expression in both the sNGT and sT2DM groups.
Confirmed genes that were involved in lipid metabolism, which included LDLR,
PLIN2, FADS and FASN, all had similar differential expression after RYGB surgery
in both groups. The majority of the confirmed genes involved in regulating
transcription (ATF5, CSRNP1, RCAN1) and genes involved in cell growth and
differentiation (DUSP1, TP53INP1, FNDC) had similar fold changes after RYGB
surgery in both groups. Confirmed genes that are involved in cellular transport and
cell structure (P4HA1, CRYAA, RPL10A and TMEM45B) also had similar changes in
gene expression in both the sNGT and sT2DM group.
In some cases, however, individuals with type 2 diabetes tended to have a greater
magnitude in differential gene expression. Many of the confirmed genes involved in
inflammation and blood coagulation (CRP, SERPINE1, and A2M,) had a greater
magnitude in fold change in individuals with type 2 diabetes after RYGB surgery.
Likewise, fold changes in UPP2 expression was greater in individuals who had a
remission of type 2 diabetes. Although most of the genes involved in transport had
similar differential expression in both groups, expression of SLC2A3 and HBA1 was
only significantly decreased in individuals who had a remission of type 2 diabetes.
Confirmed genes that are involved in xenobiotic metabolism (MT1H and MT1F)
tended to have a greater decrease in gene expression in the sNGT group. The
exception to this was CYP1A2, which was only significantly increased in the sT2DM
group after RYGB surgery.
171
Table 5-4. Confirmed genes that were differentially expressed after RYGB surgery.
Highlighted genes are likely affected by changes in diabetic or insulin-resistant status.
sNGT (Post/Pre)
Gene*
Fold Change
95% CI
INHBE
-2.5
-15.4 to 2.4
ENO3
-2.3
LDLR
sT2DM (Post/Pre)
p-value
Fold Change
95% CI
p-value
0.027
-3.7
-11.1 to -1.3
0.001
-8.1 to 1.5
0.006
-2.0
-13.3 to 1.9
0.038
-2.6
-5.6 to -1.0
0.013
-2.3
-10 to 2.4
0.013
PLIN2
-1.8
-3.9 to 1.6
0.019
-1.8
-2.9 to 1.0
0.022
FADS
-1.5
-6.70 to 2.20
0.152
-1.9
- 12.1 to 5.1
0.108
FASN
-1.7
-8.3 to 3.0
0.112
-2.1
-6.1 to 1.4
0.003
UPP2
-2.5
-12.3 to 1.8
0.118
-3.5
-25 to 1.9
0.009
CRP
-3.2
-23.3 to 1.9
0.03
-5.9
-62.5 to 1.4
0.002
IL1RN
-1.7
-6.8 to 2.2
0.085
-1.9
-5.4 to 1.3
0.010
IL1RAP
1.3
-2.4 to 3.6
0.196
2.0
1.1 to 3.4
0.000
SERPINE1
-6.9
-14.9 to 16.5
0.054
-21.3
-166.7 to -1.8
0.001
A2M
1.6
-2.6 to 6.5
0.060
2.3
1.2 to 3.9
0.002
ATF3
-4.6
-76.9 to 3.8
0.030
-11.4
-58.8 to -1.5
0.001
ATF5
-1.7
-5.4 to 1.6
0.033
-2.1
-3.7 to -1.1
0.000
BHLHE40
-1.5
-5.6 to 2.6
0.175
-3.1
-8.3 to 1.0
0.001
CSRNP1
-2.3
-10.1 to 4.1
0.055
-2.3
-8.1 to 1.5
0.004
CDKN1A
-2.0
-6.0 to 1.3
0.013
-1.9
-6.1 to 1.4
0.008
RCAN1
-2.6
-12.5 to 2.2
0.023
-2.6
-5.6 to -1.1
0.001
DUSP1
-2.3
-8.8 to 1.7
0.20
-2.6
-9.4 to 2.1
0.010
FNDC5
-2.7
-13.2 to 1.3
0.007
-2.5
-20 to 1.4
0.004
TP53INP1
-2.3
-8.1 to 1.7
0.012
-2.2
-4.9 to 1.4
0.002
IGFB2
3.3
-2.5 to 24.1
0.030
7.5
1.5 to 66.1
0.002
MT1F
-3.1
-13.9 to 1.8
0.006
-2.2
-14.2 to 2.4
0.032
MT1H
-5.7
-52.6 to 1.8
0.004
-2.3
-35.7 to 3.3
0.08
CYP1A2
1.1
-4.5 to 4.2
0.842
2.3
-1.4 to 9.0
0.008
SLC2A3
-1.7
-7.5 to 1.4
0.115
-3.0
-17.9 to -1.4
0.000
SHBG
1.8
-2.1 to 8.2
0.064
3.6
1.2 to 11.3
0.001
OSTalpha
1.7
-1.5 to 4.0
0.018
2.4
1.2 to 4.2
0.000
P4HA1
3.0
-3.1 to 12.9
0.015
2.6
-1.9 to 11.1
0.007
CRYAA
-1.5
-7.9 to 3.6
0.262
-1.9
-9.7 to 2.1
0.054
HBA1
-1.6
-13.2 to 3.5
0.266
-3.1
-12.5 to 1.2
0.002
RPL10A
1.3
-1.3 to 2.5
0.03
1.3
-1.2 to 2.1
0.025
SERPINA11
1.4
-2.2 to 2.9
0.120
2.2
1.1 to 4.9
0.000
TMEM154
-2.1
-10.8 to 1.9
0.040
-4.9
-27.8 to 1.7
0.004
TMEM45B
-2.1
-13.0 to 4.2
0.078
-2.4
-19.2 to 2.7
0.047
* Gene fold change, confidence interval and p-values were calculated using REST 2009.
172
Several genes that were differentially regulated after RYGB surgery belong to
biological processes involved in type 2 diabetes or were affected by presence of
diabetes. Figure 5.4-2 presents the mean gene expression (2-ΔCt) of these genes
relative to 18s before and after RYGB surgery. INHBE codes for a secreted protein
involved in regulation of insulin secretion. Mean INHBE expression tended to be
higher in the sT2DM group before RYGB surgery, although this did not reach
statistical significance. INHBE expression had a statistically significant decrease in
both groups, while individuals with type 2 diabetes tended to have a greater decrease
in INHBE expression (Figure 5.4-2). After RYGB surgery, INHBE expression was the
same in both the sNGT and sT2DM group. IGFB2 expression tended to be lower in
the sT2DM group before RYGB surgery (p=0.09, students t-test). Although both the
sNGT and sT2DM group had a significant increase in IGFB2 expression, the
magnitude of the increase was greater in individuals who had type 2 diabetes (Figure
5.4-2). Expression of IL1RAP was also lower in the sT2DM group before RYGB
surgery (p=0.01, students t-test) and had a statistically significant increase after
RYGB surgery in the sT2DM group only (Figure 5.4-2).
ATF3, BHLHE40 and CDKN1A all appeared to be affected by presence of diabetes.
ATF3 is involved in regulation of gluconeogenesis and it tended to be expressed at a
higher level in the sT2DM group, although this was not statistically significant.
However, individuals with type 2 diabetes had a -11.36 fold decreased in ATF3
expression, which was higher than the -5.6 fold decrease seen in the sNGT group.
Expression of BHLHE40 also tended to be higher in the sT2DM group at the time of
RYGB surgery (p=0.07, students t-test) and the change in expression after RYGB
surgery was only statistically significant in the sT2DM group. The magnitude of the
decrease in the sT2DM group (-3.1 fold) was greater than that seen in the sNGT group
(-1.5 fold), which was statistically significant (p=0.04, 2 way ANOVA). Although
CDKN1A expression was decreased to a similar extent in both groups, the sT2DM
group had a statistically significant (p=0.006) higher expression even after RYGB
surgery (Figure 5.4-2).
173
Figure 5.4-2 Mean ± SE INHBE, IGFB2, ATF3, BHLHE40, CDKN1A, and IL1RAP expression
presented as 2-∆Ct normalised to 18S in the sNGT (●, n=8) and the sT2DM (○, n=8) group before and
after RYGB surgery. *Gene expression change was statistically significant (p<0.05, paired t-test)).
174
5.4.2 Differentially expressed genes between individuals with normal glucose
tolerance and type 2 diabetes that were confirmed by TLDA RT-qPCR
Genes that were found to be differentially expressed between the sNGT and sT2DM
with microarray analysis were also confirmed with TLDA RT-qPCR. Figure 5.4-3
presents gene expression relative to 18S. Regardless of whether they had RYGB
surgery, individuals with type 2 diabetes had significantly lower hepatic mRNA
expression of CYP2C19 (-2.9 fold, 95% CI -11.5 to 2.0), DBH (-2.7 fold, 95% CI -25
to 2.3) and IGFALS (-1.6 fold, 95% CI -6.0 to 1.9) in comparison to individuals with
normal glucose tolerance (Figure 5.4-3). Conversely, hepatic mRNA expression of
IGJ (2.6 fold, 95% CI -2.2 to 20.3), LGALS4 (2.5 fold, 95% CI -1.68 to 10.0)),
PYROXD2 (3.4 fold, 95% CI -1.3 to 11.2) and SAA1 (3.0 fold, 95% CI -2.3 to 15.9)
was significantly elevated in individuals with type 2 diabetes (Figure 5.4-3).
Figure 5.4-3. Mean ± SE CYP2C19, DBH, IGFALS, IGJ, LGALS4, PYROXD2, and SAA1 expression
(2-ΔCt) in the sNGT (●, n=8) and the sT2DM (○, n=8) before and after RYGB surgery. *The difference
in gene expression was statistically significant (p<0.05, t-test).
175
5.4.3 LGALS4 protein abundance is elevated in individuals with type 2 diabetes
Elevated gene mRNA expression does not always translate directly to elevated protein
abundance. LGALS4 protein abundance was assayed by western blotting to further
confirm the quality of the microarray data. LGALS4 protein abundance was assayed
in the study groups used previously in this thesis which consisted of individuals with
normal glucose tolerance (NGT) or type 2 diabetes (T2DM). Individuals with type 2
diabetes tended to have higher abundance of LGALS4 protein in comparison to
individuals with normal glucose tolerance (Figure 5.4-4). LGALS4 may be related to
increased concentrations of carbohydrates like glucose. Out of the 5 clinical
parameters recorded (HOMA-IR, BMI, HbA1c, fasting plasma insulin and glucose),
LGALS4 protein abundance only positively correlated with fasting plasma glucose.
Figure 5.4-4. (A) Mean + SE liver LGALS4 protein abundance (relative to actin) of NGT (n=17) and
T2DM (n=26) groups. See Appendix for complete western blot. (B) Plot of liver LGALS4 protein
relative abundance (log) vs fasting plasma glucose concentration (log).
176
5.5 Discussion
The progression from insulin resistance to frank hyperglycaemia and type 2 diabetes
is a long and complex process. Although animal models have advanced our
understating of this progression, lack of human studies in key tissue such as the liver
has limited our understanding of the holistic view of type 2 diabetes pathogenesis.
Furthermore, limited access to tissue such as liver may have hindered the recognition
of novel genes or molecular pathways involved in this disease. RYGB surgery
improves insulin sensitivity and thus provides a model that can help identify novel
genes related to insulin resistance or type 2 diabetes.
In this study, we used microarray analysis to assay global liver gene expression at
RYGB surgery and ~16 months later. All individuals had a significant decrease in
fasting plasma insulin levels and an improvement in insulin sensitivity. Individuals
who had type 2 diabetes at RYGB surgery were in remission by the second operation.
Grouping individuals according to diabetic status allowed us to identify candidate
genes affected by presence or absence of hyperglycaemia and hyperinsulinemia. A
complete review of differential gene expression after RYGB surgery and how it
relates to the phenotypical changes seen in type 2 diabetes is available in Chapter 6.
The following discussion is focused on genes that were confirmed with TLDA RTqPCR and were differentially regulated after remission of type 2 diabetes. In all cases
TLDA RT-qPCR showed a greater change in gene expression, most likely related to
the saturation of the beads on the microarray chip, and the higher dynamic range
inherent in RT-qPCR.
A key observation in this study was differential expression of genes that code for
secreted proteins involved in mediating insulin secretion or glucose homeostasis.
Expression of a recently identified activin, INHBE,570 tended to be higher in
individuals with type 2 diabetes and decreased after RYGB surgery (Figure 5.4-2).
The liver secretes INHBE in response to insulin stimulus, which then goes on to act
on the pancreas to inhibit insulin secretion.571-573 These data suggest that in a natural
state, INHBE acts as a break for excessive insulin secretion. However because of the
hyperinsulinaemia present in individuals with insulin resistance, pathological
overexpression of INHBE may result in negative effects on the pancreatic cell.
177
Indeed, overexpression of INHBE in mice undermined growth of pancreatic exocrine
cells.573 Raised expression of INHBE in individuals with marked insulin resistance
may contribute to the gradual loss of beta cell function that eventually precipitates
overt diabetes. DeFronzo and Abdul-Ghani (2011) have suggested that development
of pharmacological intervention that preserves 20-30% of pancreatic β-cell function
would prevent progression from impaired glucose tolerance to type 2 diabetes.229
INHBE is a promising target for such treatments.
IGFBP2 is another gene secreted form the liver that had differential expression after
RYGB surgery. IGFBP2 has a role in mediating metabolic homeostasis574 and
investigators have documented increased expression after significant weight loss
following RYGB surgery.545 In this study, IGFBP2 expression tended to be lower in
individuals with T2DM at RYGB surgery in comparison to individuals who had
normal glucose tolerance and increased 7.5 fold after remission of diabetes. Our
results agree with that of Li et al. (2012) who showed circulating IGFBP2 increased
within 24 hours and remained raised up to a year after bariatric surgery.575 Several
animal models have shown IGFBP2 to have anti-diabetic effects.576, 577 A study by
Hedbacker et al. (2010) found liver-specific IGFBP2 overexpression reversed
diabetes in insulin resistant or obese mice.576 The same study also found that leptin
increased IGFB2 levels, suggesting that IGFBP2 signalling is responsible for leptins
effect on metabolic homeostasis.576 These findings have great implications given that
raised CRP levels induce leptin resistance by inhibiting leptin signalling.578 Notably,
CRP had decreased expression after RYGB surgery, particular in those individual
who had remission of T2DM (Table 5-4), while in the same individuals IGFBP2
expression increased. This presents a potential mechanism whereby inhibition of
leptin signalling by CRP decreases IGFBP2 levels resulting in worsened glycaemic
control.
Multiple investigators have used genome wide association studies to identify up to 18
genes that are consistently associated with type 2 diabetes.584,
585
Several of these
genes mediate pancreatic β-cell function, which is in agreement with the differential
INHBE expression observed by us and with the hypothesis that β-cell dysfunction is
critical for progression to type 2 diabetes.579, 580
178
Of all the genes identified by us, only IGFBP2 has been associated with type 2
diabetes,
and
several
genome-wide
association
studies
having
identified
polymorphisms near the IGFBP2 gene as a risk factor for type 2 diabetes.581, 582
Because RYGB surgery causes dramatic changes in intestinal physiology, identifying
genes directly involved in regulation of metabolism can be problematic. Indeed,
expression of SDS decreased after RYGB surgery most likely due to dietary
cobalamin (vitamin B12) deficiency,583 which is a well known complication of
bariatric surgery.584-586 Confirmed genes had similar expression regardless of diabetic
status, suggesting that changes in gut physiology, dietary habits and weight loss are
responsible for most of the differential gene expression. Both our study and that of
Elam et al. (2009) showed changes in expression of liver genes involved in lipid
metabolism, inflammation, cell structure, xenobiotic metabolism and cell growth after
RYGB surgery. Similarly, a study by Tamboli et al. (2011) exploring differential gene
expression in skeletal muscle after RYGB surgery with or without omentectomy
showed a decrease in inflammatory gene expression.544 Unlike previously published
studies however, our cohort clearly distinguished between individuals with normal
glucose tolerance or type 2 diabetes. Availability of liver tissue from the same
individual before and after significant increases in insulin sensitivity allowed us to
identify many differentially expressed genes that belonged to pathological processes
associated with T2DM.
Although both groups had a decrease in CDKN1A expression after RYGB surgery, its
expression was higher in individuals who had type 2 diabetes. A study by ScottDrecshel et al. (2013) associated hyperglycaemia with increased CDKN1A expression,
which caused a slowed cell cycle progression leading to reduced cellular
proliferation.587 Indeed, genes such as DUSP1, TP53INP1, GADD45B, and DDIT4
are all involved in negative regulation of cell growth and were decreased after RYGB
surgery. Individuals with type 2 diabetes tended to have a greater fold change in genes
that regulate inflammation or coagulation (CRP, SERPINE1, and A2M ). IL1RN and
IL1RAP, both of which are involved in the IL-1 signalling pathway, were
differentially expressed after RYGB surgery. IL-1 is a proinflammatory cytokine that
plays a central role in chronic inflammatory diseases588 and has recently been linked
to fat-liver cross talk in obesity.350 IL1RN is thought to inhibit proinflammatory IL-1
179
signalling by binding to the IL-1 receptor and preventing its association with the
IL1RAP coreceptor.589-591 IL1RN has been found to be increased in diet-induced
obesity,592, 593 and is likely increased to counter the proinflammatory state observed in
obesity and type 2 diabetes. The decrease in IL1RN expression after RYGB surgery is
consistent with the marked weight loss and improvement in inflammatory state that
has been documented by others.594, 595 IL1RAP is an essential component required for
IL-1 signalling through the IL-1 receptor596, 597 and has been implicated in regulation
of insulin sensitivity, pancreatic β-cell mass and leptin resistance.592, 598 IL1RAP had a
lower level of expression in individuals with type 2 diabetes and was increased after
RYGB surgery individuals who experienced remission of type 2 diabetes.
Several microarray studies have shown that insulin regulates a vast array of gene
expression.548,
599-602
There are many similarities between liver biological processes
that are affected by RYGB and those that change in skeletal muscle tissue after insulin
infusion.548, 600 In both instances, a large portion of the differential gene expression
was observed in genes which code for proteins that regulate transcription factor
binding or activity. Coletta et al. (2008)548 identified 12 genes that were differentially
regulated in parallel with our study, while four of those genes (ATF3, FOSB,
BHLHB40, and TRIB1) were involved in regulation of transcription. The similarity
between the two studies is not surprising as all individuals in this study had a decrease
in circulating insulin concentrations, which was more pronounced in those who
experienced remission of type 2 diabetes (Table 2-3).
However, differential regulation of ATF3 and BHLHE40 in the liver is more relevant
to glucose homeostasis. In this study, gene expression of ATF3 and BHLHE40 tended
to be higher in individuals with type 2 diabetes. Interestingly, increased ATF3
expression in the liver of transgenic mice led to impaired glucose homeostasis
because of its ability to inhibit gluconeogenesis.603 ATF3 expression was decreased by
-11.4 fold in individuals who experienced remission of type 2 diabetes whereas it was
decreased -4.6 fold in those individuals that only had an improvement in insulin
sensitivity. Decreased ATF3 expression is likely related to the changes in insulin
sensitivity, although the exact mechanism that involves ATF3 regulation of
gluconeogenesis remains unknown.
180
BHLHE40 (alternatively known as BHLHB2) is a pleiotropic transcription factor that
has roles in cellular process that range from metabolism to regulation of circadian
rhythm.604 Like, ATF3 Expression of BHLHE40 was decreased to a greater extent in
individuals who had remission of type 2 diabetes. Although BHLHE40 expression is
increased by insulin,548, 599, 600, 605 glucose is also able to induce BHLHE40 expression
through ChREBP activation.606 Consequently, the resolved hyperinsulinemia and
hyperglycaemia observed in individuals who have a remission of diabetes may be
responsible for the decreased BHLHE40 expression. Overexpression of BHLHE40 in
individuals with type 2 diabetes has several implications. First, BHLHE40 was found
to inhibit glucose and ChREBP-mediated activation of lipogenic genes, suggesting it
may play an important role in preventing excessive lipogenesis.606 Second, BHLHE40
may have an effect on regulating hepatic glucose metabolism through its ability to
suppress CLOCK/BMAL1 enhanced promoter activity.607 CLOCK and BMAL-1
constitute the core of the molecular clock and have been found to regulate glucose
metabolism at the liver.608-610 The relationship between metabolism and the circadian
rhythm is further strengthened by recent data that shows insulin action has a circadian
rhythm, which when disrupted caused predisposition to obesity and insulin
resistance.611 Differential BHLHE40 expression after remission of diabetes is
consistent with the increasing relevance of circadian rhythm disruption in type 2
diabetes.612, 613
The cohort used in this study also offered the opportunity to identify differential liver
gene expression between individuals with or without type 2 diabetes. DBH was
expressed at a lower level in individuals with type 2 diabetes and has previously been
associated with diabetes.614 Expression of SAA1 was increased in individuals with
type 2 diabetes and has been proposed as a potential biomarker for insulin
resistance.615 LGALS4 and PYROXD2 had higher expression in individuals with
T2DM regardless of whether they underwent RYGB surgery and are for the first time
associated with T2DM. In addition, LGASL4 protein was confirmed to be elevated in
a different set of individuals with type 2 diabetes and had a positive correlation with
glucose (Figure 5.4-4). LGALS4 (also known as galectin-4) is a part of the galectin
subfamily, which are carbohydrate binding proteins with diverse functions that range
from cell adhesion to intercellular signalling.616 LGALS4 mediates intestinal
inflammation617 and has a role in innate immunity because it binds to carbohydrate
181
moieties on the surface of bacteria resulting in rapid loss of viability.618, 619 Although
the functional characteristics of PYROXD2 are relatively unknown, its ability to
metabolise trimethylamine suggests a role in the interactions between the gut
microbiome and host.620, 621 Interestingly, the interactions between gut flora and host
metabolism have become topical for metabolic disease, while changes in gut flora
after RYGB surgery suggest the microbiome has an important role in T2DM
pathogenesis.622, 623
To conclude, although RYGB surgery induces multiple changes in gene expression,
by restricting analysis of these changes to those individuals with or without type 2
diabetes we have identified genes that may have a key role in the progression of
insulin resistance to type 2 diabetes. Of all the genes changes observed, INHBE is of
particular interest as a potential therapeutic target that might be critical to preventing
or delaying progression from insulin resistance to type 2 diabetes. Decreased
expression of INHBE in the liver after RYGB may again highlight the importance of
liver insulin resistance in type 2 diabetes and provides evidence for a hitherto
unrecognised mechanism by which the liver regulates insulin secretion.
Decreased expression of BHLHE40 only in those individuals who had remission of
type 2 diabetes provides a link between disruption of the circadian rhythm and
dysregulated metabolic homeostasis. This finding is particularly relevant given the
recent evidence showing insulin action to have a circadian rhythm and the increasing
recognition of the association between circadian rhythm disruption and insulin
resistance. The expression of several genes was markedly different between
individuals with normal glucose tolerance or type 2 diabetes. Although many of these,
including LGALS4 and PYROXD2, have no known relationship to type 2 diabetes, the
increasing awareness of the gut microbiome and its relevance to metabolic
homeostasis may be of real importance.
182
CHAPTER SIX: Global changes in hepatic gene
expression after RYGB surgery reflect the pathology
of type 2 diabetes
183
Obesity-associated type 2 diabetes is a polygenic disease caused by a combination of
environmental and genetic factors.215 The central notion governing development of
type 2 diabetes is the progressive insulin resistance that causes impaired glucose
tolerance eventually leading to pancreatic β-cell failure and fasting hyperglycemia. In
the last 50 years there has been a growing knowledge of the molecular processes that
drive the progression to type 2 diabetes. Chief among them are lipotoxicity and
inflammation, both of which are thought to contribute equally to the development of
insulin resistance.
RYGB surgery induces drastic changes in diet and involves significant manipulation
of the gastrointestinal system, so it is not surprising that there are changes in gene
expression regulating a wide range of biological processes. Genes such as LDLR,
FASN, PLIN2, SQLE, FADS1 and ACSL4 regulate lipid biosynthesis or transport and
were downregulated after RYGB surgery. This mirrors the dramatic weight loss seen
after surgery and is likely associated with the decreased caloric intake or change in
diet. Genes involved in cell adhesion such as CNTNAP2 and RND3 were
downregulated, whereas genes involved in extracellular matrix remodelling such as
PTHA1 were upregulated after RYGB surgery. Infiltration of the liver by FFA, prior
to RYGB, leading to non-alcoholic fatty liver disease (NAFLD) may cause
differential regulation of genes involved in cell adhesion and ECM remodelling.
Several studies investigating gene expression in NAFLD have linked genes involved
in liver cell adhesion and lipid metabolism with increased liver fat content.624-626
Although oxidative stress is recognized as another contributing factor in diabetes,
only in the last decade has endoplasmic reticulum stress and more specifically, the
unfolded protein response (UPR) been acknowledged as a key factor in development
of insulin resistance. Similarly, the disruption of the circadian rhythm and the gut
microbiome are also increasingly becoming recognized as key factors that contribute
to development of type 2 diabetes. This chapter provides a brief literature review of
the key gene changes observed in liver tissue after RYGB surgery or after remission
of type 2 diabetes and how they relate to these pathological processes.
184
6.1 Glucose metabolism
Dysregulation of gluconeogenesis is thought to contribute to the fasting
hyperglycaemia seen in type 2 diabetes.19,
23, 105, 196
SIK1 and ATF3 negatively
regulate gluconeogenesis and have diminished expression after RYGB surgery. SIK1
is a part of the SIK family proteins that have a wide variety of biological functions
including an increasingly important role in metabolic regulation.627 SIK1 has been
shown to inhibit gluconeogenesis by regulating CREB-mediated transcription of
gluconeogenic genes628,
629
and lipogenesis by regulating SREBP-1c-mediated
transcription of lipogenic genes.630 ATF3 also inhibits gluconeogenesis by repressing
transcription of PCK1.603 The decreased expression of SIK1 and ATF3 is likely
associated with the normalisation of overactive lipogenesis and gluconeogenesis seen
after RYGB surgery.
PCK1 positively regulates gluconeogenesis because it codes for an enzyme (PEPCK)
that catalyzes one of the first steps in the gluconeogenic reaction. The inhibitory
action of insulin on PCK1 transcription has been well established,
190, 631-634
while
silencing PCK1 expression in the liver has been shown to improve glucose
homeostasis, insulin sensitivity and dyslipidemia in diabetic mice.536PCK1 expression
was only increased in those individuals who experienced remission of diabetes (Table
5-2). This is likely linked to the decreased expression of ATF3, which was more
pronounced in individuals who had remission of type 2 diabetes. Even so, the
increased expression of PCK1 after normalisation of hyperglycaemia initially appears
to be paradoxical.
It is critical to note, however, that both the pre and post RYGB liver biopsies were
taken under fasting conditions and there is evidence to suggest that insulin regulation
of PCK1 transcription may differ in the fed and fasting state.535 Although the ability
of insulin to regulate PCK1 mRNA expression was impaired in ad libitum insulinresistant zucker fatty rats, the impairment was partially restored after an overnight
fast.535 Furthermore, individuals in this study who had remission of diabetes also
displayed a marked decrease in fasting insulin concentration which likely had a role in
inhibiting PCK1 transcription. Taken together these data suggest that regulation of
gluconeogenesis is a complex process that is regulated by multiple checks and counter
185
balances. Elucidating how SIK1, ATF3 and PCK1 act in concert to regulate
gluconeogenesis will aid in the understanding the pathogenesis of type 2 diabetes.
6.2 Regulation of metabolic homeostasis via leptin signalling
Leptin signalling through the leptin receptor (LEPR) is involved in satiety and
metabolic homeostasis. Expression of LEPR is decreased after RYGB surgery. In
healthy individuals leptin prevents obesity via LEPR by stimulating glucose uptake
and fatty acid oxidation.635,
636
Several mutations of the LEPR gene have been
associated with type 2 diabetes and the metabolic syndrome.637-640 Interestingly, the
liver was recently suggested to have a role in modulating leptin action641 via the
production of soluble leptin receptors.642 LEPR expression was shown to increase
after treating mice with leptin which consequently increased soluble leptin receptors
levels. Obesity is associated with increased levels of circulating leptin and bariatric
surgery has been shown to significantly decrease leptin levels.643-646 It can be inferred
that the decreased LEPR expression observed in this study is likely linked to the
decreased level of leptin seen after bariatric surgery. How this affects leptin signalling
is yet to be determined. Nonetheless, leptin signalling is known to have a significant
role to play in type 2 diabetes and the associated metabolic syndrome. Specifically,
increased levels of circulating leptin in the face of obesity have inspired the concept
of “leptin resistance”.647-649 Akin to insulin resistance, leptin resistance may have a
role to play in insulin sensitivity.650
Data presented in Chapter 5 provides evidence for a hitherto unconfirmed relationship
between leptin, CRP and IGFB2. Elevated CRP levels have been shown to induce
leptin resistance by inhibiting leptin signalling.578 Notably, expression of CRP was
decreased after RYGB surgery, particular in those individual who had remission of
type 2 diabetes (Table 5-4, Chapter 5). Although its expression is increased in liver
and adipose tissue in severely obese individuals,269, 651 its role in type 2 diabetes is
controversial. Some studies have shown CRP to be associated with increased risk of
developing type 2 diabetes,652, 653 while others have not.651, 654
186
Even though the mechanism behind the involvement of CRP in type 2 diabetes is yet
to be elucidated, its ability to inhibit leptin signalling578 is an interesting finding in
light of a recently identified relationship between leptin and IGFB2 levels.
Leptin has been shown to increase levels of IGFB2,576 which has a well documented
role in glucose homeostasis.574 In this study, IGFBP2 expression was increased after
RYGB surgery, which again was more pronounced in those individuals who had
remission of type 2 diabetes. A likely scenario in individuals with type 2 diabetes may
be that elevated levels of CRP inhibit leptin signalling, thus decreasing IGFB2 levels
which would have detrimental affect on glucose homeostasis. Accordingly, data
presented in Chapter 5 show elevated expression of CRP and depressed expression of
IGBF2 during diabetes, which is normalised after RYGB surgery and remission of
diabetes. The observation that there is a relationship between leptin, CRP, and IGFB2
is in accord with the increasingly evident relationship between inflammation and
metabolic homeostasis.303, 655
6.3 Lipotoxicity and non-alcoholic fatty liver disease
Dysfunctional lipid homeostasis in the context of obesity has been recognized as a
key pathology in type 2 diabetes. Accumulation of lipids in insulin target tissues,
otherwise known as lipotoxicity, disrupts insulin singling and is one of the
contributing factors that lead to insulin resistance. In the liver, lipotoxicity manifests
itself as non-alcoholic fatty liver disease (NAFLD),656 which when reversed has been
shown to improve insulin sensitivity and normalize hyperglycemia.256, 257 Similarly,
NAFLD has been shown to improve after bariatric surgery and weight loss.657, 658
In this study several genes that are associated with lipid homoeostasis and
accumulation were differentially regulated after RYGB surgery. Sterol regulatory
binding proteins (SREBPs) are transcription factors central to mediating lipid
catabolism and lipid biogenesis in the liver.200,
320
They regulate up to 30 genes
directly involved in uptake and synthesis of fatty acids, triglycerides, cholesterol and
phospholipids.200
187
LDLR and FASN are involved in the SREBP pathway and were both decreased after
RYGB surgery. The LDLR gene codes for a cell surface glycoprotein that is involved
in mediating blood cholesterol levels by removing LDL particles from the
circulation.659 Insulin has been shown to mediate the levels of LDLR expression via its
660,
influence on SREBP-1 which in turn mediates activation of the LDLR promoter.
661
Furthermore, LDLR has been found to associate with the insulin receptor which
diminishes its ability to bind LDL particles.662 Insulin stimulation causes
disassociation of the LDLR and IR receptor thereby increasing the capacity of LDLR
to bind LDL particles.662 FASN catalyzes the last step in fatty acid biosynthesis663 and
is regulated by SREBP.664-666 Increased FASN expression in adipose tissue has been
linked with type 2 diabetes667 while increased FASN expression has also been shown
in NAFLD.668 Consequently the decreased expression of LDLR and FASN is likely
associated with the decreased levels of insulin seen after RYGB surgery and the
consequent improvement in lipid homeostasis.
PLIN2, a marker of lipid accumulation, had decreased expression after RYGB surgery
and remission of diabetes. PLIN2 coats lipid droplets,669 and is thought to be involved
in their intracellular mobilization.670,
671
Lipid droplets are intracellular and
cytoplasmic structures that store triglycerides in times of energy abundance.672,
673
They are implicated in metabolic disease and their role in diabetes and NAFLD is
well documented.674, 675 Increased PLIN2 expression is associated with increased lipid
droplets and fatty liver in both humans676,
677
and animal models.678 PLIN2
overexpression increased lipid accumulation without induction of other lipogenic
genes, suggesting PLIN2 alone is capable of causing lipid accumulation in cells.679
Conversely, PLIN2 null mice were resistant to diet-induce fatty liver and obesity680,
while knocking down PLIN2 in obese mice reversed hepatic steatosis,
hypertriglyceridemia and insulin resistance.681, 682 This evidence, along with our data,
suggests PLIN2 may be involved in diabetes-associated lipotoxicity.
The exact mechanisms behind the negative affect of NAFLD on insulin regulation of
glucose homeostasis are still unknown. Nonetheless, KLF6 has been identified as a
candidate molecule that links NAFLD to dysregulated glucose homeostasis. In this
study KLF6 expression was decreased after RYGB surgery (Table 5-2, Chapter 5).
188
KLF6 is a pleiotropic transcription factor that was shown to be involved in the
regulation of cell proliferation, differentiation and tumerogenesis.683 It was found to
be rapidly induced in hepatic stellate cells after injury684 and has since been associated
with liver fibrosis and NAFLD.685, 686 KLF6 was recently implicated in regulation of
hepatic glucose metabolism and insulin sensitivity via its interaction with
glucokinase.687 Furthermore, elevated KLF6 expression was associated with elevated
activation of PPARα signalling.688 Because PPARα deficiency was shown to protect
mice against insulin resistance induced by a high-fat diet, it can be inferred that
elevated expression of KLF6 has a detrimental affect on glucose and lipid
homeostasis.689
6.4 Inflammation and blood coagulation
Obesity-induced inflammation can occur via several mechanisms. Recruitment of
adipose tissue macrophages (ATMs) by chemokine signalling is thought to play a key
role.280-283 FFAs and other metabolites like glucose are also thought to have a role in
activating the inflamasome,303 which leads to induction of NF-κβ,333 TNFα and IL-1β
among others.333, 334 Furthermore, adipose derived IL-1β was shown to act on the liver
by supporting ectopic lipid accumulation in hepatocytes.350
In this study, multiple genes that have been shown to be involved in inflammation
were differentially regulated after RYGB surgery. IL1RN and IL1RAP, both of which
modulate the strength of IL-1 signalling, were differentially regulated after RYGB
surgery. This is likely associated with a possible decrease of activity in the IL-1
signalling pathway. Expression of IL-32 was also decreased after RYGB surgery. IL32 is a recently identified proinflammatory cytokine that increases expression of
TNFα and IL-1β and has been found to be have an auto-inflammatory relationship via
its synergy with TNFα.690 The decrease in IL32 expression after RYGB surgery is
consistent with the improvements observed in metabolic homeostasis.
Dysregulated blood coagulation is another pathological condition seen in obesityrelated type 2 diabetes. Cardiovascular risk is increased in individuals with type 2
diabetes691 while changes in the homeostatic mechanism have long been associated
with this disease.692, 693
189
In this study, expression of SERPINE-1 (alternatively known as PAI-1) was decreased
after RYGB surgery and remission of type 2 diabetes. SERPINE-1 is a circulatory
protein that is involved in regulation of the fibrinolytic system in blood by inhibiting
both urokinase-type and tissue-type plasminogen activators.694 Regulation of
SERPINE-1 transcription is under the control of a vast array of stimuli, including
proinflammatory
cytokines,
TNFα,
lipids,
growth
factors
and
insulin.695
Dysregulation of the fibrinolytic system, often seen in type 2 diabetes and its
complications, is associated with increased levels of SERPINE-1 expression.695,
696
Hyperinsulinemia has long been suspected as a risk factor for the development of
vascular disease16,
697
and SERPINE-1 was singled out as a major player.698 The
connection between hyperinsulinemia and insulin induction of SERPINE-1
expression699 may be a possible mechanism that links obesity, diabetes and
cardiovascular disease.696
The effects of inflammation are partially mediated by activation of transcription
factors that increase expression of proinflammatory cytokines. Enhanced activation of
liver NF-κβ and activating protein-1 (AP-1) was observed in obese patients with
NAFLD, and both were positively correlated with insulin resistance.700 AP-1 activity
can be induced by a complex set of factors that includes growth factors such as insulin
and pro-inflammatory factors such as IL-1.701-703 The AP-1 family consists of
homodimers and heterodimers of Jun (v-Jun, c-Jun, JunB, JunD), Fos (v-Fos, c-Fos,
FosB, Fra1, Fra2) or activating transcription factor (AFT2, ATF3, B-ATF, JDP1 and
JDP2).704, 705 In this study, several members of the family of transcription factors were
differentially regulated after RYGB surgery. Gene expression of c-JUN, JUND,
FOSB, and ATF3 were all decreased after RYGB surgery.
AP-1 and its constituents control a wide variety of biological processes including cell
proliferation, differentiation, apoptosis.705 JUN is a key activator of hepatocyte
proliferation and is involved in liver regeneration,706 while JUND and FOSB were
shown to be involved in wound healing and liver fibrosis.707,
708
Furthermore, JUN
and JUND were shown to be critical in protecting cells from oxidative stress.709 As
oxidative stress has been linked to the pathogenesis of insulin resistance and type 2
diabetes,710, 711 it is highly likely that the changes seen in AP-1 expression may be
related to improvement in the inflammatory state and oxidative stress.
190
Like AP-1, ATF5 expression, which is also induced by a variety of cellular stresses,712
was decreased after RYGB surgery. Although little is known of the functional
consequences of ATF5 induction in the liver, it has been shown to be involved in
cellular survival,713 proliferation and differentiation.712, 714, 715 In the liver ATF5 has
been shown to play a key role in modulating expression of CYP2B6 under stress
conditions such as amino acid limitation or chemically-induced stress.716 Increased
expression of ATF5 induced an increase in expression of CY2B6, a member of the
cytochrome
P450
family
that
is
involved
in
oxidative
metabolism
of
endobiotic/xenobiotics.716
Several members of the metallothionein family had decreased gene expression after
RYGB surgery (Table 5-2). Metallothioneins (MT) have a high affinity for metal ions
and have been implicated in regulation of zinc and copper metabolism in addition to
having anti-oxidant properties which protect against oxidative damage.717 A study by
Coletta et
al. (2008),
in
which
healthy
individuals were treated with
supraphysiological levels of insulin for four hours, showed that gene expression of
MT proteins in muscle was increased following insulin stimulation.548 This suggested
that insulin alone can activate the anti-oxidant defence system and may be a
compensatory mechanism to the oxidative stress caused by increased free radicals
from autoxidation of sugars and unsaturated lipids.710 MT’s may also play an
important role in metabolic homeostasis. Polymorphisms in MT genes, and in
particular MT1A, were associated with type 2 diabetes mellitus,718 while MT2A gene
expression was found to be increased in adipose tissue of individuals with type 2
diabetes in comparison to lean individuals.719
Further investigations revealed that TNFα increased MT2A mRNA levels and treating
3T3-L1 cells with MT2A protein significantly inhibited glucose uptake.719 There is
good evidence to suggest that MT’s have a significant role to play in the pathogenesis
of insulin resistance and type 2 diabetes, but whether this role is central to the
development or remission of the disease is yet to be established.
191
Oxidative stress caused by over-nutrition induces anti-oxidant and inflammation
pathways. Until relatively recently the mechanism linking lipotoxicity and
inflammation to insulin resistance had eluded the scientific community. In the last
decade, however, endoplasmic reticulum stress has emerged as a key process in
obesity and diabetes.308, 720
6.5 Endoplasmic reticulum stress
The endoplasmic reticulum (ER) is responsible for protein, lipid and sterol synthesis.
In addition, it mediates protein folding and post translational modifications, both of
which are necessary in order for proteins to carry out their cellular functions.721
Stressful conditions, such as hyperinsulinemia or excess nutrient intake, cause
overproduction of proteins, which overloads the ER and leads to accumulation of
unfolded proteins. This in turn activates the unfolded protein response (UPR),309
which acts to alleviate the unfolded protein overload via several responses. First it
prevents the accumulation of further unfolded proteins by decreasing protein
synthesis.722 Second, it increases expression of chaperon proteins that accelerate
protein folding.721, 723 Finally, if ER function is severely impaired, the organelle elicits
apoptotic signals that eliminate damaged cells.724, 725
ER stress and the UPR have been recognised as key processes in type 2 diabetes that
link obesity-related inflammation to insulin resistance.244, 308, 720 FFAs were shown to
induce ER stress in cultured hepatocytes and pancreatic β cells,
726
whereas multiple
markers of ER stress were activated in adipose tissue of obese individuals.727
Furthermore, ER stress was shown to decrease in adipose and liver tissue of
individuals who had substantial weight loss after bariatric surgery.326 Similarly, in this
study several genes that are involved in processes regulated by the UPR were
differentially expressed after RYGB surgery. Expression of EEF1G, which mediates
elongation during protein translation, was increased after RYGB surgery. Depressed
EEF1G expression in the presence of obesity and type 2 diabetes is consistent with
the involvement of ER stress and its ability to decrease protein synthesis.722
Expression of BRSK2 and DDIT4, both of which were shown to be regulated by ER
stress,728, 729 was decreased after RYGB surgery.
192
DDIT4 is a stress response gene activated by hypoxia and other stimuli.730-732 Insulin
has been shown to induce DDIT4 expression in both skeletal muscle
733
and adipose
cells734 and has consequently been shown to have a role in the insulin signalling
pathway in adipocytes.735 GADD45B, DUPS1 and TP53INP1 all have roles in
regulating apoptosis and had decreased expression after RYGB surgery. GADD45B
has been shown to be induced by NF-κB during acute inflammation and is thought to
be involved in repressing cell death736 and mediating protective effects on cells
against DNA damage.737 DUSP1 belongs to the family of MAPK phosphatases that
inactivate the MAPKs and thus negatively regulate MAPK signalling.738 It was found
to be differently regulated by ER stress and had a pivotal role in ER stress-induced
apoptosis.739 Even though DUSP1 null mice are resistant to diet-induced obesity, they
develop glucose intolerance suggesting it has a complex role in regulating metabolic
homeostasis.740
Taken together, these data further confirm the involvement of ER stress and UPR in
insulin resistance and type 2 diabetes. Furthermore, ER stress at the liver may be
critical to the development of secondary peripheral insulin resistance, with hepatic
insulin resistance being a primary event in the progression to type 2 diabetes.319
6.6 The liver and inter organ communication
Despite the key roles of muscle and adipose tissue in insulin resistance, the liver is
increasingly being recognized as the organ of prime importance in the progression to
type 2 diabetes. Clinical evidence has shown that resolution of liver, but not
peripheral, insulin resistance is associated with remission of type 2 diabetes after
bariatric surgery,257,
448, 475-477
whereas improvement of NAFLD after a low calorie
diet has been associated with normalization of fasting hyperglycaemia.256,
257
Although the molecular mechanisms behind this are unclear, several studies offer
some insight. Activation of the UPR by ER stress was shown to mediate cross talk
between the liver and peripheral organ insulin sensitivity via IGFB3.319 At the same
time, iLIRKO mice, which develop peripheral insulin resistance in muscle tissue after
primary hepatic insulin resistance due to IR ablation, also had elevated levels of
IGFBP3.123
193
In this study, several liver genes that code for secreted proteins had differential
expression after RYGB surgery. These genes act on other insulin target tissue and
have been found to regulate metabolic homeostasis and insulin secretion. This adds
another dimension to the multiple lines of evidence that suggest the liver may be of
prime importance in the progression to type 2 diabetes. Interestingly, two members of
the IGFBP family (IGFBP2 and IGFBP5), which modulate the activity of IGF-I and
IGF-II,741 were increased after RYGB surgery (Table 5-2). As discussed previously,
IGFBP2 is increasingly being associated with regulation of metabolic homeostasis.574,
576, 577
IGFBP5, however, is a secreted protein that has been shown to have a role in
promoting growth and proliferation of the pancreatic β-cell.742 The role the liver plays
in regulating insulin release from the pancreas is of great importance considering the
final event in the progression to fasting hyperglycaemia and type 2 diabetes is thought
to be β-cell failure. Indeed, expression of several other genes which were found to
have roles in regulating insulin secretion was also changed after RYGB surgery. Most
notable of these is INHBE, which as discussed previously, may be a potential
therapeutic target for preventing progression to type 2 diabetes.
The fact that INHBE is secreted by the liver and acts directly on the pancreas to
regulate insulin secretion571-573 demonstrates again the central role the liver appears to
have in regulating the consequences of insulin action. Another gene that may affect
pancreatic β-cell size is MAP2K, expression of which was increased after RYGB
surgery. Imai et al. (2008) have shown that liver ERK1/2 activation by MAP2K can
induce pancreatic β-cell growth via neuronal signals from the liver.743 This suggests
that increased liver MAP2K1 expression after RYGB may increase β-cell mass,
which in turn could contribute to the observed increase in insulin sensitivity.
FNDC5 is another recently identified gene that exemplifies the central role of the
liver. It was identified by Boström et al. (2012)744 as a hormone that is cleaved and
secreted into the circulation. FNDC5 was initially found to be secreted from muscle as
Irisin and early work suggested that its primary function is to induce browning of
white adipose fat (WAT).744-746 Brown adipose tissue (BAT) has been found to be
lower in obese individuals,747 while increasing the amount of BAT in rodents reduced
body weight and improved glucose homeostasis.748
194
FNDC5 is induced by exercise 744, 749, 750 and has consequently been suggested to be a
possible treatment for obesity and insulin resistance.751 However, data presented here
shows that FNDC5 expression is decreased in the liver tissue of individuals who
undergo RYGB surgery. This is in accord with the study of Huh et al. (2012) which
showed a similar decrease in FNDC5 expression levels in muscle following bariatric
surgery.750 Further work is needed to elucidate the exact mechanisms governing
FNDC5 function and the relevance of its expression in the liver.
Evidently the liver has a critical role in inter-organ communication that leads to
regulation of glucose homeostasis and secretion of insulin that is only now becoming
recognised. The central role it occupies as a master regulator of metabolic
homeostasis is also apparent in its regulation of circadian metabolic activity.752
6.7 Circadian rhythm
Regulation of metabolism in the liver follows diurnal patterns that are governed by
the light/dark cycle and food availability.752 The modern lifestyle of high caloric
intake and sleep deprivation can impair the central circadian clock, but the impact it
has on the molecular clock in the liver is unclear.
Nonetheless, disruption of the circadian rhythm has been documented to increase risk
of obesity and diabetes.753-757 Interestingly, hepatic insulin sensitivity may also have a
diurnal pattern. At the time of writing, a key study by Shi et al. (2013) demonstrated
that insulin action has a circadian rhythm and that disrupting its rhythm increased the
risk of acquiring insulin resistance and obesity.758 Although the link between
circadian disruption and insulin resistance has been established, the molecular
mechanisms governing molecular clock participation in insulin-mediated metabolism
are yet to be fully identified. This study presents three potential candidate genes that
are involved in the circadian rhythm and may have a role in the pathogenesis of type 2
diabetes.
Several genes involved in molecular clock regulation of metabolic homeostasis were
differentially expressed after RYGB surgery. Expression of BHLHE40 and CCRN4L
was decreased after RYGB surgery, whereas expression of ARNTL was increased.
195
All of these genes have been found to have an impact on regulation of metabolic
homeostasis. BHLHE40, as discussed previously, regulates glucose metabolism
through its ability to suppress CLOCK/BMAL1 enhanced promoter activity.607,
610
CCRN4L is one of a host of genes associated with control of circadian rhythm and
has been implicated in regulating the metabolic processes and inflammation.759, 760 It
has deadenylase activity and was shown to remove poly(A) tails from mRNA causing
destabilization of mRNA and providing post transcriptional silencing.761,
762
Mice
lacking CCRN4L have reduced expression of lipogeneic genes and are resistant to
diet-induced obesity and hepatic steatosis.763 Recently, CCRN4L from adipose tissue
was shown to negatively correlate with plasma insulin levels and HOMA-IR measure
of insulin resistance,764 suggesting it has a key role to play in type 2 diabetes.
Decreased expression of BHLHE40 and CCRN4L is most likely associated with the
observed decreases in lipogenic gene expression and improvement of glucose
homeostasis observed after RYGB surgery.
Interestingly ARNTL (alternatively know as BMAL1) is another gene involved in
circadian rhythm regulation that was found to regulate CCRN4L expression.765
Whereas CCRN4L decreases after RYGB surgery, ARNTL expression increased.
ARNTL is a major component of the positive and negative transcriptional feed back
loops that belong to the intracellular clock mechanism driving rhythmic gene
expression.766 ARNTL null mice were shown to have increased levels of triglycerides,
free fatty acids and cholesterol which resulted in ectopic fat formation in the liver and
skeletal muscle.767 Furthermore, ARNTL knock out mice were locked into a state
where insulin action was at its lowest,758 suggesting ARNTL may be a molecular link
between the circadian rhythm and regulation of insulin action. The fact that it
increases after RYGB surgery is consistent with this hypothesis.
BHLHE40, CCRN4L and ARNTL are presented here as candidate genes that provide
a link between disrupted circadian rhythm and the aberrant regulation of metabolic
homeostasis seen in type 2 diabetes. Further work on the molecular mechanisms
behind this process will add to our understating of type 2 diabetes and may lead to
improvement in therapies that both prevent and treat type 2 diabetes.
196
6.8 Concluding remarks
Using the RYGB surgery as a human model to study type 2 diabetes has yielded new
associations and strengthened old concepts. The pathological role lipotoxicity,
inflammation, and ER stress have in type 2 diabetes was evident in the decreased
expression of genes involved in these processes after remission of the disease.
Evidence was also provided for the previously hypothesised link between
inflammation and metabolism via the reaction of CRP, IGFB2 and leptin signalling.
Furthermore, the association between the disruption of the circadian rhythm and
increased risk of obesity and insulin resistance was observed in the three genes
involved in regulating circadian gene expression that were differentially expressed
after RYGB surgery (BHLHE40, CCRN4L, and ARNTL).
The hypothesis that the liver may be of prime importance in the progression to type 2
diabetes is supported by the differential expression of genes that code for secreted
proteins that act on other insulin target tissues. Most notable of which is the liver’s
ability to regulate insulin secretion by the pancreas via INHBE.
In conclusion, use of human models is a rare and priceless tool for the study of
disease progression. Although use of animal models can dissect key pathways, it is
difficult to design a model that substitutes for the variability of the human population
in both genetic and environmental aspects. Type 2 diabetes, which was initially
strongly associated with obesity, is now associated with multiple pathological process
that includes lipotoxicity, inflammation, ER stress and disruption of the circadian
rhythm. Data presented here identifies the liver as a key organ in the progression to
type 2 diabetes and identifies several candidate genes that may have critical roles in
these processes.
197
198
CHAPTER SEVEN: Final discussion and experimental
limitations
199
Humanity is in the midst of an insidious pandemic of type 2 diabetes mellitus. Fuelled
by improvements in life expectancy and living standards, the prevalence of type 2
diabetes is predicted to increase to ~552 million by the year 2030,1, 2 at which point it
is projected to be the 7th leading cause of death worldwide.768 Our understanding of
type 2 diabetes pathogenesis, or lack thereof, is hindering the effective treatment of
this disease. Currently there are no known pharmacological interventions that can
induce remission of type 2 diabetes. Bariatric surgery, however, has been shown to
cause rapid remission of type 2 diabetes and insulin resistance.430, 431, 448 This surgery
presents researchers with an opportunity rarely afforded to other fields whereby tissue
from human individuals can be sampled before and after remission of type 2 diabetes.
Although animal models have improved our understanding of some of the molecular
processes in type 2 diabetes, they do not adequately model the complex development
of metabolic syndrome and diabetes in humans. Transgenic animal models often have
overexpression or complete ablation of a single gene, whereas high fat diet-induced
diabetes models do not mirror the complex nature of human feeding behaviour. With
this in mind, the use of the RYGB operation provides a unique opportunity to study
the disease and its remission in humans and was used in this study to produce several
novel findings which are summarised in the following text.
Insulin resistance is the fundamental pathology present in type 2 diabetes and has
different consequences in the various insulin target tissues. The liver is responsible for
a wide variety of biological process regulated by insulin which includes: cell growth,
lipid and glucose homeostasis, and regulation of insulin concentrations throughout the
body via hepatic insulin clearance. Unregulated hepatic gluconeogenesis is thought to
be a critical component of the fasting hyperglycaemia in late stage type 2 diabetes,19,
21, 23, 515
and multiple lines of evidence suggest that hepatic insulin resistance may lead
to secondary peripheral insulin resistance and glucose intolerance.123,
319, 348
Tissue
specific IR knockout mice show that of the three traditional insulin target tissues
involved in glucose homeostasis, the liver may be of greatest importance.111, 116, 123, 516
The liver also appears to be a vital organ in the remission of type 2 diabetes after
RYGB surgery.448, 476, 477, 481
This thesis focused on the liver and the critical role it has in regulation of metabolic
homeostasis. Liver tissue was biopsied from individuals undergoing RYGB surgery
200
and again from a subset of individuals at a subsequent operation. RNA and protein
extracted from this liver tissue was used to investigate levels of molecules previously
implicated as well as to identify novel molecular processes involved in the
pathogenesis of type 2 diabetes. To achieve this, all individuals included in the study
cohort were metabolically characterised and grouped according to presence of normal
glucose tolerance (NGT group) or type 2 diabetes (T2DM group).
Mean fasting plasma glucose and HbA1c levels were normal in the NGT group and
significantly elevated in the T2DM group (Table 2-2). Although insulin resistance
was present in varying severity in all groups, the T2DM group was the most insulin
resistant (Table 2-2). After RYGB surgery, however, both groups had a significant
decrease in fasting plasma insulin concentrations and HOMA-IR values (Figure 2.32A and C). Resolution of hepatic insulin resistance was associated with remission of
diabetes in 88% of individuals, which is similar to remission rates previously
reported.14 Those individuals who had a subsequent liver biopsy (sNGT and sT2DM
group), and were used to model diabetes remission, were shown to be representative
of the main study population (Figures 2.3-1, 2.3-2 and 2.3-3). Importantly, there were
no differences in BMI between any of the groups at any time point measured allowing
us to control for confounding variables such as weight loss.
However, there are several limitations regarding the study cohort. There is no control
group of individuals that did not have diabetes or RYGB surgery. This was
unavoidable because of the difficulty in sampling liver tissue from human individuals.
Even though we made comparisons between individuals with and without type 2
diabetes, this was done to delineate differences in diabetes and normal glucose
tolerance rather than to control for experimental manipulation. Human populations are
notoriously variable and even though this is one of the strengths of this thesis, it is
also one of its weaknesses. Because of the relative rarity of liver tissue biopsied from
the same individual before and after surgery there was little flexibility for controlling
for study group demographics. Individuals with type 2 diabetes were on average older
than those with normal glucose tolerance perhaps reflecting the progressive profile of
this disease. Furthermore, there were a greater proportion of females in the study
cohort. Although gender does effect expression of certain genes, this is unlikely to be
the case with genes that are involved in metabolic homeostasis.
201
Another caveat is the fact that almost half of the individuals with type 2 diabetes were
on some form of pharmacotherapy, which may have affected some of the gene
changes observed in the microarray analysis, particularly those involved in xenobiotic
metabolism. Even so, the varied therapies the individuals with type 2 diabetes were on
before remission of diabetes did not appear to affect the conclusions reached in this
thesis. Finally, the sample size for the individuals who had a second liver biopsy was
small which presents another limitation, although this is somewhat mitigated by the
fact that the data is paired, which increased the power of the study. Overall the data
presented here are of considerable value to the study of type 2 diabetes.
7.1 Insulin processing at the liver: a new role for ENPP1
The liver is the first organ to receive insulin secreted from the pancreas and hepatic
insulin clearance is critical in regulating the peripheral insulin concentration. Hepatic
insulin clearance has been shown to increase after RYGB surgery.482 Data presented
in Chapter 3 shows that this is not due to changes in levels of CEACAM-1 and IDE,
both of which mediate the rate of insulin clearance via regulation of insulin
internalization66 and degradation63 respectively.
A major finding in this study is the identification of a novel relationship between liver
ENPP1 and insulin signalling at the liver. ENPP1 protein has previously been found
to be increased in muscle, adipose and skin tissue of insulin-resistant individuals and
its ability to inhibit insulin signalling has lead to the hypothesis that it may cause
insulin resistance.167-170 Because of this we assayed ENPP1 mRNA expression and
protein abundance in individuals with and without type 2 diabetes before and after
RYGB surgery. Somewhat surprisingly, in this study, ENPP1 protein levels were
lower in individuals with type 2 diabetes when compared to those individuals with
normal glucose tolerance. Furthermore, those individuals who experienced remission
of type 2 diabetes after RYGB surgery had an associated increase in ENPP1 protein
levels.
Although this is the reverse of what has been reported previously in the literature, this
and other evidence suggests that the increase in ENPP1 after remission of diabetes
may be associated with its purported role as a desensitizer of insulin signalling.503
202
Considering the liver is exposed to supraphysiological levels of insulin during first
phase insulin secretion, a mechanism which prevents overstimulation of the biological
pathways may be necessary for normal liver function. The role of ENPP1 as a
desensitizer of insulin signalling has previously been hypothesised by Menzaghi et al.
(2003) who showed that insulin stimulation promotes ENPP1 recruitment to the cell
membrane where it can inhibit insulin signalling.503 Data presented in Chapter 3 that
shows ENPP1 is inversely correlated to HOMA-IR levels further supports its role as a
natural modulator of insulin signalling and suggests that the higher the insulin
sensitivity the more liver ENPP1 is needed to dampen insulin signalling.
Indirect evidence also suggests that ENPP1 may be closely associated with the first
phase insulin response. Individuals with type 2 diabetes have a diminished first phase
insulin response,52 which has been shown to return to normal after gastric bypass
surgery.471 At the same time, our data shows that ENPP1 is lower in individuals with
type 2 diabetes and increases after RYGB surgery. Taken together, these data suggest
that one of the roles of liver ENPP1 may be to act as a natural desensitizer of insulin
signalling in response to first phase insulin secretion.
Although the work presented in this thesis suggests that ENPP1 has natural
modulating action on insulin signalling strength, it cannot yet be concluded that
ENPP1 has no role in insulin-resistant pathology. Indeed, the relatively rare Q121
variant has been shown to have a greater inhibitory effect on insulin receptor than the
more common K121 variant506 and the ENPP1 K121Q polymorphism has been
associated with type 2 diabetes.59, 60 Further research on the role of ENPP1 in type 2
diabetes will likely focus on the frequency of the K121Q polymorphism, although
data presented in this thesis suggest that ENPP1 may not be as relevant as previously
thought.
There are some limitations to the study design and experimental methods used in
Chapter 3. Although the technology used in this thesis has been tried and tested, in
some cases better alternatives are available. Western blotting is an excellent technique
to assay semi-quantitative levels of protein. However for increased accuracy and
sensitivity, enzyme linked immunosorbent assay (ELISA) should have been used to
assay for differences in protein concentration between study groups.
203
Due to funding and time constraints; however, this was not achievable. Nonetheless
the relationship between liver ENPP1 protein content and hepatic insulin processing is
evident with the western blot data provided.
7.2 Insulin receptor isoforms in diabetes: detrimental affects of IRA overexpression on regulation of glucose homeostasis
Another important finding of this work is the observation that the liver IR isoform
ratio is altered in type 2 diabetes and is normalised after RYGB surgery. When they
were first identified the IR isoforms were quickly shown to have different functional
characteristics. Multiple studies have investigated whether the IR isoform ratio is
altered in muscle and fat tissue of individual with type 2 diabetes or insulin resistance,
but differences in technique, and variability of study population led to conflicting
results in the literature.418, 419, 421, 423, 424, 426, 518
Early work on the functional characteristics showed that IR-A bound and
internalized373,
374
insulin more efficiently, whereas insulin binding to the IR-B
isoform induced stronger activation of the IR tyrosine kinase.376-378 It is generally
accepted that IR-A mediates mitogenic signalling (because of its association with fetal
and cancer cells),367,
368, 382
and IR-B predominantly mediates metabolic signalling
(because of its association with metabolically active tissue).365,
366
Because of the
purported metabolic signalling through the IR-B isoform, it is not surprising the liver
IR-B:A ratio was found to be 9.8 in normal lean individuals.366 However, data
reported in Chapter 4 shows that the IR B:A ratio is decreased during type 2 diabetes.
Obese individuals with type 2 diabetes had a significantly lower IR-B:A ratio (5.2)
than obese individuals without type 2 diabetes (6.6). In addition, the IR-B:A ratio
increased from 5.4 to 8.6 only in those individuals who experienced remission of
diabetes.
This change in IR-B:A ratio was caused by a significantly reduced expression of IR-A
isoform, suggesting that abnormally elevated levels of IR-A may have functional
consequences for insulin signalling during type 2 diabetes.
204
Indeed, overexpressing IR-A isoform in HepG2 cells had no affect on AKT activation
and reduced insulin’s ability to inhibit PCK1 transcription, whereas overexpressing
IR-B caused a 50% increase in AKT activation and tended to improve insulin
downregulation of PCK1.
Consequently it can be inferred that the abrogated IR-B:A ratio in type 2 diabetes may
have a critical role in dysregulated gluconeogenesis. In accordance with our data,
elevated expression of IR-A in muscle tissue of individuals with MD1, which is
characterised by severe hyperinsulinaemia, has also been associated with impaired
glucose homeostasis.415 Furthermore, because IR-A is a predominant mitogenic
receptor, an increasing awareness of the link between hyperinsulinemia and
carcinogenesis428 provides circumstantial evidence for a relationship between
hyperinsulinemia and increased IR-A expression. Although elevated levels of IR-A
appear to be associated with hyperinsulinaemia, treating Hep G2 cells with
supraphysiological levels of insulin did not induce changes in IR-A isoform
expression (Figure 4.3-6). The stimuli that regulate alternative IR splicing are
unknown, but are likely to be associated with the metabolic milieu observed in insulin
There are several caveats with using Hep G2 cells to model normal human liver cells.
Although Hep G2 cells are a well characterised liver cell line used by many
investigators to model insulin signalling,506, 769-775 they poorly reflect the behaviour of
differentiated liver tissue. Unlike normal human liver cells, which have 10 times more
IR-B expression than IR-A (IR-B:A ratio of 9.8), HepG2 cells have twice as much IRA than IR-B (IR-B:A ratio of 0.5, Figure 4.3-6). Hep G2 cells need
supraphysiological concentrations of insulin to stimulate molecules such as AKT,
possibly because of the much reduced IR-B:A ratio. Such high concentrations of
insulin will also activate IGF-I receptors and IGF-I/INSR hybrids thus potentially
confounding the conclusions made from our data. But the constant expression of IGFIR across all transfected cells should control for this factor. Further, because Hep G2
cells originated from a hepatocellular carcinoma, it is unknown whether they kept the
same mechanism by which insulin may regulate PCK1 expression. However, the
empty vector transfected Hep G2 cells had a significant increase in AKT
phosphorylation and inhibition of PCK1 transcription 5 minutes and 12 hours after
205
treatment with insulin respectively, which is consistent with previously reported
work.378, 390, 506 To improve the experiment, a time course of AKT phosphorylation at
lower insulin concentrations would have provided stronger evidence for the purported
harmful effects of increased IR-A expression. Another experimental limitation is the
lack of data relating to the protein levels of the two isoforms. Although an antibody to
a peptide corresponding to the exon 11 of IR-B isoform was produced, it was found to
be non-specific and could not be used to assay levels of IR-B protein. Nevertheless,
IR isoform mRNA levels have previously been shown to mirror the relative levels of
the two proteins on the cell surface.420, 424
All the caveats withstanding, the conclusions reached still have some impact on
insulin’s ability to regulate metabolic signalling through the two IR isoforms. A
transgenic mouse model, although technically challenging, would be useful to fully
understand the effect of increased IR-A expression on insulin’s ability to regulate
metabolic signalling and is a avenue for further work.
It is difficult to ascertain whether the altered liver IR-B:A ratio observed in type 2
diabetes is the cause or result of insulin resistance, but the data presented suggest that
it has a role in dysregulated glucose homeostasis. Although the proteins that regulate
alternative IR isoform splicing have been identified (CUG-BP1, SRp20 and
SF2/ASF),361 the molecule or molecules that regulate their activity are still unknown.
Further work should aim to elucidate the relationship between hyperinsulinaemia, the
spliceosome and metabolic homeostasis. The role of the spliceosome in regulating
variable signalling in different tissue and the consequences of this on metabolic
homeostasis is worthy of research in and of itself. It adds another dimension to the
pathogenesis of type 2 diabetes, a better understanding of which will lead to more
efficient treatments of this disease. Data presented in Chapter 4 reinvigorates the role
of IR isoforms in the pathogenesis of type 2 diabetes and suggests that aberrant
regulation of IR isoform in metabolically active tissue has adverse affects on insulin
signalling.
206
7.3 Changes in gene expression after remission of diabetes reveal
new roles for the liver in regulating metabolic homeostasis
From the data presented so far it is clear that using RYGB surgery with a study cohort
controlled for degree of insulin resistance is an excellent human model of type 2
diabetes. Nowhere is this more evident than in the microarray data presented in
Chapter 5, which shows that the majority of genes that were differentially regulated
after RYGB surgery are known to be implicated in the multiple pathological processes
involved in type 2 diabetes. Not surprisingly, expression of genes involved in
lipotoxicity, inflammation and ER stress were decreased after RYGB surgery,
particularly in those individuals who had remission of type 2 diabetes.
Although there were many significant observations, two of them are of particular
interest and present novel data that offer insights into the pathogenesis of type 2
diabetes. The apparent inter-organ communication between the liver and the pancreas
presents a hitherto unconfirmed relationship whereby the liver can not only mediate
pancreas β cell size, but can also regulate insulin secretion. IGFB5, which is made and
secreted by the liver, is increased after RYGB surgery and has been shown to induce
pancreatic cell growth.742 Similarly, expression of MAP2K1, which was shown to be
involved in a pathway that stimulates pancreatic cell growth via neuronal signals from
the liver,743 was also increased after remission of type 2 diabetes.
INHBE, a newly identified activin which is involved in liver-pancreas cross talk
presents as a promising candidate for treatments that prevent progression from
impaired glucose tolerance to type 2 diabetes. In this study, expression of liver
INHBE tended to be higher in individuals with type 2 diabetes and decreased after
RYGB surgery. Recent work on the functional characteristics of INHBE showed that
it is secreted by the liver in response to insulin stimulus and acts on the pancreas to
inhibit insulin secretion.571-573 Furthermore, overexpression of INHBE in mice
impaired growth of pancreatic exocrine cells.573 INHBE secretion by the liver appears
to be a natural feedback response to elevated levels of insulin and may contribute to
the gradual loss of beta cell function that eventually precipitates diabetes. To test this
hypothesis, we would measure INHBE protein concentrations from hepatic portal vein
blood to confirm its high concentrations in individual with insulin resistance.
207
Following this, we would treat human pancreatic cells with INHBE protein to test
what concentration of INHBE can induce β-cell failure. Finally, a transgenic animal
model of insulin resistance in which we can induce INHBE overexpression would
provide strong evidence for the role of INHBE in diabetes progression.
The microarray data also presented several candidate genes that are involved in
circadian like regulation of metabolic homeostasis. Although disruption of the
circadian rhythm is increasingly being associated with an increased risk of
development of insulin resistance or type 2 diabetes,753-757 the molecular mechanism
governing this link is yet to be indentified. Data presented in Chapter 5 identifies
BHLHE40, CCRN4L and ARNTL as candidate genes that may be directly involved in
the pathogenesis of type 2 diabetes. In particular, expression of BHLHE40 was only
significantly decreased in those individuals who had a remission of type 2 diabetes.
All three genes have critical roles in the molecular clock and transgenic animal
models have shown them to be critical in appropriate regulation of metabolic
homeostasis.606,
607, 611, 763, 767
Although the increasing association of the circadian
rhythm with type 2 diabetes is relatively recent, data presented here reveals three
genes that may be directly involved with insulin’s purported diurnal variation.758
However, the gene changes observed in the microarray data also suffer from the
causality paradigm in that it cannot be ascertained whether the variable gene
expression contributes to or is a result of type 2 diabetes. This is a common limitation
of microarray analysis which is exacerbated by the multitude of genes that have no
known link to type 2 diabetes or are yet to be fully characterised. For example,
LGALS4 and PYROXD2 are for the first time linked to type 2 diabetes in this study.
Although they may be associated with host defence and the gut microbiome, lack of
functional work in this thesis prevents us from drawing any further conclusions.
Microarray data can often be overwhelming, but a thorough literature search and
investigation of the significant gene changes has presented multiple avenues for future
research. The most promising of which would be full characterisation of INHBE’s
involvement in regulation of insulin secretion and its role in the progression to type 2
diabetes.
208
Further, this thesis demonstrates the polygenic nature of type 2 diabetes by the array
of gene changes involved in multiple pathological processes. Involvement of
lipotoxicity, inflammation and ER stress has been well documented, while the
importance of the gut microbiome and disruption of the circadian rhythm increased in
importance with data presented here. Investigation and functional characterisation of
any of the genes identified and how they interact in health and during progression to
type 2 diabetes will greatly aid our understanding of this disease. Using the RYGB
surgery as a model of type 2 diabetes and technologies such as exon junction arrays,
whole genome sequencing, methylation and miRNA analysis will also allow for a
complete picture of changes in gene expression, epigenetic and transcriptomic
regulation that will eventually lead to better understanding of type 2 diabetes.
7.4 Conclusion
Type 2 diabetes is the culmination of multiple pathological processes that results from
abnormal feeding, activity and circadian behaviour. To our knowledge, this is the first
study that has used the RYGB surgery as a human model of type 2 diabetes to conduct
a focused investigation on molecules previously implicated in the pathogenesis of said
disease. Novel findings around the role of liver ENPP1 and the IR isoforms in insulin
signalling have enhanced our understanding of the insulin signalling pathway and
through that, the pathogenesis of type 2 diabetes. In addition, the improvement of
liver insulin resistance that is associated with remission of type 2 diabetes allowed for
identification of novel relationships, the most promising of which is the involvement
of INHBE in type 2 diabetes progression. As with most research, this thesis provides
some answers but raises even more questions. Nonetheless, the evidence presented
here develops the field of type 2 diabetes research further and provides new avenues
for future studies.
209
210
REFRENCES
1.
Shaw JE, Sicree RA, Zimmet PZ. Global estimates of the prevalence of
diabetes for 2010 and 2030. Diabetes Research and Clinical Practice. 2010;87:4-14.
2.
Whiting DR, Guariguata L, Weil C, Shaw J. IDF diabetes atlas: global
estimates of the prevalence of diabetes for 2011 and 2030. Diabetes Research and
Clinical Practice. 2011;94:311-21.
3.
Alberti KG, Zimmet P, Shaw J. Metabolic syndrome--a new world-wide
definition. A Consensus Statement from the International Diabetes Federation.
Diabetic Medicine. 2006;23:469-80.
4.
Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et
al. Harmonizing the metabolic syndrome: a joint interim statement of the International
Diabetes Federation Task Force on Epidemiology and Prevention; National Heart,
Lung, and Blood Institute; American Heart Association; World Heart Federation;
International Atherosclerosis Society; and International Association for the Study of
Obesity. Circulation. 2009;120:1640-5.
5.
Shand BI, Scott RS, Lewis JG, Elder PA, Frampton CM. Comparison of
indices of insulin resistance with metabolic syndrome classifications to predict the
development of impaired fasting glucose in overweight and obese subjects: a 3-year
prospective study. Int J Obes (Lond). 2009;33:1274-9.
6.
DeFronzo RA. Current issues in the treatment of type 2 diabetes. Overview of
newer agents: where treatment is going. American Journal of Medicine.
2010;123:S38-48.
7.
Saydah SH, Fradkin J, Cowie CC. Poor control of risk factors for vascular
disease among adults with previously diagnosed diabetes. JAMA. 2004;291:335-42.
8.
Duckworth W, Abraira C, Moritz T, Reda D, Emanuele N, Reaven PD, et al.
Glucose control and vascular complications in veterans with type 2 diabetes. New
England Journal of Medicine. 2009;360:129-39.
9.
Gerstein HC, Miller ME, Byington RP, Goff DC, Jr., Bigger JT, Buse JB, et
al. Effects of intensive glucose lowering in type 2 diabetes. New England Journal of
Medicine. 2008;358:2545-59.
10.
Patel A, MacMahon S, Chalmers J, Neal B, Billot L, Woodward M, et al.
Intensive blood glucose control and vascular outcomes in patients with type 2
diabetes. New England Journal of Medicine. 2008;358:2560-72.
11.
Ministry of Health. Report on New Zealand Cost-of-Illness Studies on longterm conditions. Wellington: Ministry of Health. 2009.
211
12.
Pories W, Swanson M, MacDonald K, Long S, Morris B, Barakat H, et al.
Who whould have though it? An operation proves to be the most effective therapy for
adult-onset diabetes mellitus. Annals of Surgery. 1995;222:339-52.
13.
Buchwald H, Avidor Y, Braunwald E, Jensen MD, Pories W, Fahrbach K, et
al. Bariatric surgery: a systematic review and meta-analysis. JAMA. 2004;292:172437.
14.
Buchwald H, Estok R, Fahrbach K, Banel D, Jensen MD, Pories WJ, et al.
Weight and type 2 diabetes after bariatric surgery: systematic review and metaanalysis. American Journal of Medicine. 2009;122:248-56 e5.
15.
Dixon JB, Zimmet P, Alberti KG, Rubino F. Bariatric surgery: an IDF
statement for obese Type 2 diabetes. Diabetic Medicine. 2011;28:628-42.
16.
Reaven GM. Banting lecture 1988. Role of insulin resistance in human
disease. Diabetes. 1988;37:1595-607.
17.
Reaven GM. Pathophysiology of insulin resistance in human disease.
Physiological Reviews. 1995;75:473-86.
18.
Warram JH, Martin BC, Krolewski AS, Soeldner JS, Kahn CR. Slow glucose
removal rate and hyperinsulinemia precede the development of type II diabetes in the
offspring of diabetic parents. Annals of Internal Medicine. 1990;113:909-15.
19.
DeFronzo RA, Ferrannini E, Simonson DC. Fasting hyperglycemia in noninsulin-dependent diabetes mellitus: contributions of excessive hepatic glucose
production and impaired tissue glucose uptake. Metabolism: Clinical and
Experimental. 1989;38:387-95.
20.
Weyer C, Bogardus C, Pratley RE. Metabolic characteristics of individuals
with impaired fasting glucose and/or impaired glucose tolerance. Diabetes.
1999;48:2197-203.
21.
Bock G, Chittilapilly E, Basu R, Toffolo G, Cobelli C, Chandramouli V, et al.
Contribution of hepatic and extrahepatic insulin resistance to the pathogenesis of
impaired fasting glucose: role of increased rates of gluconeogenesis. Diabetes.
2007;56:1703-11.
22.
Mangiafico SP, Lim SH, Neoh S, Massinet H, Joannides CN, Proietto J, et al.
A primary defect in glucose production alone cannot induce glucose intolerance
without defects in insulin secretion. Journal of Endocrinology. 2011;210:335-47.
23.
Consoli A, Nurjhan N, Capani F, Gerich J. Predominant role of
gluconeogenesis in increased hepatic glucose production in NIDDM. Diabetes.
1989;38:550-7.
24.
Kahn SE, Prigeon RL, McCulloch DK, Boyko EJ, Bergman RN, Schwartz
MW, et al. Quantification of the relationship between insulin sensitivity and beta-cell
212
function in human subjects. Evidence for a hyperbolic function. Diabetes.
1993;42:1663-72.
25.
Buse JB, Polonsky KS, Burant CF. Type 2 Diabetes Mellitus. In: Melmed S,
Polonsky KS, Larsen PR, Kronenberg HM, editors. Williams Textbook of
Endocrinology. 12 ed. Philadelphia, PA, USA: Elsevier Saunders; 2011. p. 1371-435.
26.
Weyer C, Bogardus C, Mott DM, Pratley RE. The natural history of insulin
secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes
mellitus. Journal of Clinical Investigation. 1999;104:787-94.
27.
Abdul-Ghani MA, Jenkinson CP, Richardson DK, Tripathy D, DeFronzo RA.
Insulin secretion and action in subjects with impaired fasting glucose and impaired
glucose tolerance: results from the Veterans Administration Genetic Epidemiology
Study. Diabetes. 2006;55:1430-5.
28.
Reaven GM, Hollenbeck CB, Chen YD. Relationship between glucose
tolerance, insulin secretion, and insulin action in non-obese individuals with varying
degrees of glucose tolerance. Diabetologia. 1989;32:52-5.
29.
Porksen N, Munn S, Steers J, Vore S, Veldhuis J, Butler P. Pulsatile insulin
secretion accounts for 70% of total insulin secretion during fasting. American Journal
of Physiology. 1995;269:E478-88.
30.
Porksen N, Nyholm B, Veldhuis JD, Butler PC, Schmitz O. In humans at least
75% of insulin secretion arises from punctuated insulin secretory bursts. American
Journal of Physiology. 1997;273:E908-14.
31.
Matthews DR, Naylor BA, Jones RG, Ward GM, Turner RC. Pulsatile insulin
has greater hypoglycemic effect than continuous delivery. Diabetes. 1983;32:617-21.
32.
Paolisso G, Scheen AJ, Giugliano D, Sgambato S, Albert A, Varricchio M, et
al. Pulsatile insulin delivery has greater metabolic effects than continuous hormone
administration in man: importance of pulse frequency. Journal of Clinical
Endocrinology and Metabolism. 1991;72:607-15.
33.
O'Rahilly S, Turner RC, Matthews DR. Impaired pulsatile secretion of insulin
in relatives of patients with non-insulin-dependent diabetes. New England Journal of
Medicine. 1988;318:1225-30.
34.
Kjems LL, Kirby BM, Welsh EM, Veldhuis JD, Straume M, McIntyre SS, et
al. Decrease in beta-cell mass leads to impaired pulsatile insulin secretion, reduced
postprandial hepatic insulin clearance, and relative hyperglucagonemia in the minipig.
Diabetes. 2001;50:2001-12.
35.
Laedtke T, Kjems L, Porksen N, Schmitz O, Veldhuis J, Kao PC, et al.
Overnight inhibition of insulin secretion restores pulsatility and proinsulin/insulin
ratio in type 2 diabetes. Am J Physiol Endocrinol Metab. 2000;279:E520-8.
213
36.
Lang DA, Matthews DR, Burnett M, Turner RC. Brief, irregular oscillations
of basal plasma insulin and glucose concentrations in diabetic man. Diabetes.
1981;30:435-9.
37.
Polonsky KS, Given BD, Van Cauter E. Twenty-four-hour profiles and
pulsatile patterns of insulin secretion in normal and obese subjects. Journal of
Clinical Investigation. 1988;81:442-8.
38.
Curry DL, Bennett LL, Grodsky GM. Dynamics of insulin secretion by the
perfused rat pancreas. Endocrinology. 1968;83:572-84.
39.
Getty L, Hamilton-Wessler M, Ader M, Dea MK, Bergman RN. Biphasic
insulin secretion during intravenous glucose tolerance test promotes optimal
interstitial insulin profile. Diabetes. 1998;47:1941-7.
40.
Pratley RE, Weyer C. The role of impaired early insulin secretion in the
pathogenesis of Type II diabetes mellitus. Diabetologia. 2001;44:929-45.
41.
Rorsman P, Renstrom E. Insulin granule dynamics in pancreatic beta cells.
Diabetologia. 2003;46:1029-45.
42.
Daniel S, Noda M, Straub SG, Sharp GW. Identification of the docked granule
pool responsible for the first phase of glucose-stimulated insulin secretion. Diabetes.
1999;48:1686-90.
43.
Henquin JC. Regulation of insulin secretion: a matter of phase control and
amplitude modulation. Diabetologia. 2009;52:739-51.
44.
Eliasson L, Abdulkader F, Braun M, Galvanovskis J, Hoppa MB, Rorsman P.
Novel aspects of the molecular mechanisms controlling insulin secretion. J Physiol.
2008;586:3313-24.
45.
Luzi L, DeFronzo RA. Effect of loss of first-phase insulin secretion on hepatic
glucose production and tissue glucose disposal in humans. American Journal of
Physiology. 1989;257:E241-6.
46.
Elahi D, Meneilly GS, Minaker KL, Andersen DK, Rowe JW. Escape of
hepatic glucose production during hyperglycemic clamp. American Journal of
Physiology. 1989;257:E704-11.
47.
Steiner KE, Mouton SM, Bowles CR, Williams PE, Cherrington AD. The
relative importance of first- and second-phase insulin secretion in countering the
action of glucagon on glucose turnover in the conscious dog. Diabetes. 1982;31:96472.
48.
Brunzell JD, Robertson RP, Lerner RL, Hazzard WR, Ensinck JW, Bierman
EL, et al. Relationships between fasting plasma glucose levels and insulin secretion
during intravenous glucose tolerance tests. Journal of Clinical Endocrinology and
Metabolism. 1976;42:222-9.
214
49.
Calles-Escandon J, Robbins DC. Loss of early phase of insulin release in
humans impairs glucose tolerance and blunts thermic effect of glucose. Diabetes.
1987;36:1167-72.
50.
Davies MJ, Rayman G, Grenfell A, Gray IP, Day JL, Hales CN. Loss of the
first phase insulin response to intravenous glucose in subjects with persistent impaired
glucose tolerance. Diabetic Medicine. 1994;11:432-6.
51.
Mitrakou A, Kelley D, Mokan M, Veneman T, Pangburn T, Reilly J, et al.
Role of reduced suppression of glucose production and diminished early insulin
release in impaired glucose tolerance. New England Journal of Medicine.
1992;326:22-9.
52.
Gerich JE. Is reduced first-phase insulin release the earliest detectable
abnormality in individuals destined to develop type 2 diabetes? Diabetes. 2002;51
Suppl 1:S117-21.
53.
Del Prato S, Tiengo A. The importance of first-phase insulin secretion:
implications for the therapy of type 2 diabetes mellitus. Diabetes/Metabolism
Research and Reviews. 2001;17:164-74.
54.
Steiner KE, Mouton SM, Williams PE, Lacy WW, Cherrington AD. Relative
importance of first- and second-phase insulin secretion in glucose homeostasis in
conscious dog. II. Effects on gluconeogenesis. Diabetes. 1986;35:776-84.
55.
Song SH, McIntyre SS, Shah H, Veldhuis JD, Hayes PC, Butler PC. Direct
measurement of pulsatile insulin secretion from the portal vein in human subjects.
Journal of Clinical Endocrinology and Metabolism. 2000;85:4491-9.
56.
Meier JJ, Veldhuis JD, Butler PC. Pulsatile insulin secretion dictates systemic
insulin delivery by regulating hepatic insulin extraction in humans. Diabetes.
2005;54:1649-56.
57.
Bonora E, Zavaroni I, Coscelli C, Butturini U. Decreased hepatic insulin
extraction in subjects with mild glucose intolerance. Metabolism: Clinical and
Experimental. 1983;32:438-46.
58.
Sando H, Lee YS, Iwamoto Y, Ikeuchi M, Kosaka K. Isoproterenol-stimulated
C-peptide and insulin secretion in diabetic and nonobese normal subjects: decreased
hepatic extraction of endogenous insulin in diabetes. Journal of Clinical
59.
Jones CN, Pei D, Staris P, Polonsky KS, Chen YD, Reaven GM. Alterations in
the glucose-stimulated insulin secretory dose-response curve and in insulin clearance
in nondiabetic insulin-resistant individuals. Journal of Clinical Endocrinology and
Metabolism. 1997;82:1834-8.
60.
Bonora E, Zavaroni I, Manicardi V, Coscelli C, Butturini U. Further evidence
that insulin metabolism is a major determinant of peripheral insulin response to oral
215
glucose in subjects with mild glucose intolerance. Journal of Endocrinological
Investigation. 1986;9:371-4.
61.
Rudovich NN, Rochlitz HJ, Pfeiffer AF. Reduced hepatic insulin extraction in
response to gastric inhibitory polypeptide compensates for reduced insulin secretion
in normal-weight and normal glucose tolerant first-degree relatives of type 2 diabetic
patients. Diabetes. 2004;53:2359-65.
62.
Hovorka R, Powrie JK, Smith GD, Sonksen PH, Carson ER, Jones RH. Fivecompartment model of insulin kinetics and its use to investigate action of chloroquine
in NIDDM. American Journal of Physiology. 1993;265:E162-75.
63.
Duckworth WC, Bennett RG, Hamel FG. Insulin degradation: progress and
potential. Endocrine Reviews. 1998;19:608-24.
64.
DeAngelis AM, Heinrich G, Dai T, Bowman TA, Patel PR, Lee SJ, et al.
Carcinoembryonic antigen-related cell adhesion molecule 1: a link between insulin
and lipid metabolism. Diabetes. 2008;57:2296-303.
65.
Poy MN, Yang Y, Rezaei K, Fernstrom MA, Lee AD, Kido Y, et al.
CEACAM1 regulates insulin clearance in liver. Nature Genetics. 2002;30:270-6.
66.
Najjar SM. Regulation of insulin action by CEACAM1. Trends Endocrinol
Metab. 2002;13:240-5.
67.
Luo W, Wood CG, Earley K, Hung MC, Lin SH. Suppression of
tumorigenicity of breast cancer cells by an epithelial cell adhesion molecule (CCAM1): the adhesion and growth suppression are mediated by different domains.
Oncogene. 1997;14:1697-704.
68.
Soni P, Lakkis M, Poy MN, Fernstrom MA, Najjar SM. The differential
effects of pp120 (Ceacam 1) on the mitogenic action of insulin and insulin-like
growth factor 1 are regulated by the nonconserved tyrosine 1316 in the insulin
receptor. Molecular and Cellular Biology. 2000;20:3896-905.
69.
Formisano P, Najjar SM, Gross CN, Philippe N, Oriente F, Kern-Buell CL, et
al. Receptor-mediated internalization of insulin. Potential role of pp120/HA4, a
substrate of the insulin receptor kinase. Journal of Biological Chemistry.
1995;270:24073-7.
70.
Park SY, Cho YR, Kim HJ, Hong EG, Higashimori T, Lee SJ, et al.
Mechanism of glucose intolerance in mice with dominant negative mutation of
CEACAM1. Am J Physiol Endocrinol Metab. 2006;291:E517-24.
71.
Leissring MA, Selkoe DJ. Structural biology: enzyme target to latch on to.
Nature. 2006;443:761-2.
72.
Fakhrai-Rad H, Nikoshkov A, Kamel A, Fernstrom M, Zierath JR, Norgren S,
et al. Insulin-degrading enzyme identified as a candidate diabetes susceptibility gene
in GK rats. Human Molecular Genetics. 2000;9:2149-58.
216
73.
Abdul-Hay SO, Kang D, McBride M, Li L, Zhao J, Leissring MA. Deletion of
insulin-degrading enzyme elicits antipodal, age-dependent effects on glucose and
insulin tolerance. PLoS One. 2011;6:e20818.
74.
Rudovich N, Pivovarova O, Fisher E, Fischer-Rosinsky A, Spranger J, Mohlig
M, et al. Polymorphisms within insulin-degrading enzyme (IDE) gene determine
insulin metabolism and risk of type 2 diabetes. J Mol Med (Berl). 2009;87:1145-51.
75.
DeFronzo RA. Lilly lecture 1987. The triumvirate: beta-cell, muscle, liver. A
collusion responsible for NIDDM. Diabetes. 1988;37:667-87.
76.
Defronzo RA. Banting Lecture. From the triumvirate to the ominous octet: a
new paradigm for the treatment of type 2 diabetes mellitus. Diabetes. 2009;58:773-95.
77.
Konrad D, Rudich A, Klip A. Insulin-mediated regulation of glucose
metabolism. In: Kumar S, O'Rahilly S, editors. Insulin Resistance: Insulin Action and
its disturbances in disease. Chichester: John Wiley and Sons; 2005. p. 63-85.
78.
Insulin Resistance: Insulin Action and Its Disturbances in Disease. Kumar S,
O'Rahilly S, editors2005.
79.
Lewis GF, Zinman B, Groenewoud Y, Vranic M, Giacca A. Hepatic glucose
production is regulated both by direct hepatic and extrahepatic effects of insulin in
humans. Diabetes. 1996;45:454-62.
80.
Rebrin K, Steil GM, Getty L, Bergman RN. Free fatty acid as a link in the
regulation of hepatic glucose output by peripheral insulin. Diabetes. 1995;44:1038-45.
81.
Plum L, Belgardt BF, Bruning JC. Central insulin action in energy and glucose
homeostasis. Journal of Clinical Investigation. 2006;116:1761-6.
82.
Schwartz MW, Woods SC, Porte D, Jr., Seeley RJ, Baskin DG. Central
nervous system control of food intake. Nature. 2000;404:661-71.
83.
Matsuda M, Liu Y, Mahankali S, Pu Y, Mahankali A, Wang J, et al. Altered
hypothalamic function in response to glucose ingestion in obese humans. Diabetes.
1999;48:1801-6.
84.
Clark MG, Wallis MG, Barrett EJ, Vincent MA, Richards SM, Clerk LH, et al.
Blood flow and muscle metabolism: a focus on insulin action. Am J Physiol
Endocrinol Metab. 2003;284:E241-58.
85.
Leibiger B, Leibiger IB, Moede T, Kemper S, Kulkarni RN, Kahn CR, et al.
Selective insulin signaling through A and B insulin receptors regulates transcription
of insulin and glucokinase genes in pancreatic beta cells. Molecular Cell. 2001;7:55970.
86.
Wallace TM, Matthews DR. The assessment of insulin resistance in man.
Diabetic Medicine. 2002;19:527-34.
217
87.
Bergman RN, Finegood DT, Ader M. Assessment of insulin sensitivity in
vivo. Endocrine Reviews. 1985;6:45-86.
88.
Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC.
Homeostasis model assessment: insulin resistance and beta-cell function from fasting
plasma glucose and insulin concentrations in man. Diabetologia. 1985;28:412-9.
89.
McAuley KA, Williams SM, Mann JI, Walker RJ, Lewis-Barned NJ, Temple
LA, et al. Diagnosing insulin resistance in the general population. Diabetes Care.
2001;24:460-4.
90.
Braatvedt G, Gamble G, Kyle C. Metabolic characteristics of patients with
apparently normal
fasting plasma glucose. The New Zealand Medical Journal. 2006;119.
91.
Geloneze B, Vasques AC, Stabe CF, Pareja JC, Rosado LE, Queiroz EC, et al.
HOMA1-IR and HOMA2-IR indexes in identifying insulin resistance and metabolic
syndrome: Brazilian Metabolic Syndrome Study (BRAMS). Arquivos Brasileiros de
Endocrinologia e Metabologia. 2009;53:281-7.
92.
Davis SN, Monti L, Piatti PM, Moller N, Ng L, Coppack S, et al. Estimates of
insulin action in normal, obese and NIDDM man: comparison of insulin and glucose
infusion test, CIGMA, minimal model and glucose clamp techniques. Diabetes
Research. 1993;23:1-18.
93.
Bonora E, Targher G, Alberiche M, Bonadonna RC, Saggiani F, Zenere MB,
et al. Homeostasis model assessment closely mirrors the glucose clamp technique in
the assessment of insulin sensitivity: studies in subjects with various degrees of
glucose tolerance and insulin sensitivity. Diabetes Care. 2000;23:57-63.
94.
Beck-Nielsen H, Alford F, Hother-Nielsen O. Insulin Resistance in Glucose
Disposal and Production in Man with Specific Reference to Metabolic Syndrome and
Type 2 Diabetes. In: Kumar S, O'Rahilly S, editors. Insulin Resistance Insulin action
and its disturbances in disease. West Sussex: Wiley; 2005. p. 155-79.
95.
DeFronzo RA, Gunnarsson R, Bjorkman O, Olsson M, Wahren J. Effects of
insulin on peripheral and splanchnic glucose metabolism in noninsulin-dependent
(type II) diabetes mellitus. Journal of Clinical Investigation. 1985;76:149-55.
96.
Pendergrass M, Bertoldo A, Bonadonna R, Nucci G, Mandarino L, Cobelli C,
et al. Muscle glucose transport and phosphorylation in type 2 diabetic, obese
nondiabetic, and genetically predisposed individuals. Am J Physiol Endocrinol Metab.
2007;292:E92-100.
97.
Vaag A, Alford F, Henriksen FL, Christopher M, Beck-Nielsen H. Multiple
defects of both hepatic and peripheral intracellular glucose processing contribute to
the hyperglycaemia of NIDDM. Diabetologia. 1995;38:326-36.
218
98.
Groop LC, Bonadonna RC, DelPrato S, Ratheiser K, Zyck K, Ferrannini E, et
al. Glucose and free fatty acid metabolism in non-insulin-dependent diabetes mellitus.
Evidence for multiple sites of insulin resistance. Journal of Clinical Investigation.
1989;84:205-13.
99.
Groop LC, Bonadonna RC, Simonson DC, Petrides AS, Shank M, DeFronzo
RA. Effect of insulin on oxidative and nonoxidative pathways of free fatty acid
metabolism in human obesity. American Journal of Physiology. 1992;263:E79-84.
100. Rebrin K, Steil GM, Mittelman SD, Bergman RN. Causal linkage between
insulin suppression of lipolysis and suppression of liver glucose output in dogs.
Journal of Clinical Investigation. 1996;98:741-9.
101. Williamson JR, Kreisberg RA, Felts PW. Mechanism for the stimulation of
gluconeogenesis by fatty acids in perfused rat liver. Proceedings of the National
Academy of Sciences of the United States of America. 1966;56:247-54.
102. Bergman RN. New concepts in extracellular signaling for insulin action: the
single gateway hypothesis. Recent Progress in Hormone Research. 1997;52:359-85;
discussion 85-7.
103. Cherrington AD. Banting Lecture 1997. Control of glucose uptake and release
by the liver in vivo. Diabetes. 1999;48:1198-214.
104. Perriello G, Pampanelli S, Del Sindaco P, Lalli C, Ciofetta M, Volpi E, et al.
Evidence of increased systemic glucose production and gluconeogenesis in an early
stage of NIDDM. Diabetes. 1997;46:1010-6.
105. Magnusson I, Rothman DL, Katz LD, Shulman RG, Shulman GI. Increased
rate of gluconeogenesis in type II diabetes mellitus. A 13C nuclear magnetic
resonance study. Journal of Clinical Investigation. 1992;90:1323-7.
106. Lewis GF, Carpentier A, Vranic M, Giacca A. Resistance to insulin's acute
direct hepatic effect in suppressing steady-state glucose production in individuals with
type 2 diabetes. Diabetes. 1999;48:570-6.
107. Gastaldelli A, Toschi E, Pettiti M, Frascerra S, Quinones-Galvan A, Sironi
AM, et al. Effect of physiological hyperinsulinemia on gluconeogenesis in
nondiabetic subjects and in type 2 diabetic patients. Diabetes. 2001;50:1807-12.
108. Krauss RM. Lipids and lipoproteins in patients with type 2 diabetes. Diabetes
Care. 2004;27:1496-504.
109. Adeli K, Taghibiglou C, Van Iderstine SC, Lewis GF. Mechanisms of hepatic
very low-density lipoprotein overproduction in insulin resistance. Trends in
Cardiovascular Medicine. 2001;11:170-6.
110. Lewis GF, Carpentier A, Adeli K, Giacca A. Disordered fat storage and
mobilization in the pathogenesis of insulin resistance and type 2 diabetes. Endocrine
Reviews. 2002;23:201-29.
219
111. Biddinger SB, Kahn CR. From mice to men: insights into the insulin
resistance syndromes. Annual Review of Physiology. 2006;68:123-58.
112. Bluher M, Michael MD, Peroni OD, Ueki K, Carter N, Kahn BB, et al.
Adipose tissue selective insulin receptor knockout protects against obesity and
obesity-related glucose intolerance. Dev Cell. 2002;3:25-38.
113. Bruning JC, Gautam D, Burks DJ, Gillette J, Schubert M, Orban PC, et al.
Role of brain insulin receptor in control of body weight and reproduction. Science.
2000;289:2122-5.
114. Bruning JC, Michael MD, Winnay JN, Hayashi T, Horsch D, Accili D, et al. A
muscle-specific insulin receptor knockout exhibits features of the metabolic syndrome
of NIDDM without altering glucose tolerance. Molecular Cell. 1998;2:559-69.
115. Kulkarni RN, Bruning JC, Winnay JN, Postic C, Magnuson MA, Kahn CR.
Tissue-specific knockout of the insulin receptor in pancreatic beta cells creates an
insulin secretory defect similar to that in type 2 diabetes. Cell. 1999;96:329-39.
116. Michael MD, Kulkarni RN, Postic C, Previs SF, Shulman GI, Magnuson MA,
et al. Loss of insulin signaling in hepatocytes leads to severe insulin resistance and
progressive hepatic dysfunction. Molecular Cell. 2000;6:87-97.
117. Gu H, Marth JD, Orban PC, Mossmann H, Rajewsky K. Deletion of a DNA
polymerase beta gene segment in T cells using cell type-specific gene targeting.
Science. 1994;265:103-6.
118. Obici S, Feng Z, Karkanias G, Baskin DG, Rossetti L. Decreasing
hypothalamic insulin receptors causes hyperphagia and insulin resistance in rats.
Nature Neuroscience. 2002;5:566-72.
119. Lauro D, Kido Y, Castle AL, Zarnowski MJ, Hayashi H, Ebina Y, et al.
Impaired glucose tolerance in mice with a targeted impairment of insulin action in
muscle and adipose tissue. Nature Genetics. 1998;20:294-8.
120. Mauvais-Jarvis F, Virkamaki A, Michael MD, Winnay JN, Zisman A,
Kulkarni RN, et al. A model to explore the interaction between muscle insulin
resistance and beta-cell dysfunction in the development of type 2 diabetes. Diabetes.
2000;49:2126-34.
121. Accili D, Drago J, Lee EJ, Johnson MD, Cool MH, Salvatore P, et al. Early
neonatal death in mice homozygous for a null allele of the insulin receptor gene.
Nature Genetics. 1996;12:106-9.
122. Okamoto H, Nakae J, Kitamura T, Park BC, Dragatsis I, Accili D. Transgenic
rescue of insulin receptor-deficient mice. Journal of Clinical Investigation.
2004;114:214-23.
220
123. Escribano O, Guillen C, Nevado C, Gomez-Hernandez A, Kahn CR, Benito
M. Beta-Cell hyperplasia induced by hepatic insulin resistance: role of a liverpancreas endocrine axis through insulin receptor A isoform. Diabetes. 2009;58:820-8.
124. McGarry JD. What if Minkowski had been ageusic? An alternative angle on
diabetes. Science. 1992;258:766-70.
125. Kahn CR, Flier JS, Bar RS, Archer JA, Gorden P, Martin MM, et al. The
syndromes of insulin resistance and acanthosis nigricans. Insulin-receptor disorders in
man. New England Journal of Medicine. 1976;294:739-45.
126. Dumont JE, Dremier S, Pirson I, Maenhaut C. Cross signaling, cell specificity,
and physiology. Am J Physiol Cell Physiol. 2002;283:C2-28.
127. Taniguchi CM, Emanuelli B, Kahn CR. Critical nodes in signalling pathways:
insights into insulin action. Nat Rev Mol Cell Biol. 2006;7:85-96.
128. Ullrich A, Schlessinger J. Signal transduction by receptors with tyrosine
kinase activity. Cell. 1990;61:203-12.
129. Ebina Y, Ellis L, Jarnagin K, Edery M, Graf L, Clauser E, et al. The human
insulin receptor cDNA: the structural basis for hormone-activated transmembrane
signalling. Cell. 1985;40:747-58.
130. Ullrich A, Bell JR, Chen EY, Herrera R, Petruzzelli LM, Dull TJ, et al.
Human insulin receptor and its relationship to the tyrosine kinase family of
oncogenes. Nature. 1985;313:756-61.
131. Dupont J, LeRoith D. Insulin and insulin-like growth factor I receptors:
similarities and differences in signal transduction. Hormone Research. 2001;55 Suppl
2:22-6.
132. Kahn CR, White MF. The insulin receptor and the molecular mechanism of
insulin action. Journal of Clinical Investigation. 1988;82:1151-6.
133. Roth RA, Cassell DJ. Insulin receptor: evidence that it is a protein kinase.
Science. 1983;219:299-301.
134. Kasuga M, Karlsson FA, Kahn CR. Insulin stimulates the phosphorylation of
the 95,000-dalton subunit of its own receptor. Science. 1982;215:185-7.
135. Cai D, Dhe-Paganon S, Melendez PA, Lee J, Shoelson SE. Two new
substrates in insulin signaling, IRS5/DOK4 and IRS6/DOK5. Journal of Biological
Chemistry. 2003;278:25323-30.
136. Fantin VR, Lavan BE, Wang Q, Jenkins NA, Gilbert DJ, Copeland NG, et al.
Cloning, tissue expression, and chromosomal location of the mouse insulin receptor
substrate 4 gene. Endocrinology. 1999;140:1329-37.
221
137. Lavan BE, Lane WS, Lienhard GE. The 60-kDa phosphotyrosine protein in
insulin-treated adipocytes is a new member of the insulin receptor substrate family.
Journal of Biological Chemistry. 1997;272:11439-43.
138. Sun XJ, Rothenberg P, Kahn CR, Backer JM, Araki E, Wilden PA, et al.
Structure of the insulin receptor substrate IRS-1 defines a unique signal transduction
protein. Nature. 1991;352:73-7.
139. Sun XJ, Wang LM, Zhang Y, Yenush L, Myers MG, Jr., Glasheen E, et al.
Role of IRS-2 in insulin and cytokine signalling. Nature. 1995;377:173-7.
140. White MF. The IRS-signalling system: a network of docking proteins that
mediate insulin action. Molecular and Cellular Biochemistry. 1998;182:3-11.
141. Gustafson TA, He W, Craparo A, Schaub CD, O'Neill TJ. Phosphotyrosinedependent interaction of SHC and insulin receptor substrate 1 with the NPEY motif of
the insulin receptor via a novel non-SH2 domain. Molecular and Cellular Biology.
1995;15:2500-8.
142. Baumann CA, Ribon V, Kanzaki M, Thurmond DC, Mora S, Shigematsu S, et
al. CAP defines a second signalling pathway required for insulin-stimulated glucose
transport. Nature. 2000;407:202-7.
143. Wick MJ, Dong LQ, Hu D, Langlais P, Liu F. Insulin receptor-mediated
p62dok tyrosine phosphorylation at residues 362 and 398 plays distinct roles for
binding GTPase-activating protein and Nck and is essential for inhibiting insulinstimulated activation of Ras and Akt. Journal of Biological Chemistry.
2001;276:42843-50.
144. Lehr S, Kotzka J, Herkner A, Sikmann A, Meyer HE, Krone W, et al.
Identification of major tyrosine phosphorylation sites in the human insulin receptor
substrate Gab-1 by insulin receptor kinase in vitro. Biochemistry. 2000;39:10898-907.
145. Giovannone B, Scaldaferri ML, Federici M, Porzio O, Lauro D, Fusco A, et al.
Insulin receptor substrate (IRS) transduction system: distinct and overlapping
signaling potential. Diabetes/Metabolism Research and Reviews. 2000;16:434-41.
146. Sciacchitano S, Taylor SI. Cloning, tissue expression, and chromosomal
localization of the mouse IRS-3 gene. Endocrinology. 1997;138:4931-40.
147. Tamemoto H, Kadowaki T, Tobe K, Yagi T, Sakura H, Hayakawa T, et al.
Insulin resistance and growth retardation in mice lacking insulin receptor substrate-1.
Nature. 1994;372:182-6.
148. Araki E, Lipes MA, Patti ME, Bruning JC, Haag B, 3rd, Johnson RS, et al.
Alternative pathway of insulin signalling in mice with targeted disruption of the IRS-1
gene. Nature. 1994;372:186-90.
149. Kubota N, Tobe K, Terauchi Y, Eto K, Yamauchi T, Suzuki R, et al.
Disruption of insulin receptor substrate 2 causes type 2 diabetes because of liver
222
insulin resistance and lack of compensatory beta-cell hyperplasia. Diabetes.
2000;49:1880-9.
150. Withers DJ, Gutierrez JS, Towery H, Burks DJ, Ren JM, Previs S, et al.
Disruption of IRS-2 causes type 2 diabetes in mice. Nature. 1998;391:900-4.
151. Liu SC, Wang Q, Lienhard GE, Keller SR. Insulin receptor substrate 3 is not
essential for growth or glucose homeostasis. Journal of Biological Chemistry.
1999;274:18093-9.
152. Fantin VR, Wang Q, Lienhard GE, Keller SR. Mice lacking insulin receptor
substrate 4 exhibit mild defects in growth, reproduction, and glucose homeostasis. Am
J Physiol Endocrinol Metab. 2000;278:E127-33.
153. Caro JF, Amatruda JM. Insulin receptors in hepatocytes: postreceptor events
mediate down regulation. Science. 1980;210:1029-31.
154. Gavin JR, 3rd, Roth J, Neville DM, Jr., de Meyts P, Buell DN. Insulindependent regulation of insulin receptor concentrations: a direct demonstration in cell
culture. Proceedings of the National Academy of Sciences of the United States of
America. 1974;71:84-8.
155. Knutson VP, Ronnett GV, Lane MD. Control of insulin receptor level in 3T3
cells: effect of insulin-induced down-regulation and dexamethasone-induced upregulation on rate of receptor inactivation. Proceedings of the National Academy of
Sciences of the United States of America. 1982;79:2822-6.
156. Okabayashi Y, Maddux BA, McDonald AR, Logsdon CD, Williams JA,
Goldfine ID. Mechanisms of insulin-induced insulin-receptor downregulation.
Decrease of receptor biosynthesis and mRNA levels. Diabetes. 1989;38:182-7.
157. Ramos FJ, Langlais PR, Hu D, Dong LQ, Liu F. Grb10 mediates insulinstimulated degradation of the insulin receptor: a mechanism of negative regulation.
Am J Physiol Endocrinol Metab. 2006;290:E1262-6.
158. Youngren JF. Regulation of insulin receptor function. Cellular and Molecular
Life Sciences. 2007;64:873-91.
159. Friedman JE, Ishizuka T, Liu S, Farrell CJ, Bedol D, Koletsky RJ, et al.
Reduced insulin receptor signaling in the obese spontaneously hypertensive Koletsky
rat. American Journal of Physiology. 1997;273:E1014-23.
160. Tsou RC, Bence KK. Central regulation of metabolism by protein tyrosine
phosphatases. Front Neurosci. 2012;6:192.
161. Drake PG, Posner BI. Insulin receptor-associated protein tyrosine
phosphatase(s): role in insulin action. Molecular and Cellular Biochemistry.
1998;182:79-89.
223
162. Delibegovic M, Zimmer D, Kauffman C, Rak K, Hong EG, Cho YR, et al.
Liver-specific deletion of protein-tyrosine phosphatase 1B (PTP1B) improves
metabolic syndrome and attenuates diet-induced endoplasmic reticulum stress.
Diabetes. 2009;58:590-9.
163. Bence KK. Hepatic PTP1B Deficiency: The Promise of a Treatment for
Metabolic Syndrome? J Clin Metab Diabetes. 2010;1:27-33.
164. Ueki K, Kondo T, Kahn CR. Suppressor of cytokine signaling 1 (SOCS-1) and
SOCS-3 cause insulin resistance through inhibition of tyrosine phosphorylation of
insulin receptor substrate proteins by discrete mechanisms. Molecular and Cellular
Biology. 2004;24:5434-46.
165. Emanuelli B, Peraldi P, Filloux C, Chavey C, Freidinger K, Hilton DJ, et al.
SOCS-3 inhibits insulin signaling and is up-regulated in response to tumor necrosis
factor-alpha in the adipose tissue of obese mice. Journal of Biological Chemistry.
2001;276:47944-9.
166. Maddux BA, Goldfine ID. Membrane glycoprotein PC-1 inhibition of insulin
receptor function occurs via direct interaction with the receptor alpha-subunit.
Diabetes. 2000;49:13-9.
167. Frittitta L, Youngren J, Vigneri R, Maddux BA, Trischitta V, Goldfine ID.
PC-1 content in skeletal muscle of non-obese, non-diabetic subjects: relationship to
insulin receptor tyrosine kinase and whole body insulin sensitivity. Diabetologia.
1996;39:1190-5.
168. Frittitta L, Youngren JF, Sbraccia P, D'Adamo M, Buongiorno A, Vigneri R,
et al. Increased adipose tissue PC-1 protein content, but not tumour necrosis factoralpha gene expression, is associated with a reduction of both whole body insulin
sensitivity and insulin receptor tyrosine-kinase activity. Diabetologia. 1997;40:282-9.
169. Frittitta L, Spampinato D, Solini A, Nosadini R, Goldfine ID, Vigneri R, et al.
Elevated PC-1 content in cultured skin fibroblasts correlates with decreased in vivo
and in vitro insulin action in nondiabetic subjects: evidence that PC-1 may be an
intrinsic factor in impaired insulin receptor signaling. Diabetes. 1998;47:1095-100.
170. Maddux BA, Sbraccia P, Kumakura S, Sasson S, Youngren J, Fisher A, et al.
Membrane glycoprotein PC-1 and insulin resistance in non-insulin-dependent diabetes
mellitus. Nature. 1995;373:448-51.
171. Goldfine ID, Maddux BA, Youngren JF, Reaven G, Accili D, Trischitta V, et
al. The role of membrane glycoprotein plasma cell antigen 1/ectonucleotide
pyrophosphatase phosphodiesterase 1 in the pathogenesis of insulin resistance and
related abnormalities. Endocrine Reviews. 2008;29:62-75.
172. Dong H, Maddux BA, Altomonte J, Meseck M, Accili D, Terkeltaub R, et al.
Increased hepatic levels of the insulin receptor inhibitor, PC-1/NPP1, induce insulin
resistance and glucose intolerance. Diabetes. 2005;54:367-72.
224
173. Zhou HH, Chin CN, Wu M, Ni W, Quan S, Liu F, et al. Suppression of PC1/ENPP-1 expression improves insulin sensitivity in vitro and in vivo. European
Journal of Pharmacology. 2009;616:346-52.
174. Pizzuti A, Frittitta L, Argiolas A, Baratta R, Goldfine ID, Bozzali M, et al. A
polymorphism (K121Q) of the human glycoprotein PC-1 gene coding region is
strongly associated with insulin resistance. Diabetes. 1999;48:1881-4.
175. Costanzo BV, Trischitta V, Di Paola R, Spampinato D, Pizzuti A, Vigneri R,
et al. The Q allele variant (GLN121) of membrane glycoprotein PC-1 interacts with
the insulin receptor and inhibits insulin signaling more effectively than the common K
allele variant (LYS121). Diabetes. 2001;50:831-6.
176. Shepherd PR, Withers DJ, Siddle K. Phosphoinositide 3-kinase: the key
switch mechanism in insulin signalling. Biochemical Journal. 1998;333 ( Pt 3):47190.
177. Escobedo JA, Navankasattusas S, Kavanaugh WM, Milfay D, Fried VA,
Williams LT. cDNA cloning of a novel 85 kd protein that has SH2 domains and
regulates binding of PI3-kinase to the PDGF beta-receptor. Cell. 1991;65:75-82.
178. Hiles ID, Otsu M, Volinia S, Fry MJ, Gout I, Dhand R, et al.
Phosphatidylinositol 3-kinase: structure and expression of the 110 kd catalytic
subunit. Cell. 1992;70:419-29.
179. Myers MG, Jr., Backer JM, Sun XJ, Shoelson S, Hu P, Schlessinger J, et al.
IRS-1 activates phosphatidylinositol 3'-kinase by associating with src homology 2
domains of p85. Proceedings of the National Academy of Sciences of the United
States of America. 1992;89:10350-4.
180. Carpenter CL, Cantley LC. Phosphoinositide kinases. Current Opinion in Cell
Biology. 1996;8:153-8.
181. Traynor-Kaplan AE, Harris AL, Thompson BL, Taylor P, Sklar LA. An
inositol tetrakisphosphate-containing phospholipid in activated neutrophils. Nature.
1988;334:353-6.
182. Saltiel AR, Kahn CR. Insulin signalling and the regulation of glucose and lipid
metabolism. Nature. 2001;414:799-806.
183. Lietzke SE, Bose S, Cronin T, Klarlund J, Chawla A, Czech MP, et al.
Structural basis of 3-phosphoinositide recognition by pleckstrin homology domains.
Molecular Cell. 2000;6:385-94.
184. Alessi DR, James SR, Downes CP, Holmes AB, Gaffney PR, Reese CB, et al.
Characterization of a 3-phosphoinositide-dependent protein kinase which
phosphorylates and activates protein kinase Balpha. Current Biology. 1997;7:261-9.
225
185. James SR, Downes CP, Gigg R, Grove SJ, Holmes AB, Alessi DR. Specific
binding of the Akt-1 protein kinase to phosphatidylinositol 3,4,5-trisphosphate
without subsequent activation. Biochemical Journal. 1996;315 ( Pt 3):709-13.
186. Cross DA, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA. Inhibition
of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature.
1995;378:785-9.
187. Newgard CB, Brady MJ, O'Doherty RM, Saltiel AR. Organizing glucose
disposal: emerging roles of the glycogen targeting subunits of protein phosphatase-1.
Diabetes. 2000;49:1967-77.
188. Pilkis SJ, Granner DK. Molecular physiology of the regulation of hepatic
gluconeogenesis and glycolysis. Annual Review of Physiology. 1992;54:885-909.
189. Hall RK, Wang XL, George L, Koch SR, Granner DK. Insulin represses
phosphoenolpyruvate carboxykinase gene transcription by causing the rapid
disruption of an active transcription complex: a potential epigenetic effect. Molecular
Endocrinology. 2007;21:550-63.
190. Hanson RW, Reshef L. Regulation of phosphoenolpyruvate carboxykinase
(GTP) gene expression. Annual Review of Biochemistry. 1997;66:581-611.
191. Nakae J, Kitamura T, Silver DL, Accili D. The forkhead transcription factor
Foxo1 (Fkhr) confers insulin sensitivity onto glucose-6-phosphatase expression.
192. Nakae J, Park BC, Accili D. Insulin stimulates phosphorylation of the
forkhead transcription factor FKHR on serine 253 through a Wortmannin-sensitive
pathway. Journal of Biological Chemistry. 1999;274:15982-5.
193. Barzilai N, Rossetti L. Role of glucokinase and glucose-6-phosphatase in the
acute and chronic regulation of hepatic glucose fluxes by insulin. Journal of
Biological Chemistry. 1993;268:25019-25.
194. Yamada K, Noguchi T. Nutrient and hormonal regulation of pyruvate kinase
gene expression. Biochemical Journal. 1999;337 ( Pt 1):1-11.
195. Cheng Z, White MF. Targeting Forkhead box O1 from the concept to
metabolic diseases: lessons from mouse models. Antioxid Redox Signal. 2011;14:64961.
196. Yoon JC, Puigserver P, Chen G, Donovan J, Wu Z, Rhee J, et al. Control of
hepatic gluconeogenesis through the transcriptional coactivator PGC-1. Nature.
2001;413:131-8.
197. Puigserver P, Rhee J, Donovan J, Walkey CJ, Yoon JC, Oriente F, et al.
Insulin-regulated hepatic gluconeogenesis through FOXO1-PGC-1alpha interaction.
Nature. 2003;423:550-5.
226
198. Holm C, Osterlund T, Laurell H, Contreras JA. Molecular mechanisms
regulating hormone-sensitive lipase and lipolysis. Annual Review of Nutrition.
2000;20:365-93.
199. Munday MR, Hemingway CJ. The regulation of acetyl-CoA carboxylase--a
potential target for the action of hypolipidemic agents. Advances in Enzyme
Regulation. 1999;39:205-34.
200. Horton JD, Goldstein JL, Brown MS. SREBPs: activators of the complete
program of cholesterol and fatty acid synthesis in the liver. Journal of Clinical
201. Azzout-Marniche D, Becard D, Guichard C, Foretz M, Ferre P, Foufelle F.
Insulin effects on sterol regulatory-element-binding protein-1c (SREBP-1c)
transcriptional activity in rat hepatocytes. Biochemical Journal. 2000;350 Pt 2:38993.
202. Fleischmann M, Iynedjian PB. Regulation of sterol regulatory-element binding
protein 1 gene expression in liver: role of insulin and protein kinase B/cAkt.
Biochemical Journal. 2000;349:13-7.
203. Matsumoto M, Ogawa W, Teshigawara K, Inoue H, Miyake K, Sakaue H, et
al. Role of the insulin receptor substrate 1 and phosphatidylinositol 3-kinase signaling
pathway in insulin-induced expression of sterol regulatory element binding protein 1c
and glucokinase genes in rat hepatocytes. Diabetes. 2002;51:1672-80.
204. Chen G, Liang G, Ou J, Goldstein JL, Brown MS. Central role for liver X
receptor in insulin-mediated activation of Srebp-1c transcription and stimulation of
fatty acid synthesis in liver. Proceedings of the National Academy of Sciences of the
United States of America. 2004;101:11245-50.
205. Shimomura I, Bashmakov Y, Ikemoto S, Horton JD, Brown MS, Goldstein JL.
Insulin selectively increases SREBP-1c mRNA in the livers of rats with
streptozotocin-induced diabetes. Proceedings of the National Academy of Sciences of
the United States of America. 1999;96:13656-61.
206. Horton JD, Bashmakov Y, Shimomura I, Shimano H. Regulation of sterol
regulatory element binding proteins in livers of fasted and refed mice. Proceedings of
the National Academy of Sciences of the United States of America. 1998;95:5987-92.
207. Ogawa W, Matozaki T, Kasuga M. Role of binding proteins to IRS-1 in
insulin signalling. Mol Cell Biochem. 1998;182:13-22.
208. Maassen JA, Burgering BM, Medema RH, Osterop AP, van der Zon GC,
Moller W, et al. The role of ras proteins in insulin signal transduction. Horm Metab
Res. 1992;24:214-8.
209. Sakaue M, Bowtell D, Kasuga M. A dominant-negative mutant of mSOS1
inhibits insulin-induced Ras activation and reveals Ras-dependent and -independent
insulin signaling pathways. Mol Cell Biol. 1995;15:379-88.
227
210. Waters SB, Yamauchi K, Pessin JE. Functional expression of insulin receptor
substrate-1 is required for insulin-stimulated mitogenic signaling. J Biol Chem.
1993;268:22231-4.
211. Rose DW, Saltiel AR, Majumdar M, Decker SJ, Olefsky JM. Insulin receptor
substrate 1 is required for insulin-mediated mitogenic signal transduction. Proc Natl
Acad Sci U S A. 1994;91:797-801.
212. Kosztolanyi G. Leprechaunism/Donohue syndrome/insulin receptor gene
mutations: a syndrome delineation story from clinicopathological description to
molecular understanding. European Journal of Pediatrics. 1997;156:253-5.
213. West RJ, Lloyd JK, Turner WM. Familial insulin-resistant diabetes, multiple
somatic anomalies, and pineal hyperplasia. Archives of Disease in Childhood.
1975;50:703-8.
214. Rabson SM, Mendenhall EN. Familial hypertrophy of pineal body,
hyperplasia of adrenal cortex and diabetes mellitus; report of 3 cases. American
Journal of Clinical Pathology. 1956;26:283-90.
215. Kahn SE, Hull RL, Utzschneider KM. Mechanisms linking obesity to insulin
resistance and type 2 diabetes. Nature. 2006;444:840-6.
216. Ogden CL, Carroll MD, B.K. K, Flegal KM. Prevalence of Obesity in the
United States, 2009–2010. NCHS Data Brief [serial on the Internet]. 2012.
217. Ministry of Health. A Portrait of Health: Key results of the 2006/07 New
Zealand Health Survey. Wellington: Ministry of Health2008.
218. University of Otago and Ministry of Health. A Focus on Nutrition: Key
findings of the 2008/09 New Zealand Adult Nutrition Survey. Wellington: Ministry of
Health2011.
219. Stefan N, Kantartzis K, Machann J, Schick F, Thamer C, Rittig K, et al.
Identification and characterization of metabolically benign obesity in humans.
Archives of Internal Medicine. 2008;168:1609-16.
220. Karelis AD.
2008;372:1281-3.
Metabolically
healthy
but
obese
individuals.
Lancet.
221. Karelis AD, Faraj M, Bastard JP, St-Pierre DH, Brochu M, Prud'homme D, et
al. The metabolically healthy but obese individual presents a favorable inflammation
profile. Journal of Clinical Endocrinology and Metabolism. 2005;90:4145-50.
222. Iacobellis G, Ribaudo MC, Zappaterreno A, Iannucci CV, Leonetti F.
Prevalence of uncomplicated obesity in an Italian obese population. Obesity Research.
2005;13:1116-22.
228
223. Brochu M, Tchernof A, Dionne IJ, Sites CK, Eltabbakh GH, Sims EA, et al.
What are the physical characteristics associated with a normal metabolic profile
despite a high level of obesity in postmenopausal women? Journal of Clinical
224. Steil GM, Trivedi N, Jonas JC, Hasenkamp WM, Sharma A, Bonner-Weir S,
et al. Adaptation of beta-cell mass to substrate oversupply: enhanced function with
normal gene expression. Am J Physiol Endocrinol Metab. 2001;280:E788-96.
225. Jetton TL, Lausier J, LaRock K, Trotman WE, Larmie B, Habibovic A, et al.
Mechanisms of compensatory beta-cell growth in insulin-resistant rats: roles of Akt
kinase. Diabetes. 2005;54:2294-304.
226. Chen C, Hosokawa H, Bumbalo LM, Leahy JL. Mechanism of compensatory
hyperinsulinemia in normoglycemic insulin-resistant spontaneously hypertensive rats.
Augmented enzymatic activity of glucokinase in beta-cells. Journal of Clinical
227. Liu YQ, Jetton TL, Leahy JL. beta-Cell adaptation to insulin resistance.
Increased pyruvate carboxylase and malate-pyruvate shuttle activity in islets of
nondiabetic Zucker fatty rats. Journal of Biological Chemistry. 2002;277:39163-8.
228. Kahn SE. Clinical review 135: The importance of beta-cell failure in the
development and progression of type 2 diabetes. Journal of Clinical Endocrinology
and Metabolism. 2001;86:4047-58.
229. DeFronzo RA, Abdul-Ghani MA. Preservation of beta-cell function: the key to
diabetes prevention. Journal of Clinical Endocrinology and Metabolism.
2011;96:2354-66.
230. Knowles NG, Landchild MA, Fujimoto WY, Kahn SE. Insulin and amylin
release are both diminished in first-degree relatives of subjects with type 2 diabetes.
Diabetes Care. 2002;25:292-7.
231. Klein S, Fabbrini E, Romijn JA. Obesity. In: Melmed S, Polonsky KS, Larsen
PR, Kronenberg HM, editors. Williams Textbook of Endocrinology. 12 ed.
Philadelphia, PA, USA: Elsevier Saunders; 2011. p. 1605-32.
232. Scherer PE. Adipose tissue: from lipid storage compartment to endocrine
organ. Diabetes. 2006;55:1537-45.
233. Fain JN, Madan AK, Hiler ML, Cheema P, Bahouth SW. Comparison of the
release of adipokines by adipose tissue, adipose tissue matrix, and adipocytes from
visceral and subcutaneous abdominal adipose tissues of obese humans.
234. Gregor MF, Hotamisligil GS. Inflammatory mechanisms in obesity. Annual
Review of Immunology. 2011;29:415-45.
229
235. Donath MY, Shoelson SE. Type 2 diabetes as an inflammatory disease. Nat
Rev Immunol. 2011;11:98-107.
236. Foster MT, Pagliassotti MJ. Metabolic alterations following visceral fat
removal and expansion: Beyond anatomic location. Adipocyte. 2012;1:192-9.
237. Fontana L, Eagon JC, Trujillo ME, Scherer PE, Klein S. Visceral fat adipokine
secretion is associated with systemic inflammation in obese humans. Diabetes.
2007;56:1010-3.
238. Kim SP, Ellmerer M, Van Citters GW, Bergman RN. Primacy of hepatic
insulin resistance in the development of the metabolic syndrome induced by an
isocaloric moderate-fat diet in the dog. Diabetes. 2003;52:2453-60.
239. O'Connell J, Lynch L, Cawood TJ, Kwasnik A, Nolan N, Geoghegan J, et al.
The Relationship of Omental and Subcutaneous Adipocyte Size to Metabolic Disease
in Severe Obesity. PLoS One. 2010;5:e9997.
240. Bergman RN. Non-esterified fatty acids and the liver: why is insulin secreted
into the portal vein? Diab tologia. 2000;43:946-52.
241. Fabbrini E, Tamboli RA, Magkos F, Marks-Shulman PA, Eckhauser AW,
Richards WO, et al. Surgical removal of omental fat does not improve insulin
sensitivity and cardiovascular risk factors in obese adults. Gastroenterology.
2010;139:448-55.
242. Petersen KF, Oral EA, Dufour S, Befroy D, Ariyan C, Yu C, et al. Leptin
reverses insulin resistance and hepatic steatosis in patients with severe lipodystrophy.
243. Kim JK, Gavrilova O, Chen Y, Reitman ML, Shulman GI. Mechanism of
insulin resistance in A-ZIP/F-1 fatless mice. Journal of Biological Chemistry.
2000;275:8456-60.
244. Samuel VT, Shulman GI. Mechanisms for insulin resistance: common threads
and missing links. Cell. 2012;148:852-71.
245. Boden G. Role of fatty acids in the pathogenesis of insulin resistance and
NIDDM. Diabetes. 1997;46:3-10.
246. Reaven GM, Hollenbeck C, Jeng CY, Wu MS, Chen YD. Measurement of
plasma glucose, free fatty acid, lactate, and insulin for 24 h in patients with NIDDM.
Diabetes. 1988;37:1020-4.
247. Kashyap S, Belfort R, Gastaldelli A, Pratipanawatr T, Berria R, Pratipanawatr
W, et al. A sustained increase in plasma free fatty acids impairs insulin secretion in
nondiabetic subjects genetically predisposed to develop type 2 diabetes. Diabetes.
2003;52:2461-74.
230
248. Unger RH. Lipotoxicity in the pathogenesis of obesity-dependent NIDDM.
Genetic and clinical implications. Diabetes. 1995;44:863-70.
249. Griffin ME, Marcucci MJ, Cline GW, Bell K, Barucci N, Lee D, et al. Free
fatty acid-induced insulin resistance is associated with activation of protein kinase C
theta and alterations in the insulin signaling cascade. Diabetes. 1999;48:1270-4.
250. Dresner A, Laurent D, Marcucci M, Griffin ME, Dufour S, Cline GW, et al.
Effects of free fatty acids on glucose transport and IRS-1-associated
phosphatidylinositol 3-kinase activity. Journal of Clinical Investigation.
1999;103:253-9.
251. Roden M, Price TB, Perseghin G, Petersen KF, Rothman DL, Cline GW, et al.
Mechanism of free fatty acid-induced insulin resistance in humans. Journal of
252. Levin K, Daa Schroeder H, Alford FP, Beck-Nielsen H. Morphometric
documentation of abnormal intramyocellular fat storage and reduced glycogen in
obese patients with Type II diabetes. Diabetologia. 2001;44:824-33.
253. Krssak M, Falk Petersen K, Dresner A, DiPietro L, Vogel SM, Rothman DL,
et al. Intramyocellular lipid concentrations are correlated with insulin sensitivity in
humans: a 1H NMR spectroscopy study. Diabetologia. 1999;42:113-6.
254. Angulo P. Nonalcoholic fatty liver disease. New England Journal of Medicine.
2002;346:1221-31.
255. Fabbrini E, Magkos F, Mohammed BS, Pietka T, Abumrad NA, Patterson
BW, et al. Intrahepatic fat, not visceral fat, is linked with metabolic complications of
obesity. Proceedings of the National Academy of Sciences of the United States of
America. 2009;106:15430-5.
256. Petersen KF, Dufour S, Befroy D, Lehrke M, Hendler RE, Shulman GI.
Reversal of nonalcoholic hepatic steatosis, hepatic insulin resistance, and
hyperglycemia by moderate weight reduction in patients with type 2 diabetes.
Diabetes. 2005;54:603-8.
257. Lim EL, Hollingsworth KG, Aribisala BS, Chen MJ, Mathers JC, Taylor R.
Reversal of type 2 diabetes: normalisation of beta cell function in association with
decreased pancreas and liver triacylglycerol. Diabetologia. 2011;54:2506-14.
258. Turinsky J, O'Sullivan DM, Bayly BP. 1,2-Diacylglycerol and ceramide levels
in insulin-resistant tissues of the rat in vivo. Journal of Biological Chemistry.
1990;265:16880-5.
259. Itani SI, Ruderman NB, Schmieder F, Boden G. Lipid-induced insulin
resistance in human muscle is associated with changes in diacylglycerol, protein
kinase C, and IkappaB-alpha. Diabetes. 2002;51:2005-11.
231
260. Li Y, Soos TJ, Li X, Wu J, Degennaro M, Sun X, et al. Protein kinase C Theta
inhibits insulin signaling by phosphorylating IRS1 at Ser(1101). Journal of Biological
Chemistry. 2004;279:45304-7.
261. Yu C, Chen Y, Cline GW, Zhang D, Zong H, Wang Y, et al. Mechanism by
which fatty acids inhibit insulin activation of insulin receptor substrate-1 (IRS-1)associated phosphatidylinositol 3-kinase activity in muscle. Journal of Biological
Chemistry. 2002;277:50230-6.
262. Samuel VT, Liu ZX, Wang A, Beddow SA, Geisler JG, Kahn M, et al.
Inhibition of protein kinase Cepsilon prevents hepatic insulin resistance in
nonalcoholic fatty liver disease. Journal of Clinical Investigation. 2007;117:739-45.
263. Schmitz-Peiffer C, Craig DL, Biden TJ. Ceramide generation is sufficient to
account for the inhibition of the insulin-stimulated PKB pathway in C2C12 skeletal
muscle cells pretreated with palmitate. Journal of Biological Chemistry.
1999;274:24202-10.
264. Stratford S, Hoehn KL, Liu F, Summers SA. Regulation of insulin action by
ceramide: dual mechanisms linking ceramide accumulation to the inhibition of
Akt/protein kinase B. Journal of Biological Chemistry. 2004;279:36608-15.
265. Fox TE, Houck KL, O'Neill SM, Nagarajan M, Stover TC, Pomianowski PT,
et al. Ceramide recruits and activates protein kinase C zeta (PKC zeta) within
structured membrane microdomains. Journal of Biological Chemistry.
2007;282:12450-7.
266. Blouin CM, Prado C, Takane KK, Lasnier F, Garcia-Ocana A, Ferre P, et al.
Plasma membrane subdomain compartmentalization contributes to distinct
mechanisms of ceramide action on insulin signaling. Diabetes. 2010;59:600-10.
267. Powell DJ, Hajduch E, Kular G, Hundal HS. Ceramide disables 3phosphoinositide binding to the pleckstrin homology domain of protein kinase B
(PKB)/Akt by a PKCzeta-dependent mechanism. Molecular and Cellular Biology.
2003;23:7794-808.
268. Yudkin JS, Kumari M, Humphries SE, Mohamed-Ali V. Inflammation,
obesity, stress and coronary heart disease: is interleukin-6 the link? Atherosclerosis.
2000;148:209-14.
269. Visser M, Bouter LM, McQuillan GM, Wener MH, Harris TB. Elevated Creactive protein levels in overweight and obese adults. JAMA. 1999;282:2131-5.
270. Pickup JC, Crook MA. Is type II diabetes mellitus a disease of the innate
immune system? Diab tologia. 1998;41:1241-8.
271. Pickup JC, Mattock MB, Chusney GD, Burt D. NIDDM as a disease of the
innate immune system: association of acute-phase reactants and interleukin-6 with
metabolic syndrome X. Diabetologia. 1997;40:1286-92.
232
272. Spranger J, Kroke A, Mohlig M, Hoffmann K, Bergmann MM, Ristow M, et
al. Inflammatory cytokines and the risk to develop type 2 diabetes: results of the
prospective population-based European Prospective Investigation into Cancer and
Nutrition (EPIC)-Potsdam Study. Diabetes. 2003;52:812-7.
273. Pradhan AD, Manson JE, Rifai N, Buring JE, Ridker PM. C-reactive protein,
interleukin 6, and risk of developing type 2 diabetes mellitus. JAMA. 2001;286:32734.
274. Herder C, Brunner EJ, Rathmann W, Strassburger K, Tabak AG, Schloot NC,
et al. Elevated levels of the anti-inflammatory interleukin-1 receptor antagonist
precede the onset of type 2 diabetes: the Whitehall II study. Diabetes Care.
2009;32:421-3.
275. Carstensen M, Herder C, Kivimaki M, Jokela M, Roden M, Shipley MJ, et al.
Accelerated increase in serum interleukin-1 receptor antagonist starts 6 years before
diagnosis of type 2 diabetes: Whitehall II prospective cohort study. Diabetes.
2010;59:1222-7.
276. Hotamisligil GS, Shargill NS, Spiegelman BM. Adipose expression of tumor
necrosis factor-alpha: direct role in obesity-linked insulin resistance. Science.
1993;259:87-91.
277. Saghizadeh M, Ong JM, Garvey WT, Henry RR, Kern PA. The expression of
TNF alpha by human muscle. Relationship to insulin resistance. Journal of Clinical
278. Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF, Spiegelman BM.
IRS-1-mediated inhibition of insulin receptor tyrosine kinase activity in TNF-alphaand obesity-induced insulin resistance. Science. 1996;271:665-8.
279. Aguirre V, Uchida T, Yenush L, Davis R, White MF. The c-Jun NH(2)terminal kinase promotes insulin resistance during association with insulin receptor
substrate-1 and phosphorylation of Ser(307). Journal of Biological Chemistry.
2000;275:9047-54.
280. Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ, et al. Chronic
inflammation in fat plays a crucial role in the development of obesity-related insulin
resistance. Journal of Clinical Investigation. 2003;112:1821-30.
281. Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante AW,
Jr. Obesity is associated with macrophage accumulation in adipose tissue. Journal of
282. Anderson EK, Gutierrez DA, Hasty AH. Adipose tissue recruitment of
leukocytes. Current Opinion in Lipidology. 2010;21:172-7.
283. Weisberg SP, Hunter D, Huber R, Lemieux J, Slaymaker S, Vaddi K, et al.
CCR2 modulates inflammatory and metabolic effects of high-fat feeding. Journal of
233
284. Kanda H, Tateya S, Tamori Y, Kotani K, Hiasa K, Kitazawa R, et al. MCP-1
contributes to macrophage infiltration into adipose tissue, insulin resistance, and
hepatic steatosis in obesity. Journal of Clinical Investigation. 2006;116:1494-505.
285. Meijer K, de Vries M, Al-Lahham S, Bruinenberg M, Weening D, Dijkstra M,
et al. Human primary adipocytes exhibit immune cell function: adipocytes prime
inflammation independent of macrophages. PLoS One. 2011;6:e17154.
286. Kloting N, Fasshauer M, Dietrich A, Kovacs P, Schon MR, Kern M, et al.
Insulin-sensitive obesity. Am J Physiol Endocrinol Metab. 2010;299:E506-15.
287. Solinas G, Karin M. JNK1 and IKKbeta: molecular links between obesity and
metabolic dysfunction. FASEB Journal. 2010;24:2596-611.
288. Gao Z, Zhang X, Zuberi A, Hwang D, Quon MJ, Lefevre M, et al. Inhibition
of insulin sensitivity by free fatty acids requires activation of multiple serine kinases
in 3T3-L1 adipocytes. Molecular Endocrinology. 2004;18:2024-34.
289. Yuan M, Konstantopoulos N, Lee J, Hansen L, Li ZW, Karin M, et al.
Reversal of obesity- and diet-induced insulin resistance with salicylates or targeted
disruption of Ikkbeta. Science. 2001;293:1673-7.
290. Kim JK, Kim YJ, Fillmore JJ, Chen Y, Moore I, Lee J, et al. Prevention of fatinduced insulin resistance by salicylate. Journal of Clinical Investigation.
2001;108:437-46.
291. Gao Z, Hwang D, Bataille F, Lefevre M, York D, Quon MJ, et al. Serine
phosphorylation of insulin receptor substrate 1 by inhibitor kappa B kinase complex.
292. Arkan MC, Hevener AL, Greten FR, Maeda S, Li ZW, Long JM, et al. IKKbeta links inflammation to obesity-induced insulin resistance. Nature Medicine.
2005;11:191-8.
293. Shoelson SE, Lee J, Yuan M. Inflammation and the IKK beta/I kappa B/NFkappa B axis in obesity- and diet-induced insulin resistance. International Journal of
Obesity and Related Metabolic Disorders. 2003;27 Suppl 3:S49-52.
294. Cai D, Yuan M, Frantz DF, Melendez PA, Hansen L, Lee J, et al. Local and
systemic insulin resistance resulting from hepatic activation of IKK-beta and NFkappaB. Nature Medicine. 2005;11:183-90.
295. Rohl M, Pasparakis M, Baudler S, Baumgartl J, Gautam D, Huth M, et al.
Conditional disruption of IkappaB kinase 2 fails to prevent obesity-induced insulin
resistance. Journal of Clinical Investigation. 2004;113:474-81.
296. Davis RJ. Signal transduction by the JNK group of MAP kinases. Cell.
2000;103:239-52.
234
297. Hirosumi J, Tuncman G, Chang L, Gorgun CZ, Uysal KT, Maeda K, et al. A
central role for JNK in obesity and insulin resistance. Nature. 2002;420:333-6.
298. Aguirre V, Werner ED, Giraud J, Lee YH, Shoelson SE, White MF.
Phosphorylation of Ser307 in insulin receptor substrate-1 blocks interactions with the
insulin receptor and inhibits insulin action. Journal of Biological Chemistry.
2002;277:1531-7.
299. Nakatani Y, Kaneto H, Kawamori D, Hatazaki M, Miyatsuka T, Matsuoka
TA, et al. Modulation of the JNK pathway in liver affects insulin resistance status.
300. Kaneto H, Nakatani Y, Miyatsuka T, Kawamori D, Matsuoka TA, Matsuhisa
M, et al. Possible novel therapy for diabetes with cell-permeable JNK-inhibitory
peptide. Nature Medicine. 2004;10:1128-32.
301. Larsen CM, Faulenbach M, Vaag A, Volund A, Ehses JA, Seifert B, et al.
Interleukin-1-receptor antagonist in type 2 diabetes mellitus. New England Journal of
Medicine. 2007;356:1517-26.
302. Wen H, Gris D, Lei Y, Jha S, Zhang L, Huang MT, et al. Fatty acid-induced
NLRP3-ASC inflammasome activation interferes with insulin signaling. Nat
Immunol. 2011;12:408-15.
303. Stienstra R, Tack CJ, Kanneganti TD, Joosten LA, Netea MG. The
inflammasome puts obesity in the danger zone. Cell Metab. 2012;15:10-8.
304. Donath MY, Storling J, Maedler K, Mandrup-Poulsen T. Inflammatory
mediators and islet beta-cell failure: a link between type 1 and type 2 diabetes. J Mol
Med (Berl). 2003;81:455-70.
305. Jager J, Gremeaux T, Cormont M, Le Marchand-Brustel Y, Tanti JF.
Interleukin-1beta-induced insulin resistance in adipocytes through down-regulation of
insulin receptor substrate-1 expression. Endocrinology. 2007;148:241-51.
306. Dinarello CA. Interleukin-1 in the pathogenesis and treatment of inflammatory
diseases. Blood. 2011;117:3720-32.
307. McGillicuddy FC, Harford KA, Reynolds CM, Oliver E, Claessens M, Mills
KH, et al. Lack of interleukin-1 receptor I (IL-1RI) protects mice from high-fat dietinduced adipose tissue inflammation coincident with improved glucose homeostasis.
Diabetes. 2011;60:1688-98.
308. Hotamisligil GS. Endoplasmic reticulum stress and the inflammatory basis of
metabolic disease. Cell. 2010;140:900-17.
309. Ron D, Walter P. Signal integration in the endoplasmic reticulum unfolded
protein response. Nat Rev Mol Cell Biol. 2007;8:519-29.
235
310. Bertolotti A, Zhang Y, Hendershot LM, Harding HP, Ron D. Dynamic
interaction of BiP and ER stress transducers in the unfolded-protein response. Nat
Cell Biol. 2000;2:326-32.
311. Deng J, Lu PD, Zhang Y, Scheuner D, Kaufman RJ, Sonenberg N, et al.
Translational repression mediates activation of nuclear factor kappa B by
phosphorylated translation initiation factor 2. Molecular and Cellular Biology.
2004;24:10161-8.
312. Hu P, Han Z, Couvillon AD, Kaufman RJ, Exton JH. Autocrine tumor
necrosis factor alpha links endoplasmic reticulum stress to the membrane death
receptor pathway through IRE1alpha-mediated NF-kappaB activation and downregulation of TRAF2 expression. Molecular and Cellular Biology. 2006;26:3071-84.
313. Urano F, Wang X, Bertolotti A, Zhang Y, Chung P, Harding HP, et al.
Coupling of stress in the ER to activation of JNK protein kinases by transmembrane
protein kinase IRE1. Science. 2000;287:664-6.
314. Yamazaki H, Hiramatsu N, Hayakawa K, Tagawa Y, Okamura M, Ogata R, et
al. Activation of the Akt-NF-kappaB pathway by subtilase cytotoxin through the
ATF6 branch of the unfolded protein response. Journal of Immunology.
2009;183:1480-7.
315. Li Y, Schwabe RF, DeVries-Seimon T, Yao PM, Gerbod-Giannone MC, Tall
AR, et al. Free cholesterol-loaded macrophages are an abundant source of tumor
necrosis factor-alpha and interleukin-6: model of NF-kappaB- and map kinasedependent inflammation in advanced atherosclerosis. Journal of Biological
Chemistry. 2005;280:21763-72.
316. Lee MW, Chanda D, Yang J, Oh H, Kim SS, Yoon YS, et al. Regulation of
hepatic gluconeogenesis by an ER-bound transcription factor, CREBH. Cell Metab.
2010;11:331-9.
317. Oyadomari S, Harding HP, Zhang Y, Oyadomari M, Ron D.
Dephosphorylation of translation initiation factor 2alpha enhances glucose tolerance
and attenuates hepatosteatosis in mice. Cell Metab. 2008;7:520-32.
318. Scheuner D, Song B, McEwen E, Liu C, Laybutt R, Gillespie P, et al.
Translational control is required for the unfolded protein response and in vivo glucose
homeostasis. Molecular Cell. PER2001;7:1165-76.
319. Birkenfeld AL, Lee HY, Majumdar S, Jurczak MJ, Camporez JP, Jornayvaz
FR, et al. Influence of the hepatic eukaryotic initiation factor 2alpha (eIF2alpha)
endoplasmic reticulum (ER) stress response pathway on insulin-mediated ER stress
and hepatic and peripheral glucose metabolism. Journal of Biological Chemistry.
2011;286:36163-70.
320. Raghow R, Yellaturu C, Deng X, Park EA, Elam MB. SREBPs: the crossroads
of physiological and pathological lipid homeostasis. Trends Endocrinol Metab.
2008;19:65-73.
236
321. Kammoun HL, Chabanon H, Hainault I, Luquet S, Magnan C, Koike T, et al.
GRP78 expression inhibits insulin and ER stress-induced SREBP-1c activation and
reduces hepatic steatosis in mice. Journal of Clinical Investigation. 2009;119:120115.
322. Zhang K, Wang S, Malhotra J, Hassler JR, Back SH, Wang G, et al. The
unfolded protein response transducer IRE1alpha prevents ER stress-induced hepatic
steatosis. EMBO Journal. 2011;30:1357-75.
323. Yamamoto K, Takahara K, Oyadomari S, Okada T, Sato T, Harada A, et al.
Induction of liver steatosis and lipid droplet formation in ATF6alpha-knockout mice
burdened with pharmacological endoplasmic reticulum stress. Molecular Biology of
the Cell. 2010;21:2975-86.
324. Ozcan U, Cao Q, Yilmaz E, Lee AH, Iwakoshi NN, Ozdelen E, et al.
Endoplasmic reticulum stress links obesity, insulin action, and type 2 diabetes.
Science. 2004;306:457-61.
325. Tuncman G, Hirosumi J, Solinas G, Chang L, Karin M, Hotamisligil GS.
Functional in vivo interactions between JNK1 and JNK2 isoforms in obesity and
insulin resistance. Proceedings of the National Academy of Sciences of the United
326. Gregor MF, Yang L, Fabbrini E, Mohammed BS, Eagon JC, Hotamisligil GS,
et al. Endoplasmic reticulum stress is reduced in tissues of obese subjects after weight
loss. Diabetes. 2009;58:693-700.
327. Ozcan U, Yilmaz E, Ozcan L, Furuhashi M, Vaillancourt E, Smith RO, et al.
Chemical chaperones reduce ER stress and restore glucose homeostasis in a mouse
model of type 2 diabetes. Science. 2006;313:1137-40.
328. Kars M, Yang L, Gregor MF, Mohammed BS, Pietka TA, Finck BN, et al.
Tauroursodeoxycholic Acid may improve liver and muscle but not adipose tissue
insulin sensitivity in obese men and women. Diabetes. 2010;59:1899-905.
329. Yang JS, Kim JT, Jeon J, Park HS, Kang GH, Park KS, et al. Changes in
hepatic gene expression upon oral administration of taurine-conjugated
ursodeoxycholic acid in ob/ob mice. PLoS One. 2010;5:e13858.
330. da-Silva WS, Ribich S, Arrojo e Drigo R, Castillo M, Patti ME, Bianco AC.
The chemical chaperones tauroursodeoxycholic and 4-phenylbutyric acid accelerate
thyroid hormone activation and energy expenditure. FEBS Letters. 2011;585:539-44.
331. Lee JY, Sohn KH, Rhee SH, Hwang D. Saturated fatty acids, but not
unsaturated fatty acids, induce the expression of cyclooxygenase-2 mediated through
Toll-like receptor 4. Journal of Biological Chemistry. 2001;276:16683-9.
332. Lee JY, Zhao L, Youn HS, Weatherill AR, Tapping R, Feng L, et al. Saturated
fatty acid activates but polyunsaturated fatty acid inhibits Toll-like receptor 2
237
dimerized with Toll-like receptor 6 or 1. Journal of Biological Chemistry.
2004;279:16971-9.
333. Shi H, Kokoeva MV, Inouye K, Tzameli I, Yin H, Flier JS. TLR4 links innate
immunity and fatty acid-induced insulin resistance. Journal of Clinical Investigation.
2006;116:3015-25.
334. Haversen L, Danielsson KN, Fogelstrand L, Wiklund O. Induction of
proinflammatory cytokines by long-chain saturated fatty acids in human
macrophages. Atherosclerosis. 2009;202:382-93.
335. Kawakami M, Murase T, Ogawa H, Ishibashi S, Mori N, Takaku F, et al.
Human recombinant TNF suppresses lipoprotein lipase activity and stimulates
lipolysis in 3T3-L1 cells. J Biochem. 1987;101:331-8.
336. Mattacks CA, Pond CM. Interactions of noradrenalin and tumour necrosis
factor alpha, interleukin 4 and interleukin 6 in the control of lipolysis from adipocytes
around lymph nodes. Cytokine. 1999;11:334-46.
337. Watt MJ, Carey AL, Wolsk-Petersen E, Kraemer FB, Pedersen BK, Febbraio
MA. Hormone-sensitive lipase is reduced in the adipose tissue of patients with type 2
diabetes mellitus: influence of IL-6 infusion. Diab tologia. 2005;48:105-12.
338. Wolsk E, Mygind H, Grondahl TS, Pedersen BK, van Hall G. IL-6 selectively
stimulates fat metabolism in human skeletal muscle. Am J Physiol Endocrinol Metab.
2010;299:E832-40.
339. Bezaire V, Mairal A, Anesia R, Lefort C, Langin D. Chronic TNFalpha and
cAMP pre-treatment of human adipocytes alter HSL, ATGL and perilipin to regulate
basal and stimulated lipolysis. FEBS Letters. 2009;583:3045-9.
340. Laurencikiene J, van Harmelen V, Arvidsson Nordstrom E, Dicker A,
Blomqvist L, Naslund E, et al. NF-kappaB is important for TNF-alpha-induced
lipolysis in human adipocytes. Journal of Lipid Research. 2007;48:1069-77.
341. Souza SC, de Vargas LM, Yamamoto MT, Lien P, Franciosa MD, Moss LG,
et al. Overexpression of perilipin A and B blocks the ability of tumor necrosis factor
alpha to increase lipolysis in 3T3-L1 adipocytes. Journal of Biological Chemistry.
1998;273:24665-9.
342. Tansey JT, Sztalryd C, Gruia-Gray J, Roush DL, Zee JV, Gavrilova O, et al.
Perilipin ablation results in a lean mouse with aberrant adipocyte lipolysis, enhanced
leptin production, and resistance to diet-induced obesity. Proceedings of the National
343. Martinez-Botas J, Anderson JB, Tessier D, Lapillonne A, Chang BH, Quast
MJ, et al. Absence of perilipin results in leanness and reverses obesity in Lepr(db/db)
mice. Nature Genetics. 2000;26:474-9.
238
344. Ranjit S, Boutet E, Gandhi P, Prot M, Tamori Y, Chawla A, et al. Regulation
of fat specific protein 27 by isoproterenol and TNF-alpha to control lipolysis in
murine adipocytes. Journal of Lipid Research. 2011;52:221-36.
345. Almon RR, DuBois DC, Lai W, Xue B, Nie J, Jusko WJ. Gene expression
analysis of hepatic roles in cause and development of diabetes in Goto-Kakizaki rats.
Journal of Endocrinology. 2009;200:331-46.
346. Do GM, Oh HY, Kwon EY, Cho YY, Shin SK, Park HJ, et al. Long-term
adaptation of global transcription and metabolism in the liver of high-fat diet-fed
C57BL/6J mice. Mol Nutr Food Res. 2011;55 Suppl 2:S173-85.
347. Radonjic M, de Haan JR, van Erk MJ, van Dijk KW, van den Berg SA, de
Groot PJ, et al. Genome-wide mRNA expression analysis of hepatic adaptation to
high-fat diets reveals switch from an inflammatory to steatotic transcriptional
program. PLoS One. 2009;4:e6646.
348. Kleemann R, van Erk M, Verschuren L, van den Hoek AM, Koek M,
Wielinga PY, et al. Time-resolved and tissue-specific systems analysis of the
pathogenesis of insulin resistance. PLoS One. 2010;5:e8817.
349. Stanton MC, Chen SC, Jackson JV, Rojas-Triana A, Kinsley D, Cui L, et al.
Inflammatory Signals shift from adipose to liver during high fat feeding and influence
the development of steatohepatitis in mice. J Inflamm (Lond). 2011;8:8.
350. Nov O, Shapiro H, Ovadia H, Tarnovscki T, Dvir I, Shemesh E, et al.
Interleukin-1β Regulates Fat-Liver Crosstalk in Obesity by Auto-Paracrine
Modulation of Adipose Tissue Inflammation and Expandability. PLoS One.
2013;8:e53626.
351. Brown MS, Goldstein JL. Selective versus total insulin resistance: a
pathogenic paradox. Cell Metab. 2008;7:95-6.
352. Shimomura I, Matsuda M, Hammer RE, Bashmakov Y, Brown MS, Goldstein
JL. Decreased IRS-2 and increased SREBP-1c lead to mixed insulin resistance and
sensitivity in livers of lipodystrophic and ob/ob mice. Molecular Cell. 2000;6:77-86.
353. Boni-Schnetzler M, Boller S, Debray S, Bouzakri K, Meier DT, Prazak R, et
al. Free fatty acids induce a proinflammatory response in islets via the abundantly
expressed interleukin-1 receptor I. Endocrinology. 2009;150:5218-29.
354. Lee AH, Heidtman K, Hotamisligil GS, Glimcher LH. Dual and opposing
roles of the unfolded protein response regulated by IRE1alpha and XBP1 in proinsulin
processing and insulin secretion. Proceedings of the National Academy of Sciences of
355. Patane G, Anello M, Piro S, Vigneri R, Purrello F, Rabuazzo AM. Role of
ATP production and uncoupling protein-2 in the insulin secretory defect induced by
chronic exposure to high glucose or free fatty acids and effects of peroxisome
proliferator-activated receptor-gamma inhibition. Diabetes. 2002;51:2749-56.
239
356. Rossetti L, Giaccari A, DeFronzo RA. Glucose toxicity. Diabetes Care.
1990;13:610-30.
357. Seino S, Seino M, Nishi S, Bell GI. Structure of the human insulin receptor
gene and characterization of its promoter. Proceedings of the National Academy of
358. Seino S, Bell GI. Alternative splicing of human insulin receptor messenger
RNA. Biochemical and Biophysical Research Communications. 1989;159:312-6.
359. Hernandez-Sanchez C, Mansilla A, de Pablo F, Zardoya R. Evolution of the
insulin receptor family and receptor isoform expression in vertebrates. Molecular
Biology and Evolution. 2008;25:1043-53.
360. Kosaki A, Nelson J, Webster NJ. Identification of intron and exon sequences
involved in alternative splicing of insulin receptor pre-mRNA. Journal of Biological
Chemistry. 1998;273:10331-7.
361. Webster NJ, Evans LG, Caples M, Erker L, Chew SL. Assembly of splicing
complexes on exon 11 of the human insulin receptor gene does not correlate with
splicing efficiency in-vitro. BMC Mol Biol. 2004;5:7.
362. Black DL. Mechanisms of alternative pre-messenger RNA splicing. Annual
Review of Biochemistry. 2003;72:291-336.
363. Kramer A. The structure and function of proteins involved in mammalian premRNA splicing. Annual Review of Biochemistry. 1996;65:367-409.
364. Lopez AJ. Alternative splicing of pre-mRNA: developmental consequences
and mechanisms of regulation. Annual Review of Genetics. 1998;32:279-305.
365. Mosthaf L, Grako K, Dull TJ, Coussens L, Ullrich A, McClain DA.
Functionally distinct insulin receptors generated by tissue-specific alternative
splicing. EMBO Journal. 1990;9:2409-13.
366. Moller DE, Yokota A, Caro JF, Flier JS. Tissue-specific expression of two
alternatively spliced insulin receptor mRNAs in man. Molecular Endocrinology.
1989;3:1263-9.
367. Frasca F, Pandini G, Scalia P, Sciacca L, Mineo R, Costantino A, et al. Insulin
receptor isoform A, a newly recognized, high-affinity insulin-like growth factor II
receptor in fetal and cancer cells. Molecular and Cellular Biology. 1999;19:3278-88.
368. Harrington SC, Weroha SJ, Reynolds C, Suman VJ, Lingle WL, Haluska P.
Quantifying insulin receptor isoform expression in FFPE breast tumors. Growth
Hormone and IGF Research. 2012.
240
369. Huang J, Morehouse C, Streicher K, Higgs BW, Gao J, Czapiga M, et al.
Altered expression of insulin receptor isoforms in breast cancer. PLoS One.
2011;6:e26177.
370. Denley A, Wallace JC, Cosgrove LJ, Forbes BE. The insulin receptor isoform
exon 11- (IR-A) in cancer and other diseases: a review. Hormone and Metabolic
Research. 2003;35:778-85.
371. McClain DA. Different ligand affinities of the two human insulin receptor
splice variants are reflected in parallel changes in sensitivity for insulin action.
Molecular Endocrinology. 1991;5:734-9.
372. Yamaguchi Y, Flier JS, Benecke H, Ransil BJ, Moller DE. Ligand-binding
properties of the two isoforms of the human insulin receptor. Endocrinology.
1993;132:1132-8.
373. Vogt B, Carrascosa JM, Ermel B, Ullrich A, Haring HU. The two isotypes of
the human insulin receptor (HIR-A and HIR-B) follow different internalization
kinetics. Biochemical and Biophysical Research Communications. 1991;177:1013-8.
374. Yamaguchi Y, Flier JS, Yokota A, Benecke H, Backer JM, Moller DE.
Functional properties of two naturally occurring isoforms of the human insulin
receptor in Chinese hamster ovary cells. Endocrinology. 1991;129:2058-66.
375. Li Calzi S, Choice CV, Najjar SM. Differential effect of pp120 on insulin
endocytosis by two variant insulin receptor isoforms. American Journal of
Physiology. 1997;273:E801-8.
376. Kellerer M, Lammers R, Ermel B, Tippmer S, Vogt B, Obermaier-Kusser B,
et al. Distinct alpha-subunit structures of human insulin receptor A and B variants
determine differences in tyrosine kinase activities. Biochemistry. 1992;31:4588-96.
377. Kosaki A, Webster NJ. Effect of dexamethasone on the alternative splicing of
the insulin receptor mRNA and insulin action in HepG2 hepatoma cells. Journal of
378. Kosaki A, Pillay TS, Xu L, Webster NJ. The B isoform of the insulin receptor
signals more efficiently than the A isoform in HepG2 cells. Journal of Biological
Chemistry. 1995;270:20816-23.
379. Morrione A, Valentinis B, Xu SQ, Yumet G, Louvi A, Efstratiadis A, et al.
Insulin-like growth factor II stimulates cell proliferation through the insulin receptor.
Proceedings of the National Academy of Sciences of the United States of America.
1997;94:3777-82.
380. Louvi A, Accili D, Efstratiadis A. Growth-promoting interaction of IGF-II
with the insulin receptor during mouse embryonic development. Developmental
Biology. 1997;189:33-48.
241
381. Sciacca L, Prisco M, Wu A, Belfiore A, Vigneri R, Baserga R. Signaling
differences from the A and B isoforms of the insulin receptor (IR) in 32D cells in the
presence or absence of IR substrate-1. Endocrinology. 2003;144:2650-8.
382. Pandini G, Medico E, Conte E, Sciacca L, Vigneri R, Belfiore A. Differential
gene expression induced by insulin and insulin-like growth factor-II through the
insulin receptor isoform A. Journal of Biological Chemistry. 2003;278:42178-89.
383. Gonzalez-Rodriguez A, Nevado C, Escriva F, Sesti G, Rondinone CM, Benito
M, et al. PTP1B deficiency increases glucose uptake in neonatal hepatocytes:
involvement of IRA/GLUT2 complexes. Am J Physiol Gastrointest Liver Physiol.
2008;295:G338-47.
384. Nevado C, Valverde AM, Benito M. Role of insulin receptor in the regulation
of glucose uptake in neonatal hepatocytes. Endocrinology. 2006;147:3709-18.
385. Malaguarnera R, Sacco A, Voci C, Pandini G, Vigneri R, Belfiore A.
Proinsulin binds with high affinity the insulin receptor isoform A and predominantly
activates the mitogenic pathway. Endocrinology. 2012;153:2152-63.
386. Kalli KR, Falowo OI, Bale LK, Zschunke MA, Roche PC, Conover CA.
Functional insulin receptors on human epithelial ovarian carcinoma cells: implications
for IGF-II mitogenic signaling. Endocrinology. 2002;143:3259-67.
387. Sciacca L, Mineo R, Pandini G, Murabito A, Vigneri R, Belfiore A. In IGF-I
receptor-deficient leiomyosarcoma cells autocrine IGF-II induces cell invasion and
protection from apoptosis via the insulin receptor isoform A. Oncogene.
2002;21:8240-50.
388. Vella V, Pandini G, Sciacca L, Mineo R, Vigneri R, Pezzino V, et al. A novel
autocrine loop involving IGF-II and the insulin receptor isoform-A stimulates growth
of thyroid cancer. Journal of Clinical Endocrinology and Metabolism. 2002;87:24554.
389. Uhles S, Moede T, Leibiger B, Berggren PO, Leibiger IB. Isoform-specific
insulin receptor signaling involves different plasma membrane domains. Journal of
Cell Biology. 2003;163:1327-37.
390. Giudice J, Leskow FC, Arndt-Jovin DJ, Jovin TM, Jares-Erijman EA.
Differential endocytosis and signaling dynamics of insulin receptor variants IR-A and
IR-B. Journal of Cell Science. 2011;124:801-11.
391. Bergeron JJ, Di Guglielmo GM, Baass PC, Authier F, Posner BI. Endosomes,
receptor tyrosine kinase internalization and signal transduction. Bioscience Reports.
1995;15:411-8.
392. Di Guglielmo GM, Baass PC, Ou WJ, Posner BI, Bergeron JJ.
Compartmentalization of SHC, GRB2 and mSOS, and hyperphosphorylation of Raf-1
by EGF but not insulin in liver parenchyma. EMBO Journal. 1994;13:4269-77.
242
393. Rizzo MA, Shome K, Watkins SC, Romero G. The recruitment of Raf-1 to
membranes is mediated by direct interaction with phosphatidic acid and is
independent of association with Ras. Journal of Biological Chemistry.
2000;275:23911-8.
394. Sorkin A, McClure M, Huang F, Carter R. Interaction of EGF receptor and
grb2 in living cells visualized by fluorescence resonance energy transfer (FRET)
microscopy. Current Biology. 2000;10:1395-8.
395. Ceresa BP, Kao AW, Santeler SR, Pessin JE. Inhibition of clathrin-mediated
endocytosis selectively attenuates specific insulin receptor signal transduction
pathways. Molecular and Cellular Biology. 1998;18:3862-70.
396. Hamer I, Foti M, Emkey R, Cordier-Bussat M, Philippe J, De Meyts P, et al.
An arginine to cysteine(252) mutation in insulin receptors from a patient with severe
insulin resistance inhibits receptor internalisation but preserves signalling events.
Diab tologia. 2002;45:657-67.
397. Blanquart C, Achi J, Issad T. Characterization of IRA/IRB hybrid insulin
receptors using bioluminescence resonance energy transfer. Biochemical
Pharmacology. 2008;76:873-83.
398. Soos MA, Whittaker J, Lammers R, Ullrich A, Siddle K. Receptors for insulin
and insulin-like growth factor-I can form hybrid dimers. Characterisation of hybrid
receptors in transfected cells. Biochemical Journal. 1990;270:383-90.
399. Moxham CP, Duronio V, Jacobs S. Insulin-like growth factor I receptor betasubunit heterogeneity. Evidence for hybrid tetramers composed of insulin-like growth
factor I and insulin receptor heterodimers. Journal of Biological Chemistry.
1989;264:13238-44.
400. Pandini G, Wurch T, Akla B, Corvaia N, Belfiore A, Goetsch L. Functional
responses and in vivo anti-tumour activity of h7C10: a humanised monoclonal
antibody with neutralising activity against the insulin-like growth factor-1 (IGF-1)
receptor and insulin/IGF-1 hybrid receptors. European Journal of Cancer.
2007;43:1318-27.
401. Frattali AL, Pessin JE. Relationship between alpha subunit ligand occupancy
and beta subunit autophosphorylation in insulin/insulin-like growth factor-1 hybrid
receptors. Journal of Biological Chemistry. 1993;268:7393-400.
402. Bailyes EM, Nave BT, Soos MA, Orr SR, Hayward AC, Siddle K. Insulin
receptor/IGF-I receptor hybrids are widely distributed in mammalian tissues:
quantification of individual receptor species by selective immunoprecipitation and
immunoblotting. Biochemical Journal. 1997;327 ( Pt 1):209-15.
403. Siddle K, Soos MA, Field CE, Nave BT. Hybrid and atypical insulin/insulinlike growth factor I receptors. Hormone Research. 1994;41 Suppl 2:56-64; discussion
5.
243
404. Benyoucef S, Surinya KH, Hadaschik D, Siddle K. Characterization of
insulin/IGF hybrid receptors: contributions of the insulin receptor L2 and Fn1
domains and the alternatively spliced exon 11 sequence to ligand binding and receptor
activation. Biochemical Journal. 2007;403:603-13.
405. Pandini G, Vigneri R, Costantino A, Frasca F, Ippolito A, Fujita-Yamaguchi
Y, et al. Insulin and insulin-like growth factor-I (IGF-I) receptor overexpression in
breast cancers leads to insulin/IGF-I hybrid receptor overexpression: evidence for a
second mechanism of IGF-I signaling. Clinical Cancer Research. 1999;5:1935-44.
406. Pandini G, Frasca F, Mineo R, Sciacca L, Vigneri R, Belfiore A.
Insulin/insulin-like growth factor I hybrid receptors have different biological
characteristics depending on the insulin receptor isoform involved. Journal of
407. Federici M, Porzio O, Zucaro L, Giovannone B, Borboni P, Marini MA, et al.
Increased abundance of insulin/IGF-I hybrid receptors in adipose tissue from NIDDM
patients. Molecular and Cellular Endocrinology. 1997;135:41-7.
408. Federici M, Zucaro L, Porzio O, Massoud R, Borboni P, Lauro D, et al.
Increased expression of insulin/insulin-like growth factor-I hybrid receptors in
skeletal muscle of noninsulin-dependent diabetes mellitus subjects. Journal of
409. Spampinato D, Pandini G, Iuppa A, Trischitta V, Vigneri R, Frittitta L.
Insulin/insulin-like growth factor I hybrid receptors overexpression is not an early
defect in insulin-resistant subjects. Journal of Clinical Endocrinology and
Metabolism. 2000;85:4219-23.
410. Federici M, Lauro D, D'Adamo M, Giovannone B, Porzio O, Mellozzi M, et
al. Expression of insulin/IGF-I hybrid receptors is increased in skeletal muscle of
patients with chronic primary hyperinsulinemia. Diabetes. 1998;47:87-92.
411. Morrone A, Pegoraro E, Angelini C, Zammarchi E, Marconi G, Hoffman EP.
RNA metabolism in myotonic dystrophy: patient muscle shows decreased insulin
receptor RNA and protein consistent with abnormal insulin resistance. Journal of
412. Pender C, Goldfine ID, Tanner CJ, Pories WJ, MacDonald KG, Havel PJ, et
al. Muscle insulin receptor concentrations in obese patients post bariatric surgery:
relationship to hyperinsulinemia. International Journal of Obesity and Related
Metabolic Disorders. 2004;28:363-9.
413. Wang GS, Cooper TA. Splicing in disease: disruption of the splicing code and
the decoding machinery. Nat Rev Genet. 2007;8:749-61.
414. Lopez-Bigas N, Audit B, Ouzounis C, Parra G, Guigo R. Are splicing
mutations the most frequent cause of hereditary disease? FEBS Letters.
2005;579:1900-3.
244
415. Savkur RS, Philips AV, Cooper TA. Aberrant regulation of insulin receptor
alternative splicing is associated with insulin resistance in myotonic dystrophy.
Nature Genetics. 2001;29:40-7.
416. Philips AV, Timchenko LT, Cooper TA. Disruption of splicing regulated by a
CUG-binding protein in myotonic dystrophy. Science. 1998;280:737-41.
417. Sesti G, Marini MA, Tullio AN, Montemurro A, Borboni P, Fusco A, et al.
Altered expression of the two naturally occurring human insulin receptor variants in
isolated adipocytes of non-insulin-dependent diabetes mellitus patients. Biochemical
and Biophysical Research Communications. 1991;181:1419-24.
418. Sesti G, D'Alfonso R, Vargas Punti MD, Frittitta L, Trischitta V, Liu YY, et
al. Peptide-based radioimmunoassay for the two isoforms of the human insulin
receptor. Diabetologia. 1995;38:445-53.
419. Mosthaf L, Vogt B, Haring HU, Ullrich A. Altered expression of insulin
receptor types A and B in the skeletal muscle of non-insulin-dependent diabetes
mellitus patients. Proceedings of the National Academy of Sciences of the United
420. Kellerer M, Sesti G, Seffer E, Obermaier-Kusser B, Pongratz DE, Mosthaf L,
et al. Altered pattern of insulin receptor isotypes in skeletal muscle membranes of
type 2 (non-insulin-dependent) diabetic subjects. Diab tologia. 1993;36:628-32.
421. Mosthaf L, Eriksson J, Haring HU, Groop L, Widen E, Ullrich A. Insulin
receptor isotype expression correlates with risk of non-insulin-dependent diabetes.
1993;90:2633-5.
422. Haring HU, Kellerer M, Mosthaf L. Modulation of insulin receptor signalling:
significance of altered receptor isoform patterns and mechanism of hyperglycaemiainduced receptor modulation. Diab tologia. 1994;37 Suppl 2:S149-54.
423. Norgren S, Zierath J, Wedell A, Wallberg-Henriksson H, Luthman H.
Regulation of human insulin receptor RNA splicing in vivo. Proceedings of the
National Academy of Sciences of the United States of America. 1994;91:1465-9.
424. Benecke H, Flier JS, Moller DE. Alternatively spliced variants of the insulin
receptor protein. Expression in normal and diabetic human tissues. Journal of Clinical
425. Hansen T, Bjorbaek C, Vestergaard H, Gronskov K, Bak JF, Pedersen O.
Expression of insulin receptor spliced variants and their functional correlates in
muscle from patients with non-insulin-dependent diabetes mellitus. Journal of
Clinical Endocrinology and Metabolism. 1993;77:1500-5.
426. Anderson CM, Henry RR, Knudson PE, Olefsky JM, Webster NJ. Relative
expression of insulin receptor isoforms does not differ in lean, obese, and noninsulin-
245
dependent diabetes mellitus subjects. Journal of Clinical Endocrinology and
Metabolism. 1993;76:1380-2.
427. DePirro R, Lauro R, Testa I, Ferretti I, De Martinis C, Dellatonio R.
Decreased insulin receptors but normal glucose metabolism in Duchenne muscular
dystrophy. Science. 1982;216:311-3.
428. Arcidiacono B, Iiritano S, Nocera A, Possidente K, Nevolo MT, Ventura V, et
al. Insulin resistance and cancer risk: an overview of the pathogenetic mechanisms.
Exp Diabetes Res. 2012;2012:789174.
429. Friedman MN, Sancetta AJ, Magovern GJ. The amelioration of diabetes
mellitus following subtotal gastrectomy. Surgery, Gynecology and Obstetrics.
1955;100:201-4.
430. Dixon JB, le Roux CW, Rubino F, Zimmet P. Bariatric surgery for type 2
diabetes. Lancet. 2012;379:2300-11.
431. Rubino F, Schauer PR, Kaplan LM, Cummings DE. Metabolic surgery to treat
type 2 diabetes: clinical outcomes and mechanisms of action. Annual Review of
Medicine. 2010;61:393-411.
432. Rubino F, R'Bibo S L, del Genio F, Mazumdar M, McGraw TE. Metabolic
surgery: the role of the gastrointestinal tract in diabetes mellitus. Nat Rev Endocrinol.
2010;6:102-9.
433. Adams TD, Gress RE, Smith SC, Halverson RC, Simper SC, Rosamond WD,
et al. Long-term mortality after gastric bypass surgery. New England Journal of
Medicine. 2007;357:753-61.
434. Adams TD, Hunt SC. Cancer and obesity: effect of bariatric surgery. World
Journal of Surgery. 2009;33:2028-33.
435. Sjostrom L, Gummesson A, Sjostrom CD, Narbro K, Peltonen M, Wedel H, et
al. Effects of bariatric surgery on cancer incidence in obese patients in Sweden
(Swedish Obese Subjects Study): a prospective, controlled intervention trial. Lancet
Oncol. 2009;10:653-62.
436. Sjostrom L, Lindroos AK, Peltonen M, Torgerson J, Bouchard C, Carlsson B,
et al. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric
surgery. New England Journal of Medicine. 2004;351:2683-93.
437. Sjostrom L, Narbro K, Sjostrom CD, Karason K, Larsson B, Wedel H, et al.
Effects of bariatric surgery on mortality in Swedish obese subjects. New England
Journal of Medicine. 2007;357:741-52.
438. Sjostrom L, Peltonen M, Jacobson P, Sjostrom CD, Karason K, Wedel H, et
al. Bariatric surgery and long-term cardiovascular events. JAMA. 2012;307:56-65.
246
439. Colquitt JL, Picot J, Loveman E, Clegg AJ. Surgery for obesity. Cochrane
Database Syst Rev. 2009:CD003641.
440. Dixon JB, O'Brien PE, Playfair J, Chapman L, Schachter LM, Skinner S, et al.
Adjustable gastric banding and conventional therapy for type 2 diabetes: a
randomized controlled trial. JAMA. 2008;299:316-23.
441. Mingrone G, Panunzi S, De Gaetano A, Guidone C, Iaconelli A, Leccesi L, et
al. Bariatric surgery versus conventional medical therapy for type 2 diabetes. New
England Journal of Medicine. 2012;366:1577-85.
442. Schauer PR, Kashyap SR, Wolski K, Brethauer SA, Kirwan JP, Pothier CE, et
al. Bariatric surgery versus intensive medical therapy in obese patients with diabetes.
New England Journal of Medicine. 2012;366:1567-76.
443. Chipkin SR, Goldberg RJ. Obesity surgery and diabetes: does a chance to cut
mean a chance to cure? American Journal of Medicine. 2009;122:205-6.
444. Pinkney JH, Johnson AB, Gale EA. The big fat bariatric bandwagon.
445. Buse JB, Caprio S, Cefalu WT, Ceriello A, Del Prato S, Inzucchi SE, et al.
How do we define cure of diabetes? Diabetes Care. 2009;32:2133-5.
446. Pontiroli AE, Pizzocri P, Librenti MC, Vedani P, Marchi M, Cucchi E, et al.
Laparoscopic adjustable gastric banding for the treatment of morbid (grade 3) obesity
and its metabolic complications: a three-year study. Journal of Clinical
447. Schauer PR, Burguera B, Ikramuddin S, Cottam D, Gourash W, Hamad G, et
al. Effect of laparoscopic Roux-en Y gastric bypass on type 2 diabetes mellitus.
Annals of Surgery. 2003;238:467-84; discussion 84-5.
448. Wickremesekera K, Miller G, Naotunne TD, Knowles G, Stubbs RS. Loss of
insulin resistance after Roux-en-Y gastric bypass surgery: a time course study. Obes
Surg. 2005;15:474-81.
449. Rubino F, Marescaux J. Effect of duodenal-jejunal exclusion in a non-obese
animal model of type 2 diabetes: a new perspective for an old disease. Annals of
Surgery. 2004;239:1-11.
450. Kindel TL, Yoder SM, Seeley RJ, D'Alessio DA, Tso P. Duodenal-jejunal
exclusion improves glucose tolerance in the diabetic, Goto-Kakizaki rat by a GLP-1
receptor-mediated mechanism. Journal of Gastrointestinal Surgery. 2009;13:1762-72.
451. Pacheco D, de Luis DA, Romero A, Gonzalez Sagrado M, Conde R, Izaola O,
et al. The effects of duodenal-jejunal exclusion on hormonal regulation of glucose
metabolism in Goto-Kakizaki rats. American Journal of Surgery. 2007;194:221-4.
247
452. Wang TT, Hu SY, Gao HD, Zhang GY, Liu CZ, Feng JB, et al. Ileal
transposition controls diabetes as well as modified duodenal jejunal bypass with better
lipid lowering in a nonobese rat model of type II diabetes by increasing GLP-1.
Annals of Surgery. 2008;247:968-75.
453. Laferrere B, Teixeira J, McGinty J, Tran H, Egger JR, Colarusso A, et al.
Effect of weight loss by gastric bypass surgery versus hypocaloric diet on glucose and
incretin levels in patients with type 2 diabetes. Journal of Clinical Endocrinology and
Metabolism. 2008;93:2479-85.
454. Pournaras DJ, Osborne A, Hawkins SC, Vincent RP, Mahon D, Ewings P, et
al. Remission of type 2 diabetes after gastric bypass and banding: mechanisms and 2
year outcomes. Annals of Surgery. 2010;252:966-71.
455. Reis CE, Alvarez-Leite JI, Bressan J, Alfenas RC. Role of bariatric-metabolic
surgery in the treatment of obese type 2 diabetes with body mass index <35 kg/m2: a
literature review. Diabetes Technol Ther. 2012;14:365-72.
456. Moo T-A, Rubino F. Gastrointestinal surgery as treatment for type 2 diabetes.
Current Opinion in Endocrinology, Diabetes and Obesity. 2008;15.
457. Cummings DE, Overduin J, Foster-Schubert KE, Carlson MJ. Role of the
bypassed proximal intestine in the anti-diabetic effects of bariatric surgery. Surg Obes
Relat Dis. 2007;3:109-15.
458. le Roux CW, Welbourn R, Werling M, Osborne A, Kokkinos A, Laurenius A,
et al. Gut hormones as mediators of appetite and weight loss after Roux-en-Y gastric
bypass. Annals of Surgery. 2007;246:780-5.
459. Laferrere B, Heshka S, Wang K, Khan Y, McGinty J, Teixeira J, et al. Incretin
levels and effect are markedly enhanced 1 month after Roux-en-Y gastric bypass
surgery in obese patients with type 2 diabetes. Diabetes Care. 2007;30:1709-16.
460. le Roux CW, Aylwin SJ, Batterham RL, Borg CM, Coyle F, Prasad V, et al.
Gut hormone profiles following bariatric surgery favor an anorectic state, facilitate
weight loss, and improve metabolic parameters. Annals of Surgery. 2006;243:108-14.
461. Korner J, Inabnet W, Conwell IM, Taveras C, Daud A, Olivero-Rivera L, et al.
Differential effects of gastric bypass and banding on circulating gut hormone and
leptin levels. Obesity (Silver Spring). 2006;14:1553-61.
462. Korner J, Bessler M, Cirilo LJ, Conwell IM, Daud A, Restuccia NL, et al.
Effects of Roux-en-Y gastric bypass surgery on fasting and postprandial
concentrations of plasma ghrelin, peptide YY, and insulin. Journal of Clinical
463. Rubino F, Gagner M, Gentileschi P, Kini S, Fukuyama S, Feng J, et al. The
early effect of the Roux-en-Y gastric bypass on hormones involved in body weight
regulation and glucose metabolism. Annals of Surgery. 2004;240:236-42.
248
464. Cheung GW, Kokorovic A, Lam CK, Chari M, Lam TK. Intestinal
cholecystokinin controls glucose production through a neuronal network. Cell Metab.
2009;10:99-109.
465. Knauf C, Cani PD, Kim DH, Iglesias MA, Chabo C, Waget A, et al. Role of
central nervous system glucagon-like Peptide-1 receptors in enteric glucose sensing.
Diabetes. 2008;57:2603-12.
466. Troy S, Soty M, Ribeiro L, Laval L, Migrenne S, Fioramonti X, et al.
Intestinal gluconeogenesis is a key factor for early metabolic changes after gastric
bypass but not after gastric lap-band in mice. Cell Metab. 2008;8:201-11.
467. Hayes MT, Foo J, Besic V, Tychinskaya Y, Stubbs RS. Is intestinal
gluconeogenesis a key factor in the early changes in glucose homeostasis following
gastric bypass? Obes Surg. 2011;21:759-62.
468. Mari A, Manco M, Guidone C, Nanni G, Castagneto M, Mingrone G, et al.
Restoration of normal glucose tolerance in severely obese patients after biliopancreatic diversion: role of insulin sensitivity and beta cell function. Diabetologia.
2006;49:2136-43.
469. Mingrone G, Castagneto M. Bariatric surgery: unstressing or boosting the
beta-cell? Diabetes Obes Metab. 2009;11 Suppl 4:130-42.
470. Kashyap SR, Daud S, Kelly KR, Gastaldelli A, Win H, Brethauer S, et al.
Acute effects of gastric bypass versus gastric restrictive surgery on beta-cell function
and insulinotropic hormones in severely obese patients with type 2 diabetes. Int J
Obes (Lond). 2010;34:462-71.
471. Salinari S, Bertuzzi A, Asnaghi S, Guidone C, Manco M, Mingrone G. Firstphase insulin secretion restoration and differential response to glucose load depending
on the route of administration in type 2 diabetic subjects after bariatric surgery.
472. Nannipieri M, Mari A, Anselmino M, Baldi S, Barsotti E, Guarino D, et al.
The role of beta-cell function and insulin sensitivity in the remission of type 2
diabetes after gastric bypass surgery. Journal of Clinical Endocrinology and
Metabolism. 2011;96:E1372-9.
473. Lee WJ, Lee YC, Ser KH, Chen JC, Chen SC. Improvement of insulin
resistance after obesity surgery: a comparison of gastric banding and bypass
procedures. Obes Surg. 2008;18:1119-25.
474. Camastra S, Gastaldelli A, Mari A, Bonuccelli S, Scartabelli G, Frascerra S, et
al. Early and longer term effects of gastric bypass surgery on tissue-specific insulin
sensitivity and beta cell function in morbidly obese patients with and without type 2
diabetes. Diabetologia. 2011;54:2093-102.
249
475. Foo J, Krebs J, Hayes MT, Bell D, Macartney-Coxson D, Croft T, et al.
Studies in insulin resistance following very low calorie diet and/or gastric bypass
surgery. Obes Surg. 2011;21:1914-20.
476. Lima MM, Pareja JC, Alegre SM, Geloneze SR, Kahn SE, Astiarraga BD, et
al. Acute effect of roux-en-y gastric bypass on whole-body insulin sensitivity: a study
with the euglycemic-hyperinsulinemic clamp. Journal of Clinical Endocrinology and
Metabolism. 2010;95:3871-5.
477. Campos GM, Rabl C, Peeva S, Ciovica R, Rao M, Schwarz JM, et al.
Improvement in peripheral glucose uptake after gastric bypass surgery is observed
only after substantial weight loss has occurred and correlates with the magnitude of
weight lost. Journal of Gastrointestinal Surgery. 2010;14:15-23.
478. Rees DA, Alcolado JC. Animal models of diabetes mellitus. Diabetic
Medicine. 2005;22:359-70.
479. Tietz Textbook of Clinical Chemistry and Molecular Diagnostics. 5 ed. Burtis
CA, Ashwood ER, Bruns DE, editors. St. Louis, Missouri: Elsevier Saunders; 2006.
480. Falken Y, Hellstrom PM, Holst JJ, Naslund E. Changes in glucose
homeostasis after Roux-en-Y gastric bypass surgery for obesity at day three, two
months, and one year after surgery: role of gut peptides. Journal of Clinical
481. Ballantyne GH, Farkas D, Laker S, Wasielewski A. Short-term changes in
insulin resistance following weight loss surgery for morbid obesity: laparoscopic
adjustable gastric banding versus laparoscopic Roux-en-Y gastric bypass. Obes Surg.
2006;16:1189-97.
482. Bojsen-Møller KN, Dirksen C, Jørgensen NB, Jacobsen SH, Hansen DL,
Worm D, et al. Increased Hepatic Insulin Clearance After Roux-en-Y Gastric Bypass.
Journal of Clinical Endocrinology and Metabolism. 2013;98:E1066-E71.
483. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid
guanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochemistry.
1987;162:156-9.
484. Imbeaud S, Graudens E, Boulanger V, Barlet X, Zaborski P, Eveno E, et al.
Towards standardization of RNA quality assessment using user-independent
classifiers of microcapillary electrophoresis traces. Nucleic Acids Res. 2005;33:e56.
485. Schroeder A, Mueller O, Stocker S, Salowsky R, Leiber M, Gassmann M, et
al. The RIN: an RNA integrity number for assigning integrity values to RNA
measurements. BMC Mol Biol. 2006;7:3.
486. Sharkey DJ, Scalice ER, Christy KG, Jr., Atwood SM, Daiss JL. Antibodies as
thermolabile switches: high temperature triggering for the polymerase chain reaction.
Biotechnology (N Y). 1994;12:506-9.
250
487. Longo MC, Berninger MS, Hartley JL. Use of uracil DNA glycosylase to
control carry-over contamination in polymerase chain reactions. Gene. 1990;93:1258.
488. Holland PM, Abramson RD, Watson R, Gelfand DH. Detection of specific
polymerase chain reaction product by utilizing the 5'----3' exonuclease activity of
Thermus aquaticus DNA polymerase. Proceedings of the National Academy of
489. Afonina I, Zivarts M, Kutyavin I, Lukhtanov E, Gamper H, Meyer RB.
Efficient priming of PCR with short oligonucleotides conjugated to a minor groove
binder. Nucleic Acids Res. 1997;25:2657-60.
490. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using
real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods.
2001;25:402-8.
491. Schmittgen TD, Livak KJ. Analyzing real-time PCR data by the comparative
C(T) method. Nat Protoc. 2008;3:1101-8.
492. Applied-Biosystems. User Bulletin # 2: ABI PRISM 7700 Sequence Detection
System. 1997.
493. Biosystems A. Amplification Efficiency of TaqMan Gene
Expression Assays. Applied Biosystems Application Note. 2006.
494. Mane VP, Heuer MA, Hillyer P, Navarro MB, Rabin RL. Systematic method
for determining an ideal housekeeping gene for real-time PCR analysis. J Biomol
Tech. 2008;19:342-7.
495. Bradford MM. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Analytical
Biochemistry. 1976;72:248-54.
496.
Bio-Rad. Quantity One User Guide 2006.
497. Sakoda H, Ogihara T, Anai M, Funaki M, Inukai K, Katagiri H, et al. No
correlation of plasma cell 1 overexpression with insulin resistance in diabetic rats and
3T3-L1 adipocytes. Diabetes. 1999;48:1365-71.
498. Farris W, Mansourian S, Chang Y, Lindsley L, Eckman EA, Frosch MP, et al.
Insulin-degrading enzyme regulates the levels of insulin, amyloid beta-protein, and
the beta-amyloid precursor protein intracellular domain in vivo. Proceedings of the
499. Maddux BA, Chang YN, Accili D, McGuinness OP, Youngren JF, Goldfine
ID. Overexpression of the insulin receptor inhibitor PC-1/ENPP1 induces insulin
resistance and hyperglycemia. Am J Physiol Endocrinol Metab. 2006;290:E746-9.
251
500. Eller P, Hochegger K, Wehinger A, Tancevski I, Schgoer W, Ritsch A, et al.
Hepatic ENPP1 expression is induced in diabetic rabbits. Mamm Genome.
2006;17:886-91.
501. Spampinato D, Giaccari A, Trischitta V, Costanzo BV, Morviducci L,
Buongiorno A, et al. Rats that are made insulin resistant by glucosamine treatment
have impaired skeletal muscle insulin receptor phosphorylation. Metabolism: Clinical
and Experimental. 2003;52:1092-5.
502. Youngren JF, Paik J, Barnard RJ. Impaired insulin-receptor
autophosphorylation is an early defect in fat-fed, insulin-resistant rats. Journal of
Applied Physiology. 2001;91:2240-7.
503. Menzaghi C, Di Paola R, Baj G, Funaro A, Arnulfo A, Ercolino T, et al.
Insulin modulates PC-1 processing and recruitment in cultured human cells. Am J
Physiol Endocrinol Metab. 2003;284:E514-20.
504. Frittitta L, Sbraccia P, Costanzo BV, Tassi V, D'Adamo M, Spampinato D, et
al. High insulin levels do not influence PC-1 gene expression and protein content in
human muscle tissue and hepatoma cells. Diabetes/Metabolism Research and
Reviews. 2000;16:26-32.
505. Whitehead JP, Humphreys PJ, Dib K, Goding JW, O'Rahilly S. Expression of
the putative inhibitor of the insulin receptor tyrosine kinase PC-1 in dermal fibroblasts
from patients with syndromes of severe insulin resistance. Clinical Endocrinology.
1997;47:65-70.
506. Di Paola R, Caporarello N, Marucci A, Dimatteo C, Iadicicco C, Del Guerra
S, et al. ENPP1 affects insulin action and secretion: evidences from in vitro studies.
PLoS One. 2011;6:e19462.
507. Maranghi M, Prudente S, D'Erasmo L, Morini E, Ciociola E, Coletta P, et al.
The ectonucleotide pyrophosphatase phosphodiesterase 1 (ENPP1) K121Q
polymorphism modulates the beneficial effect of weight loss on fasting glucose in
non-diabetic individuals. Nutr Metab Cardiovasc Dis. 2012.
508. McAteer JB, Prudente S, Bacci S, Lyon HN, Hirschhorn JN, Trischitta V, et
al. The ENPP1 K121Q polymorphism is associated with type 2 diabetes in European
populations: evidence from an updated meta-analysis in 42,042 subjects. Diabetes.
2008;57:1125-30.
509. Wang R, Zhou D, Xi B, Ge X, Zhu P, Wang B, et al. ENPP1/PC-1 gene
K121Q polymorphism is associated with obesity in European adult populations:
evidence from a meta-analysis involving 24,324 subjects. Biomedical and
Environmental Sciences. 2011;24:200-6.
510. Grarup N, Urhammer SA, Ek J, Albrechtsen A, Glumer C, Borch-Johnsen K,
et al. Studies of the relationship between the ENPP1 K121Q polymorphism and type 2
diabetes, insulin resistance and obesity in 7,333 Danish white subjects. Diab tologia.
2006;49:2097-104.
252
511. Zhao T, Liu Z, Zhang D, Liu Y, Yang Y, Zhou D, et al. The ENPP1 K121Q
polymorphism is not associated with type 2 diabetes or obesity in the Chinese Han
population. Journal of Human Genetics. 2011;56:12-6.
512. Lee W. The CEACAM1 expression is decreased in the liver of severely obese
patients with or without diabetes. Diagn Pathol. 2011;6:40.
513. Pivovarova O, Gogebakan O, Pfeiffer AF, Rudovich N. Glucose inhibits the
insulin-induced activation of the insulin-degrading enzyme in HepG2 cells.
514. De Meyts P. The insulin receptor isoform A: a mitogenic proinsulin receptor?
515. Boden G, Chen X, Stein TP. Gluconeogenesis in moderately and severely
hyperglycemic patients with type 2 diabetes mellitus. Am J Physiol Endocrinol
Metab. 2001;280:E23-30.
516. Kraegen EW, Clark PW, Jenkins AB, Daley EA, Chisholm DJ, Storlien LH.
Development of muscle insulin resistance after liver insulin resistance in high-fat-fed
rats. Diabetes. 1991;40:1397-403.
517. Bluher M, Patti ME, Gesta S, Kahn BB, Kahn CR. Intrinsic heterogeneity in
adipose tissue of fat-specific insulin receptor knock-out mice is associated with
differences in patterns of gene expression. Journal of Biological Chemistry.
2004;279:31891-901.
518. Norgren S, Zierath J, Galuska D, Wallberg-Henriksson H, Luthman H.
Differences in the ratio of RNA encoding two isoforms of the insulin receptor
between control and NIDDM patients. The RNA variant without Exon 11
predominates in both groups. Diabetes. 1993;42:675-81.
519. Sakurai K, Chomchan P, Rossi JJ. Silencing of gene expression in cultured
cells using small interfering RNAs. Curr Protoc Cell Biol. 2010;Chapter 27:Unit 27 1
1-8.
520.
Whither RNAi? Nat Cell Biol. 2003;5:489-90.
521. Invitrogen. Transfecting Stealth™ RNAi or siRNA into HepG2 Cells Using
Lipofectamine™ RNAiMAX 2006.
522. Birnboim HC, Doly J. A rapid alkaline extraction procedure for screening
recombinant plasmid DNA. Nucleic Acids Res. 1979;7:1513-23.
523. Kozak M. Point mutations define a sequence flanking the AUG initiator codon
that modulates translation by eukaryotic ribosomes. Cell. 1986;44:283-92.
524. Kozak M. An analysis of 5'-noncoding sequences from 699 vertebrate
messenger RNAs. Nucleic Acids Res. 1987;15:8125-48.
253
525. Don RH, Cox PT, Wainwright BJ, Baker K, Mattick JS. 'Touchdown' PCR to
circumvent spurious priming during gene amplification. Nucleic Acids Res.
1991;19:4008.
526. Breslauer KJ, Frank R, Blocker H, Marky LA. Predicting DNA duplex
stability from the base sequence. Proceedings of the National Academy of Sciences of
527. Ehrhardt C, Schmolke M, Matzke A, Knoblauch A, Will C, Wixler V, et al.
Polyethylenimine, a cost-effective transfection reagent. Signal Transduction.
2006;6:179-84.
528. Boussif O, Lezoualc'h F, Zanta MA, Mergny MD, Scherman D, Demeneix B,
et al. A versatile vector for gene and oligonucleotide transfer into cells in culture and
in vivo: polyethylenimine. Proceedings of the National Academy of Sciences of the
United States of America. 1995;92:7297-301.
529. Sonawane ND, Szoka FC, Jr., Verkman AS. Chloride accumulation and
swelling in endosomes enhances DNA transfer by polyamine-DNA polyplexes.
530. Knutson VP, Ronnett GV, Lane MD. Rapid, reversible internalization of cell
surface insulin receptors. Correlation with insulin-induced down-regulation. Journal
of Biological Chemistry. 1983;258:12139-42.
531. Norgren S, Li LS, Luthman H. Regulation of human insulin receptor RNA
splicing in HepG2 cells: effects of glucocorticoid and low glucose concentration.
Biochemical and Biophysical Research Communications. 1994;199:277-84.
532. Sell SM, Reese D, Ossowski VM. Insulin-inducible changes in insulin
receptor mRNA splice variants. Journal of Biological Chemistry. 1994;269:30769-72.
533. Sen S, Talukdar I, Webster NJ. SRp20 and CUG-BP1 modulate insulin
receptor exon 11 alternative splicing. Molecular and Cellular Biology. 2009;29:87180.
534. Jensen M, De Meyts P. Molecular mechanisms of differential intracellular
signaling from the insulin receptor. Vitamins and Hormones. 2009;80:51-75.
535. Zhang Y, Chen W, Li R, Li Y, Ge Y, Chen G. Insulin-regulated Srebp-1c and
Pck1 mRNA expression in primary hepatocytes from zucker fatty but not lean rats is
affected by feeding conditions. PLoS One. 2011;6:e21342.
536. Gómez-Valadés AG, Méndez-Lucas A, Vidal-Alabró A, Blasco FX, Chillon
M, Bartrons R, et al. Pck1 Gene Silencing in the Liver Improves Glycemia Control,
Insulin Sensitivity, and Dyslipidemia in db/db Mice. Diabetes. 2008;57:2199-210.
537. Newgard CB, Attie AD. Getting biological about the genetics of diabetes. Nat
Med. 2010;16:388-91.
254
538. Hoheisel JD. Microarray technology: beyond transcript profiling and genotype
analysis. Nat Rev Genet. 2006;7:200-10.
539. Schulze A, Downward J. Navigating gene expression using microarrays--a
technology review. Nat Cell Biol. 2001;3:E190-5.
540. Nojima K, Sugimoto K, Ueda H, Babaya N, Ikegami H, Rakugi H. Analysis of
hepatic gene expression profile in a spontaneous mouse model of type 2 diabetes
under a high sucrose diet. Endocrine Journal. 2012.
541. Lee RK, Hittel DS, Nyamandi VZ, Kang L, Soh J, Sensen CW, et al.
Unconventional microarray design reveals the response to obesity is largely tissue
specific: analysis of common and divergent responses to diet-induced obesity in
insulin-sensitive tissues. Appl Physiol Nutr Metab. 2012;37:257-68.
542. Xu Y, Ramos EJ, Middleton F, Romanova I, Quinn R, Chen C, et al. Gene
expression profiles post Roux-en-Y gastric bypass. Surgery. 2004;136:246-52.
543. Zhou H, Saito S, Piao G, Liu ZP, Wang J, Horimoto K, et al. Network
screening of Goto-Kakizaki rat liver microarray data during diabetic progression.
BMC Syst Biol. 2011;5 Suppl 1:S16.
544. Tamboli RA, Hajri T, Jiang A, Marks-Shulman PA, Williams DB, Clements
RH, et al. Reduction in inflammatory gene expression in skeletal muscle from Rouxen-Y gastric bypass patients randomized to omentectomy. PLoS One. 2011;6:e28577.
545. Elam MB, Cowan GS, Jr., Rooney RJ, Hiler ML, Yellaturu CR, Deng X, et al.
Hepatic gene expression in morbidly obese women: implications for disease
susceptibility. Obesity (Silver Spring). 2009;17:1563-73.
546. MacLaren R, Cui W, Simard S, Cianflone K. Influence of obesity and insulin
sensitivity on insulin signaling genes in human omental and subcutaneous adipose
tissue. Journal of Lipid Research. 2008;49:308-23.
547. Keller MP, Attie AD. Physiological insights gained from gene expression
analysis in obesity and diabetes. Annual Review of Nutrition. 2010;30:341-64.
548. Coletta DK, Balas B, Chavez AO, Baig M, Abdul-Ghani M, Kashyap SR, et
al. Effect of acute physiological hyperinsulinemia on gene expression in human
skeletal muscle in vivo. Am J Physiol Endocrinol Metab. 2008;294:E910-7.
549. Kacharmina JE, Crino PB, Eberwine J. Preparation of cDNA from single cells
and subcellular regions. Methods in Enzymology. 1999;303:3-18.
550. Pabon C, Modrusan Z, Ruvolo MV, Coleman IM, Daniel S, Yue H, et al.
Optimized T7 amplification system for microarray analysis. Biotechniques.
2001;31:874-9.
255
551. Van Gelder RN, von Zastrow ME, Yool A, Dement WC, Barchas JD,
Eberwine JH. Amplified RNA synthesized from limited quantities of heterogeneous
cDNA. Proceedings of the National Academy of Sciences of the United States of
America. 1990;87:1663-7.
552. Kuhn K, Baker SC, Chudin E, Lieu MH, Oeser S, Bennett H, et al. A novel,
high-performance random array platform for quantitative gene expression profiling.
Genome Research. 2004;14:2347-56.
553. Gunderson KL, Kruglyak S, Graige MS, Garcia F, Kermani BG, Zhao C, et al.
Decoding randomly ordered DNA arrays. Genome Research. 2004;14:870-7.
554. Dunning MJ, Barbosa-Morais NL, Lynch AG, Tavare S, Ritchie ME.
Statistical issues in the analysis of Illumina data. BMC Bioinformatics. 2008;9:85.
555. Bolstad BM, Irizarry RA, Astrand M, Speed TP. A comparison of
normalization methods for high density oligonucleotide array data based on variance
and bias. Bioinformatics. 2003;19:185-93.
556. Diehn M, Sherlock G, Binkley G, Jin H, Matese JC, Hernandez-Boussard T, et
al. SOURCE: a unified genomic resource of functional annotations, ontologies, and
gene expression data. Nucleic Acids Res. 2003;31:219-23.
557. Benson DA, Karsch-Mizrachi I, Lipman DJ, Ostell J, Rapp BA, Wheeler DL.
GenBank. Nucleic Acids Res. 2002;30:17-20.
558. Gasteiger E, Jung E, Bairoch A. SWISS-PROT: connecting biomolecular
knowledge via a protein database. Curr Issues Mol Biol. 2001;3:47-55.
559. Hamosh A, Scott AF, Amberger J, Bocchini C, Valle D, McKusick VA.
Online Mendelian Inheritance in Man (OMIM), a knowledgebase of human genes and
genetic disorders. Nucleic Acids Res. 2002;30:52-5.
560. Wheeler DL, Church DM, Lash AE, Leipe DD, Madden TL, Pontius JU, et al.
Database resources of the National Center for Biotechnology Information: 2002
update. Nucleic Acids Res. 2002;30:13-6.
561. Povey S, Lovering R, Bruford E, Wright M, Lush M, Wain H. The HUGO
Gene Nomenclature Committee (HGNC). Human Genetics. 2001;109:678-80.
562. Korn EL, Troendle JF, McShane LM, Simon R. Controlling the number of
false discoveries: application to high-dimensional genomic data. Journal of Statistical
Planning and Inference. 2004;124:379-98.
563. Simon R, Korn E, McShane L, Radmacher M, Wright G, Zhao Y. Design and
Analysis of DNA Microarray Investigations. 2003.
564. Wright GW, Simon RM. A random variance model for detection of
differential gene expression in small microarray experiments. Bioinformatics.
2003;19:2448-55.
256
565.
Simon R. BRB-ArrayTools User’s Manual. 2010.
566. Verducci JS, Melfi VF, Lin S, Wang Z, Roy S, Sen CK. Microarray analysis
of gene expression: considerations in data mining and statistical treatment.
Physiological Genomics. 2006;25:355-63.
567. Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical
and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society
Series B (Methodological). 1995;57:289-300.
568. Matt X, Jon S, Patricia H, Xiaoqing Y, Kathy L, Criss W, et al. DataAssist™ –
Data Analysis Software for TaqMan® Real-Time PCR Data. Proceedings of the
International MultiConference of Engineers and Computer Scientists. 2010;1.
569. Pfaffl MW, Horgan GW, Dempfle L. Relative expression software tool
(REST) for group-wise comparison and statistical analysis of relative expression
results in real-time PCR. Nucleic Acids Res. 2002;30:e36.
570. Hashimoto O, Tsuchida K, Ushiro Y, Hosoi Y, Hoshi N, Sugino H, et al.
cDNA cloning and expression of human activin betaE subunit. Molecular and
Cellular Endocrinology. 2002;194:117-22.
571. Bertolino P, Holmberg R, Reissmann E, Andersson O, Berggren PO, Ibanez
CF. Activin B receptor ALK7 is a negative regulator of pancreatic beta-cell function.
2008;105:7246-51.
572. Hashimoto O, Sekiyama K, Matsuo T, Hasegawa Y. Implication of activin E
in glucose metabolism: transcriptional regulation of the inhibin/activin betaE subunit
gene in the liver. Life Sciences. 2009;85:534-40.
573. Hashimoto O, Ushiro Y, Sekiyama K, Yamaguchi O, Yoshioka K, Mutoh K-I,
et al. Impaired growth of pancreatic exocrine cells in transgenic mice expressing
human activin βE subunit. Biochemical and Biophysical Research Communications.
2006;341:416-24.
574. Wheatcroft SB, Kearney MT. IGF-dependent and IGF-independent actions of
IGF-binding protein-1 and -2: implications for metabolic homeostasis. Trends
Endocrinol Metab. 2009;20:153-62.
575. Li Z, Martin J, Poirier P, Caron-Cantin SM, Hould FS, Marceau S, et al.
Upregulation of plasma insulin-like growth factor binding protein 2 levels after
biliopancreatic diversion in humans. Obesity (Silver Spring). 2012;20:1469-73.
576. Hedbacker K, Birsoy K, Wysocki RW, Asilmaz E, Ahima RS, Farooqi IS, et
al. Antidiabetic effects of IGFBP2, a leptin-regulated gene. Cell Metab. 2010;11:1122.
257
577. Wheatcroft SB, Kearney MT, Shah AM, Ezzat VA, Miell JR, Modo M, et al.
IGF-binding protein-2 protects against the development of obesity and insulin
resistance. Diabetes. 2007;56:285-94.
578. Chen K, Li F, Li J, Cai H, Strom S, Bisello A, et al. Induction of leptin
resistance through direct interaction of C-reactive protein with leptin. Nature
Medicine. 2006;12:425-32.
579. Ridderstrale M, Groop L. Genetic dissection of type 2 diabetes. Molecular and
Cellular Endocrinology. 2009;297:10-7.
580. Florez JC. Clinical review: the genetics of type 2 diabetes: a realistic appraisal
in 2008. Journal of Clinical Endocrinology and Metabolism. 2008;93:4633-42.
581. Cauchi S, Meyre D, Durand E, Proenca C, Marre M, Hadjadj S, et al. Post
genome-wide association studies of novel genes associated with type 2 diabetes show
gene-gene interaction and high predictive value. PLoS One. 2008;3:e2031.
582. Hertel JK, Johansson S, Raeder H, Midthjell K, Lyssenko V, Groop L, et al.
Genetic analysis of recently identified type 2 diabetes loci in 1,638 unselected patients
with type 2 diabetes and 1,858 control participants from a Norwegian populationbased cohort (the HUNT study). Diabetologia. 2008;51:971-7.
583. Ebara S, Toyoshima S, Matsumura T, Adachi S, Takenaka S, Yamaji R, et al.
Cobalamin deficiency results in severe metabolic disorder of serine and threonine in
rats. Biochimica et Biophysica Acta. 2001;1568:111-7.
584. Provenzale D, Reinhold RB, Golner B, Irwin V, Dallal GE,
Papathanasopoulos N, et al. Evidence for diminished B12 absorption after gastric
bypass: oral supplementation does not prevent low plasma B12 levels in bypass
patients. Journal of the American College of Nutrition. 1992;11:29-35.
585. Smith CD, Herkes SB, Behrns KE, Fairbanks VF, Kelly KA, Sarr MG. Gastric
acid secretion and vitamin B12 absorption after vertical Roux-en-Y gastric bypass for
morbid obesity. Annals of Surgery. 1993;218:91-6.
586. Donadelli SP, Junqueira-Franco MV, de Mattos Donadelli CA, Salgado W, Jr.,
Ceneviva R, Marchini JS, et al. Daily vitamin supplementation and hypovitaminosis
after obesity surgery. Nutrition. 2012;28:391-6.
587. Scott-Drechsel DE, Rugonyi S, Marks DL, Thornburg KL, Hinds MT.
Hyperglycemia slows embryonic growth and suppresses cell cycle via cyclin D1 and
p21. Diabetes. 2013;62:234-42.
588. Banerjee M, Saxena M. Interleukin-1 (IL-1) family of cytokines: role in type 2
diabetes. Clinica Chimica Acta. 2012;413:1163-70.
589. Granowitz EV, Clark BD, Mancilla J, Dinarello CA. Interleukin-1 receptor
antagonist competitively inhibits the binding of interleukin-1 to the type II
interleukin-1 receptor. Journal of Biological Chemistry. 1991;266:14147-50.
258
590. Dripps DJ, Brandhuber BJ, Thompson RC, Eisenberg SP. Interleukin-1 (IL-1)
receptor antagonist binds to the 80-kDa IL-1 receptor but does not initiate IL-1 signal
transduction. Journal of Biological Chemistry. 1991;266:10331-6.
591. McIntyre KW, Stepan GJ, Kolinsky KD, Benjamin WR, Plocinski JM, Kaffka
KL, et al. Inhibition of interleukin 1 (IL-1) binding and bioactivity in vitro and
modulation of acute inflammation in vivo by IL-1 receptor antagonist and anti-IL-1
receptor monoclonal antibody. Journal of Experimental Medicine. 1991;173:931-9.
592. Meier CA, Bobbioni E, Gabay C, Assimacopoulos-Jeannet F, Golay A, Dayer
JM. IL-1 receptor antagonist serum levels are increased in human obesity: a possible
link to the resistance to leptin? Journal of Clinical Endocrinology and Metabolism.
2002;87:1184-8.
593. Somm E, Cettour-Rose P, Asensio C, Charollais A, Klein M, TheanderCarrillo C, et al. Interleukin-1 receptor antagonist is upregulated during diet-induced
obesity and regulates insulin sensitivity in rodents. Diabetologia. 2006;49:387-93.
594. Viana EC, Araujo-Dasilio KL, Miguel GP, Bressan J, Lemos EM, Moyses
MR, et al. Gastric Bypass and Sleeve Gastrectomy: the Same Impact on IL-6 and
TNF-alpha. Prospective Clinical Trial. Obes Surg. 2013;23:1252-61.
595. Illan-Gomez F, Gonzalvez-Ortega M, Orea-Soler I, Alcaraz-Tafalla MS,
Aragon-Alonso A, Pascual-Diaz M, et al. Obesity and inflammation: change in
adiponectin, C-reactive protein, tumour necrosis factor-alpha and interleukin-6 after
bariatric surgery. Obes Surg. 2012;22:950-5.
596. Cullinan EB, Kwee L, Nunes P, Shuster DJ, Ju G, McIntyre KW, et al. IL-1
receptor accessory protein is an essential component of the IL-1 receptor. Journal of
Immunology. 1998;161:5614-20.
597. Wesche H, Korherr C, Kracht M, Falk W, Resch K, Martin MU. The
interleukin-1 receptor accessory protein (IL-1RAcP) is essential for IL-1-induced
activation of interleukin-1 receptor-associated kinase (IRAK) and stress-activated
protein kinases (SAP kinases). Journal of Biological Chemistry. 1997;272:7727-31.
598. Feve B, Bastard JP. The role of interleukins in insulin resistance and type 2
diabetes mellitus. Nat Rev Endocrinol. 2009;5:305-11.
599. Hansen L, Gaster M, Oakeley EJ, Brusgaard K, Damsgaard Nielsen EM,
Beck-Nielsen H, et al. Expression profiling of insulin action in human myotubes:
induction of inflammatory and pro-angiogenic pathways in relationship with glycogen
synthesis and type 2 diabetes. Biochemical and Biophysical Research
Communications. 2004;323:685-95.
600. Rome S, Clement K, Rabasa-Lhoret R, Loizon E, Poitou C, Barsh GS, et al.
Microarray profiling of human skeletal muscle reveals that insulin regulates
approximately 800 genes during a hyperinsulinemic clamp. Journal of Biological
Chemistry. 2003;278:18063-8.
259
601. Sreekumar R, Halvatsiotis P, Schimke JC, Nair KS. Gene expression profile in
skeletal muscle of type 2 diabetes and the effect of insulin treatment. Diabetes.
2002;51:1913-20.
602. Ducluzeau PH, Perretti N, Laville M, Andreelli F, Vega N, Riou JP, et al.
Regulation by insulin of gene expression in human skeletal muscle and adipose tissue.
Evidence for specific defects in type 2 diabetes. Diabetes. 2001;50:1134-42.
603. Allen-Jennings AE, Hartman MG, Kociba GJ, Hai T. The roles of ATF3 in
glucose homeostasis. A transgenic mouse model with liver dysfunction and defects in
endocrine pancreas. Journal of Biological Chemistry. 2001;276:29507-14.
604. Yamada K, Miyamoto K. Basic helix-loop-helix transcription factors,
BHLHB2 and BHLHB3; their gene expressions are regulated by multiple
extracellular stimuli. Frontiers in Bioscience. 2005;10:3151-71.
605. Yamada K, Kawata H, Shou Z, Mizutani T, Noguchi T, Miyamoto K. Insulin
induces the expression of the SHARP-2/Stra13/DEC1 gene via a phosphoinositide 3kinase pathway. Journal of Biological Chemistry. 2003;278:30719-24.
606. Iizuka K, Horikawa Y. Regulation of lipogenesis via BHLHB2/DEC1 and
ChREBP feedback looping. Biochemical and Biophysical Research Communications.
2008;374:95-100.
607. Nakashima A, Kawamoto T, Honda KK, Ueshima T, Noshiro M, Iwata T, et
al. DEC1 modulates the circadian phase of clock gene expression. Molecular and
Cellular Biology. 2008;28:4080-92.
608. Zhang EE, Liu Y, Dentin R, Pongsawakul PY, Liu AC, Hirota T, et al.
Cryptochrome mediates circadian regulation of cAMP signaling and hepatic
gluconeogenesis. Nature Medicine. 2010;16:1152-6.
609. Lamia KA, Storch KF, Weitz CJ. Physiological significance of a peripheral
tissue circadian clock. Proceedings of the National Academy of Sciences of the United
610. Rudic RD, McNamara P, Curtis AM, Boston RC, Panda S, Hogenesch JB, et
al. BMAL1 and CLOCK, two essential components of the circadian clock, are
involved in glucose homeostasis. PLoS Biol. 2004;2:e377.
611. Shi SQ, Ansari TS, McGuinness OP, Wasserman DH, Johnson CH. Circadian
disruption leads to insulin resistance and obesity. Current Biology. 2013;23:372-81.
612. Buxton OM, Cain SW, O'Connor SP, Porter JH, Duffy JF, Wang W, et al.
Adverse metabolic consequences in humans of prolonged sleep restriction combined
with circadian disruption. Sci Transl Med. 2012;4:129ra43.
613. Green CB, Takahashi JS, Bass J. The meter of metabolism. Cell.
2008;134:728-42.
260
614. Munoz A, Serrano C, Garcia-Estan J, Quesada T, Miras Portugal MT. Effect
of diabetic hyperglycemia and other sugars on plasma dopamine-beta-hydroxylase
activity. Diabetes. 1984;33:1127-32.
615. Scheja L, Heese B, Zitzer H, Michael MD, Siesky AM, Pospisil H, et al.
Acute-phase serum amyloid A as a marker of insulin resistance in mice. Exp Diabetes
Res. 2008;2008:230837.
616. Huflejt ME, Leffler H. Galectin-4 in normal tissues and cancer.
Glycoconjugate Journal. 2004;20:247-55.
617. Paclik D, Danese S, Berndt U, Wiedenmann B, Dignass A, Sturm A. Galectin4 controls intestinal inflammation by selective regulation of peripheral and mucosal T
cell apoptosis and cell cycle. PLoS One. 2008;3:e2629.
618. Stowell SR, Arthur CM, Dias-Baruffi M, Rodrigues LC, Gourdine JP,
Heimburg-Molinaro J, et al. Innate immune lectins kill bacteria expressing blood
group antigen. Nature Medicine. 2010;16:295-301.
619. Liu FT, Bevins CL. A sweet target for innate immunity. Nature Medicine.
2010;16:263-4.
620. Montoliu I, Genick U, Ledda M, Collino S, Martin FP, le Coutre J, et al.
Current status on genome-metabolome-wide associations: an opportunity in nutrition
research. Genes Nutr. 2013;8:19-27.
621. Harris K, Kassis A, Major G, #232, ve, Chou CJ. Is the Gut Microbiota a New
Factor Contributing to Obesity and Its Metabolic Disorders? Journal of Obesity.
2012;2012:14.
622. Tremaroli V, Backhed F. Functional interactions between the gut microbiota
and host metabolism. Nature. 2012;489:242-9.
623. Sweeney TE, Morton JM. The Human Gut Microbiome: A Review of the
Effect of Obesity and Surgically Induced Weight Loss. JAMA Surg. 2013:1-7.
624. Wang G, Li B, Hao Y, Zhi J, He C, Xu C. Correlation analysis between gene
expression profile of high-fat emulsion-induced non-alcoholic fatty liver and liver
regeneration in rat. Cell Biology International. 2013;37:917-28.
625. Xu C, Wang G, Hao Y, Zhi J, Zhang L, Chang C. Correlation analysis
between gene expression profile of rat liver tissues and high-fat emulsion-induced
nonalcoholic fatty liver. Digestive Diseases and Sciences. 2011;56:2299-308.
626. Greco D, Kotronen A, Westerbacka J, Puig O, Arkkila P, Kiviluoto T, et al.
Gene expression in human NAFLD. Am J Physiol Gastrointest Liver Physiol.
2008;294:G1281-7.
261
627. Berdeaux R. Metabolic regulation by salt inducible kinases. Frontiers in
Biology. 2011;6:231-41.
628. Dentin R, Hedrick S, Xie J, Yates J, 3rd, Montminy M. Hepatic glucose
sensing via the CREB coactivator CRTC2. Science. 2008;319:1402-5.
629. Koo SH, Flechner L, Qi L, Zhang X, Screaton RA, Jeffries S, et al. The CREB
coactivator TORC2 is a key regulator of fasting glucose metabolism. Nature.
2005;437:1109-11.
630. Yoon YS, Seo WY, Lee MW, Kim ST, Koo SH. Salt-inducible kinase
regulates hepatic lipogenesis by controlling SREBP-1c phosphorylation. Journal of
631. O'Brien RM, Streeper RS, Ayala JE, Stadelmaier BT, Hornbuckle LA. Insulinregulated gene expression. Biochemical Society Transactions. 2001;29:552-8.
632. Chu DT, Davis CM, Chrapkiewicz NB, Granner DK. Reciprocal regulation of
gene transcription by insulin. Inhibition of the phosphoenolpyruvate carboxykinase
gene and stimulation of gene 33 in a single cell type. Journal of Biological Chemistry.
1988;263:13007-11.
633. Granner D, Andreone T, Sasaki K, Beale E. Inhibition of transcription of the
phosphoenolpyruvate carboxykinase gene by insulin. Nature. 1983;305:549-51.
634. Quinn PG, Yeagley D. Insulin regulation of PEPCK gene expression: a model
for rapid and reversible modulation. Curr Drug Targets Immune Endocr Metabol
Disord. 2005;5:423-37.
635. Minokoshi Y, Kim YB, Peroni OD, Fryer LG, Muller C, Carling D, et al.
Leptin stimulates fatty-acid oxidation by activating AMP-activated protein kinase.
Nature. 2002;415:339-43.
636. Aiston S, Agius L. Leptin enhances glycogen storage in hepatocytes by
inhibition of phosphorylase and exerts an additive effect with insulin. Diabetes.
1999;48:15-20.
637. Liao WL, Chen CC, Chang CT, Wu JY, Chen CH, Huang YC, et al. Gene
polymorphisms of adiponectin and leptin receptor are associated with early onset of
type 2 diabetes mellitus in the Taiwanese population. Int J Obes (Lond). 2012;36:7906.
638. Jeon JP, Shim SM, Nam HY, Ryu GM, Hong EJ, Kim HL, et al. Copy number
variation at leptin receptor gene locus associated with metabolic traits and the risk of
type 2 diabetes mellitus. BMC Genomics. 2010;11:426.
639. Park KS, Shin HD, Park BL, Cheong HS, Cho YM, Lee HK, et al.
Polymorphisms in the leptin receptor (LEPR)--putative association with obesity and
T2DM. Journal of Human Genetics. 2006;51:85-91.
262
640. Chen H, Charlat O, Tartaglia LA, Woolf EA, Weng X, Ellis SJ, et al.
Evidence that the diabetes gene encodes the leptin receptor: identification of a
mutation in the leptin receptor gene in db/db mice. Cell. 1996;84:491-5.
641. Cohen P, Yang G, Yu X, Soukas AA, Wolfish CS, Friedman JM, et al.
Induction of Leptin Receptor Expression in the Liver by Leptin and Food
Deprivation. Journal of Biological Chemistry. 2005;280:10034-9.
642. Yang G, Ge H, Boucher A, Yu X, Li C. Modulation of direct leptin signaling
by soluble leptin receptor. Molecular Endocrinology. 2004;18:1354-62.
643. Trakhtenbroit MA, Leichman JG, Algahim MF, Miller CC, 3rd, Moody FG,
Lux TR, et al. Body weight, insulin resistance, and serum adipokine levels 2 years
after 2 types of bariatric surgery. American Journal of Medicine. 2009;122:435-42.
644. Ballantyne GH, Gumbs A, Modlin IM. Changes in insulin resistance following
bariatric surgery and the adipoinsular axis: role of the adipocytokines, leptin,
adiponectin and resistin. Obes Surg. 2005;15:692-9.
645. Alagna S, Cossu ML, Masala A, Atzeni MM, Ruggiu M, Satta FM, et al.
Evaluation of serum leptin levels and thyroid function in morbidly obese patients
treated with bariatric surgery. Eat Weight Disord. 2003;8:95-9.
646. Geloneze B, Tambascia MA, Pareja JC, Repetto EM, Magna LA, Pereira SG.
Serum leptin levels after bariatric surgery across a range of glucose tolerance from
normal to diabetes. Obes Surg. 2001;11:693-8.
647. Rosenbaum M, Nicolson M, Hirsch J, Heymsfield SB, Gallagher D, Chu F, et
al. Effects of gender, body composition, and menopause on plasma concentrations of
leptin. Journal of Clinical Endocrinology and Metabolism. 1996;81:3424-7.
648. Considine RV, Sinha MK, Heiman ML, Kriauciunas A, Stephens TW, Nyce
MR, et al. Serum immunoreactive-leptin concentrations in normal-weight and obese
humans. New England Journal of Medicine. 1996;334:292-5.
649. Frederich RC, Hamann A, Anderson S, Lollmann B, Lowell BB, Flier JS.
Leptin levels reflect body lipid content in mice: evidence for diet-induced resistance
to leptin action. Nature Medicine. 1995;1:1311-4.
650. Yadav A, Kataria MA, Saini V. Role of leptin and adiponectin in insulin
resistance. Clinica Chimica Acta. 2013;417:80-4.
651. Anty R, Bekri S, Luciani N, Saint-Paul MC, Dahman M, Iannelli A, et al. The
inflammatory C-reactive protein is increased in both liver and adipose tissue in
severely obese patients independently from metabolic syndrome, Type 2 diabetes, and
NASH. American Journal of Gastroenterology. 2006;101:1824-33.
652. Rubio-Martin E, Soriguer F, Gutierrez-Repiso C, Garrido-Sanchez L, de
Adana MS, Garcia-Fuentes E, et al. C-reactive protein and incidence of type 2
263
diabetes in the Pizarra study. European Journal of Clinical Investigation.
2013;43:159-67.
653. Wang X, Bao W, Liu J, OuYang Y-Y, Wang D, Rong S, et al. Inflammatory
Markers and Risk of Type 2 Diabetes: A systematic review and meta-analysis.
654. Lee CC, Adler AI, Sandhu MS, Sharp SJ, Forouhi NG, Erqou S, et al.
Association of C-reactive protein with type 2 diabetes: prospective analysis and metaanalysis. Diabetologia. 2009;52:1040-7.
655. Odegaard JI, Chawla A. Pleiotropic Actions of Insulin Resistance and
Inflammation in Metabolic Homeostasis. Science. 2013;339:172-7.
656. Chalasani N, Younossi Z, Lavine JE, Diehl AM, Brunt EM, Cusi K, et al. The
diagnosis and management of non-alcoholic fatty liver disease: practice guideline by
the American Gastroenterological Association, American Association for the Study of
Liver Diseases, and American College of Gastroenterology. Gastroenterology.
2012;142:1592-609.
657. Dixon JB, Bhathal PS, Hughes NR, O'Brien PE. Nonalcoholic fatty liver
disease: Improvement in liver histological analysis with weight loss. Hepatology.
2004;39:1647-54.
658. Furuya CK, Jr., de Oliveira CP, de Mello ES, Faintuch J, Raskovski A,
Matsuda M, et al. Effects of bariatric surgery on nonalcoholic fatty liver disease:
preliminary findings after 2 years. Journal of Gastroenterology and Hepatology.
2007;22:510-4.
659. Jeon H, Blacklow SC. Structure and physiologic function of the low-density
lipoprotein receptor. Annual Review of Biochemistry. 2005;74:535-62.
660. Streicher R, Kotzka J, Muller-Wieland D, Siemeister G, Munck M, Avci H, et
al. SREBP-1 mediates activation of the low density lipoprotein receptor promoter by
insulin and insulin-like growth factor-I. Journal of Biological Chemistry.
1996;271:7128-33.
661. Sekar N, Veldhuis JD. Involvement of Sp1 and SREBP-1a in transcriptional
activation of the LDL receptor gene by insulin and LH in cultured porcine granulosaluteal cells. Am J Physiol Endocrinol Metab. 2004;287:E128-35.
662. Ramakrishnan G, Arjuman A, Suneja S, Das C, Chandra NC. The association
between insulin and low-density lipoprotein receptors. Diab Vasc Dis Res.
2012;9:196-204.
663. Maier T, Jenni S, Ban N. Architecture of mammalian fatty acid synthase at 4.5
A resolution. Science. 2006;311:1258-62.
264
664. Magana MM, Osborne TF. Two tandem binding sites for sterol regulatory
element binding proteins are required for sterol regulation of fatty-acid synthase
promoter. J Biol Chem. 1996;271:32689-94.
665. Hansmannel F, Mordier S, Iynedjian PB. Insulin induction of glucokinase and
fatty acid synthase in hepatocytes: analysis of the roles of sterol-regulatory-elementbinding protein-1c and liver X receptor. Biochemical Journal. 2006;399:275-83.
666. Latasa MJ, Moon YS, Kim KH, Sul HS. Nutritional regulation of the fatty
acid synthase promoter in vivo: sterol regulatory element binding protein functions
through an upstream region containing a sterol regulatory element. Proceedings of the
667. Berndt J, Kovacs P, Ruschke K, Kloting N, Fasshauer M, Schon MR, et al.
Fatty acid synthase gene expression in human adipose tissue: association with obesity
and type 2 diabetes. Diabetologia. 2007;50:1472-80.
668. Dorn C, Riener MO, Kirovski G, Saugspier M, Steib K, Weiss TS, et al.
Expression of fatty acid synthase in nonalcoholic fatty liver disease. Int J Clin Exp
Pathol. 2010;3:505-14.
669. Brasaemle DL, Barber T, Wolins NE, Serrero G, Blanchette-Mackie EJ,
Londos C. Adipose differentiation-related protein is an ubiquitously expressed lipid
storage droplet-associated protein. Journal of Lipid Research. 1997;38:2249-63.
670. Heid HW, Moll R, Schwetlick I, Rackwitz HR, Keenan TW. Adipophilin is a
specific marker of lipid accumulation in diverse cell types and diseases. Cell and
Tissue Research. 1998;294:309-21.
671. Targett-Adams P, Chambers D, Gledhill S, Hope RG, Coy JF, Girod A, et al.
Live cell analysis and targeting of the lipid droplet-binding adipocyte differentiationrelated protein. Journal of Biological Chemistry. 2003;278:15998-6007.
672. Ducharme NA, Bickel PE. Lipid droplets in lipogenesis and lipolysis.
673. Thiele C, Spandl J. Cell biology of lipid droplets. Current Opinion in Cell
Biology. 2008;20:378-85.
674. Greenberg AS, Coleman RA, Kraemer FB, McManaman JL, Obin MS, Puri
V, et al. The role of lipid droplets in metabolic disease in rodents and humans.
675. Okumura T. Role of lipid droplet proteins in liver steatosis. Journal of
Physiology and Biochemistry. 2011;67:629-36.
676. Fujii H, Ikura Y, Arimoto J, Sugioka K, Iezzoni JC, Park SH, et al. Expression
of perilipin and adipophilin in nonalcoholic fatty liver disease; relevance to oxidative
injury and hepatocyte ballooning. J Atheroscler Thromb. 2009;16:893-901.
265
677. Straub BK, Stoeffel P, Heid H, Zimbelmann R, Schirmacher P. Differential
pattern of lipid droplet-associated proteins and de novo perilipin expression in
hepatocyte steatogenesis. Hepatology. 2008;47:1936-46.
678. Motomura W, Inoue M, Ohtake T, Takahashi N, Nagamine M, Tanno S, et al.
Up-regulation of ADRP in fatty liver in human and liver steatosis in mice fed with
high fat diet. Biochemical and Biophysical Research Communications.
2006;340:1111-8.
679. Imamura M, Inoguchi T, Ikuyama S, Taniguchi S, Kobayashi K, Nakashima
N, et al. ADRP stimulates lipid accumulation and lipid droplet formation in murine
fibroblasts. Am J Physiol Endocrinol Metab. 2002;283:E775-83.
680. McManaman JL, Bales ES, Orlicky DJ, Jackman M, MacLean PS, Cain S, et
al. Perilipin-2 Null Mice are Protected Against Diet-Induced Obesity, Adipose
Inflammation and Fatty Liver Disease. Journal of Lipid Research. 2013.
681. Imai Y, Varela GM, Jackson MB, Graham MJ, Crooke RM, Ahima RS.
Reduction of hepatosteatosis and lipid levels by an adipose differentiation-related
protein antisense oligonucleotide. Gastroenterology. 2007;132:1947-54.
682. Varela GM, Antwi DA, Dhir R, Yin X, Singhal NS, Graham MJ, et al.
Inhibition of ADRP prevents diet-induced insulin resistance. Am J Physiol
Gastrointest Liver Physiol. 2008;295:G621-8.
683. Bieker JJ. Kruppel-like factors: three fingers in many pies. Journal of
684. Ratziu V, Lalazar A, Wong L, Dang Q, Collins C, Shaulian E, et al. Zf9, a
Kruppel-like transcription factor up-regulated in vivo during early hepatic fibrosis.
1998;95:9500-5.
685. Watanabe K, Ohnishi S, Manabe I, Nagai R, Kadowaki T. KLF6 in
nonalcoholic fatty liver disease: role of fibrogenesis and carcinogenesis.
Gastroenterology. 2008;135:309-12.
686. Miele L, Beale G, Patman G, Nobili V, Leathart J, Grieco A, et al. The
Kruppel-like factor 6 genotype is associated with fibrosis in nonalcoholic fatty liver
disease. Gastroenterology. 2008;135:282-91 e1.
687. Bechmann LP, Gastaldelli A, Vetter D, Patman GL, Pascoe L, Hannivoort
RA, et al. Glucokinase links Kruppel-like factor 6 to the regulation of hepatic insulin
sensitivity in nonalcoholic fatty liver disease. Hepatology. 2012;55:1083-93.
688. Bechmann LP, Vetter D, Ishida J, Hannivoort RA, Lang UE, Kocabayoglu P,
et al. Post-transcriptional activation of PPAR alpha by KLF6 in hepatic steatosis.
Journal of Hepatology. 2013.
266
689. Cha DR, Han JY, Su DM, Zhang Y, Fan X, Breyer MD, et al. Peroxisome
proliferator-activated receptor-alpha deficiency protects aged mice from insulin
resistance induced by high-fat diet. American Journal of Nephrology. 2007;27:47982.
690. Heinhuis B, Netea MG, van den Berg WB, Dinarello CA, Joosten LA.
Interleukin-32: a predominantly intracellular proinflammatory mediator that controls
cell activation and cell death. Cytokine. 2012;60:321-7.
691. DeFronzo RA. Insulin resistance, lipotoxicity, type 2 diabetes and
atherosclerosis: the missing links. The Claude Bernard Lecture 2009. Diabetologia.
2010;53:1270-87.
692. Kwaan HC. Changes in blood coagulation, platelet function, and plasminogenplasmin system in diabetes. Diabetes. 1992;41 Suppl 2:32-5.
693. Alzahrani S, Ajjan R. Review article: Coagulation and fibrinolysis in diabetes.
Diabetes and Vascular Disease Research. 2010;7:260-73.
694. Dimova EY, Kietzmann T. Metabolic, hormonal and environmental regulation
of plasminogen activator inhibitor-1 (PAI-1) expression: lessons from the liver.
Thrombosis and Haemostasis. 2008;100:992-1006.
695. Bastard JP, Pieroni L, Hainque B. Relationship between plasma plasminogen
activator inhibitor 1 and insulin resistance. Diabetes/Metabolism Research and
Reviews. 2000;16:192-201.
696. De Taeye B, Smith LH, Vaughan DE. Plasminogen activator inhibitor-1: a
common denominator in obesity, diabetes and cardiovascular disease. Curr Opin
Pharmacol. 2005;5:149-54.
697. Kannel WB, McGee DL. Diabetes and cardiovascular disease. The
Framingham study. JAMA. 1979;241:2035-8.
698. Wiman B. Plasminogen activator inhibitor 1 (PAI-1) in plasma: its role in
thrombotic disease. Thromb Haemost. 1995;74:71-6.
699. Banfi C, Eriksson P, Giandomenico G, Mussoni L, Sironi L, Hamsten A, et al.
Transcriptional regulation of plasminogen activator inhibitor type 1 gene by insulin:
insights into the signaling pathway. Diabetes. 2001;50:1522-30.
700. Videla LA, Tapia G, Rodrigo R, Pettinelli P, Haim D, Santibanez C, et al.
Liver NF-kappaB and AP-1 DNA binding in obese patients. Obesity (Silver Spring).
2009;17:973-9.
701. Xia Y, Makris C, Su B, Li E, Yang J, Nemerow GR, et al. MEK kinase 1 is
critically required for c-Jun N-terminal kinase activation by proinflammatory stimuli
and growth factor-induced cell migration. Proceedings of the National Academy of
267
702. Campos SP, Baumann H. Insulin is a prominent modulator of the cytokinestimulated expression of acute-phase plasma protein genes. Molecular and Cellular
Biology. 1992;12:1789-97.
703. Muegge K, Vila M, Gusella GL, Musso T, Herrlich P, Stein B, et al.
Interleukin 1 induction of the c-jun promoter. Proceedings of the National Academy
of Sciences of the United States of America. 1993;90:7054-8.
704. Karin M, Liu Z, Zandi E. AP-1 function and regulation. Current Opinion in
Cell Biology. 1997;9:240-6.
705. Shaulian E, Karin M. AP-1 as a regulator of cell life and death. Nat Cell Biol.
2002;4:E131-6.
706. Stepniak E, Ricci R, Eferl R, Sumara G, Sumara I, Rath M, et al. c-Jun/AP-1
controls liver regeneration by repressing p53/p21 and p38 MAPK activity. Genes and
Development. 2006;20:2306-14.
707. Smart DE, Green K, Oakley F, Weitzman JB, Yaniv M, Reynolds G, et al.
JunD is a profibrogenic transcription factor regulated by Jun N-terminal kinaseindependent phosphorylation. Hepatology. 2006;44:1432-40.
708. Smart DE, Vincent KJ, Arthur MJ, Eickelberg O, Castellazzi M, Mann J, et al.
JunD regulates transcription of the tissue inhibitor of metalloproteinases-1 and
interleukin-6 genes in activated hepatic stellate cells. Journal of Biological Chemistry.
2001;276:24414-21.
709. Meixner A, Karreth F, Kenner L, Penninger JM, Wagner EF. Jun and JunDdependent functions in cell proliferation and stress response. Cell Death and
Differentiation. 2010;17:1409-19.
710. Baynes JW. Role of oxidative stress in development of complications in
diabetes. Diabetes. 1991;40:405-12.
711. Houstis N, Rosen ED, Lander ES. Reactive oxygen species have a causal role
in multiple forms of insulin resistance. Nature. 2006;440:944-8.
712. Greene LA, Lee HY, Angelastro JM. The transcription factor ATF5: role in
neurodevelopment and neural tumors. Journal of Neurochemistry. 2009;108:11-22.
713. Persengiev SP, Devireddy LR, Green MR. Inhibition of apoptosis by ATFx: a
novel role for a member of the ATF/CREB family of mammalian bZIP transcription
factors. Genes and Development. 2002;16:1806-14.
714. Angelastro JM, Mason JL, Ignatova TN, Kukekov VG, Stengren GB,
Goldman JE, et al. Downregulation of activating transcription factor 5 is required for
differentiation of neural progenitor cells into astrocytes. Journal of Neuroscience.
2005;25:3889-99.
268
715. Mason JL, Angelastro JM, Ignatova TN, Kukekov VG, Lin G, Greene LA, et
al. ATF5 regulates the proliferation and differentiation of oligodendrocytes.
Molecular and Cellular Neurosciences. 2005;29:372-80.
716. Pascual M, Gomez-Lechon MJ, Castell JV, Jover R. ATF5 is a highly
abundant liver-enriched transcription factor that cooperates with constitutive
androstane receptor in the transactivation of CYP2B6: implications in hepatic stress
responses. Drug Metabolism and Disposition: The Biological Fate of Chemicals.
2008;36:1063-72.
717. Bell SG, Vallee BL. The metallothionein/thionein system: an oxidoreductive
metabolic zinc link. Chembiochem. 2009;10:55-62.
718. Yang L, Li H, Yu T, Zhao H, Cherian MG, Cai L, et al. Polymorphisms in
metallothionein-1 and -2 genes associated with the risk of type 2 diabetes mellitus and
its complications. Am J Physiol Endocrinol Metab. 2008;294:E987-92.
719. Haynes V, Connor T, Tchernof A, Vidal H, Dubois S. Metallothionein 2a gene
expression is increased in subcutaneous adipose tissue of type 2 diabetic patients.
Molecular Genetics and Metabolism. 2013;108:90-4.
720. Boden G. Endoplasmic reticulum stress: another link between obesity and
insulin resistance/inflammation? Diabetes. 2009;58:518-9.
721. Schroder M, Kaufman RJ. The mammalian unfolded protein response. Annual
Review of Biochemistry. 2005;74:739-89.
722. Harding HP, Calfon M, Urano F, Novoa I, Ron D. Transcriptional and
translational control in the Mammalian unfolded protein response. Annual Review of
Cell and Developmental Biology. 2002;18:575-99.
723. Oyadomari S, Mori M. Roles of CHOP/GADD153 in endoplasmic reticulum
stress. Cell Death and Differentiation. 2004;11:381-9.
724. Oyadomari S, Araki E, Mori M. Endoplasmic reticulum stress-mediated
apoptosis in pancreatic beta-cells. Apoptosis. 2002;7:335-45.
725. Ferri KF, Kroemer G. Organelle-specific initiation of cell death pathways. Nat
Cell Biol. 2001;3:E255-63.
726. Lai E, Bikopoulos G, Wheeler MB, Rozakis-Adcock M, Volchuk A.
Differential activation of ER stress and apoptosis in response to chronically elevated
free fatty acids in pancreatic beta-cells. Am J Physiol Endocrinol Metab.
2008;294:E540-50.
727. Sharma NK, Das SK, Mondal AK, Hackney OG, Chu WS, Kern PA, et al.
Endoplasmic reticulum stress markers are associated with obesity in nondiabetic
subjects. Journal of Clinical Endocrinology and Metabolism. 2008;93:4532-41.
269
728. Wang Y, Wan B, Li D, Zhou J, Li R, Bai M, et al. BRSK2 is regulated by ER
stress in protein level and involved in ER stress-induced apoptosis. Biochemical and
Biophysical Research Communications. 2012;423:813-8.
729. Kimball SR, Jefferson LS. Induction of REDD1 gene expression in the liver in
response to endoplasmic reticulum stress is mediated through a PERK, eIF2α
phosphorylation, ATF4-dependent cascade. Biochemical and Biophysical Research
Communications. 2012;427:485-9.
730. Ellisen LW, Ramsayer KD, Johannessen CM, Yang A, Beppu H, Minda K, et
al. REDD1, a developmentally regulated transcriptional target of p63 and p53, links
p63 to regulation of reactive oxygen species. Molecular Cell. 2002;10:995-1005.
731. Wolff NC, Vega-Rubin-de-Celis S, Xie XJ, Castrillon DH, Kabbani W,
Brugarolas J. Cell-type-dependent regulation of mTORC1 by REDD1 and the tumor
suppressors TSC1/TSC2 and LKB1 in response to hypoxia. Molecular and Cellular
Biology. 2011;31:1870-84.
732. Shoshani T, Faerman A, Mett I, Zelin E, Tenne T, Gorodin S, et al.
Identification of a novel hypoxia-inducible factor 1-responsive gene, RTP801,
involved in apoptosis. Molecular and Cellular Biology. 2002;22:2283-93.
733. Frost RA, Huber D, Pruznak A, Lang CH. Regulation of REDD1 by insulinlike growth factor-I in skeletal muscle and myotubes. Journal of Cellular
Biochemistry. 2009;108:1192-202.
734. Regazzetti C, Bost F, Le Marchand-Brustel Y, Tanti JF, Giorgetti-Peraldi S.
Insulin induces REDD1 expression through hypoxia-inducible factor 1 activation in
adipocytes. Journal of Biological Chemistry. 2010;285:5157-64.
735. Regazzetti C, Dumas K, Le Marchand-Brustel Y, Peraldi P, Tanti JF,
Giorgetti-Peraldi S. Regulated in development and DNA damage responses -1
(REDD1) protein contributes to insulin signaling pathway in adipocytes. PLoS One.
2012;7:e52154.
736. Yamamoto Y, Moore R, Flavell RA, Lu B, Negishi M. Nuclear receptor CAR
represses TNFalpha-induced cell death by interacting with the anti-apoptotic
GADD45B. PLoS One. 2010;5:e10121.
737. Zhang N, Ahsan MH, Zhu L, Sambucetti LC, Purchio AF, West DB. NFkappaB and not the MAPK signaling pathway regulates GADD45beta expression
during acute inflammation. Journal of Biological Chemistry. 2005;280:21400-8.
738. Huang CY, Tan TH. DUSPs, to MAP kinases and beyond. Cell Biosci.
2012;2:24.
739. Li B, Yi P, Zhang B, Xu C, Liu Q, Pi Z, et al. Differences in endoplasmic
reticulum stress signalling kinetics determine cell survival outcome through activation
of MKP-1. Cellular Signalling. 2011;23:35-45.
270
740. Wu JJ, Roth RJ, Anderson EJ, Hong EG, Lee MK, Choi CS, et al. Mice
lacking MAP kinase phosphatase-1 have enhanced MAP kinase activity and
resistance to diet-induced obesity. Cell Metab. 2006;4:61-73.
741. Firth SM, Baxter RC. Cellular actions of the insulin-like growth factor binding
proteins. Endocrine Reviews. 2002;23:824-54.
742. Gleason CE, Ning Y, Cominski TP, Gupta R, Kaestner KH, Pintar JE, et al.
Role of insulin-like growth factor-binding protein 5 (IGFBP5) in organismal and
pancreatic beta-cell growth. Molecular Endocrinology. 2010;24:178-92.
743. Imai J, Katagiri H, Yamada T, Ishigaki Y, Suzuki T, Kudo H, et al. Regulation
of pancreatic beta cell mass by neuronal signals from the liver. Science.
2008;322:1250-4.
744. Bostrom P, Wu J, Jedrychowski MP, Korde A, Ye L, Lo JC, et al. A PGC1alpha-dependent myokine that drives brown-fat-like development of white fat and
thermogenesis. Nature. 2012;481:463-8.
745. Castillo-Quan JI. From white to brown fat through the PGC-1alpha-dependent
myokine irisin: implications for diabetes and obesity. Dis Model Mech. 2012;5:293-5.
746. Shan T, Liang X, Bi P, Kuang S. Myostatin knockout drives browning of
white adipose tissue through activating the AMPK-PGC1alpha-Fndc5 pathway in
muscle. FASEB J. 2013.
747. van Marken Lichtenbelt WD, Vanhommerig JW, Smulders NM, Drossaerts
JM, Kemerink GJ, Bouvy ND, et al. Cold-activated brown adipose tissue in healthy
men. New England Journal of Medicine. 2009;360:1500-8.
748. Zhou Z, Yon Toh S, Chen Z, Guo K, Ng CP, Ponniah S, et al. Cidea-deficient
mice have lean phenotype and are resistant to obesity. Nature Genetics. 2003;35:4956.
749. Lecker SH, Zavin A, Cao P, Arena R, Allsup K, Daniels KM, et al. Expression
of the irisin precursor FNDC5 in skeletal muscle correlates with aerobic exercise
performance in patients with heart failure. Circ Heart Fail. 2012;5:812-8.
750. Huh JY, Panagiotou G, Mougios V, Brinkoetter M, Vamvini MT, Schneider
BE, et al. FNDC5 and irisin in humans: I. Predictors of circulating concentrations in
serum and plasma and II. mRNA expression and circulating concentrations in
response to weight loss and exercise. Metabolism: Clinical and Experimental.
2012;61:1725-38.
751. Sanchis-Gomar F, Lippi G, Mayero S, Perez-Quilis C, Garcia-Gimenez JL.
Irisin: a new potential hormonal target for the treatment of obesity and type 2
diabetes. J Diabetes. 2012;4:196.
752. Tong X, Yin L. Circadian Rhythms in Liver Physiology and Liver Diseases.
Comprehensive Physiology: John Wiley & Sons, Inc.; 2013.
271
753. Maury E, Ramsey KM, Bass J. Circadian rhythms and metabolic syndrome:
from experimental genetics to human disease. Circulation Research. 2010;106:44762.
754. Gimble JM, Sutton GM, Bunnell BA, Ptitsyn AA, Floyd ZE. Prospective
influences of circadian clocks in adipose tissue and metabolism. Nat Rev Endocrinol.
2011;7:98-107.
755. Antunes LC, Levandovski R, Dantas G, Caumo W, Hidalgo MP. Obesity and
shift work: chronobiological aspects. Nutr Res Rev. 2010;23:155-68.
756. Scheer FA, Hilton MF, Mantzoros CS, Shea SA. Adverse metabolic and
cardiovascular consequences of circadian misalignment. Proceedings of the National
757. Knutson KL, Ryden AM, Mander BA, Van Cauter E. Role of sleep duration
and quality in the risk and severity of type 2 diabetes mellitus. Archives of Internal
Medicine. 2006;166:1768-74.
758. Shi S-q, Ansari TS, McGuinness Owen P, Wasserman David H, Johnson
Carl H. Circadian Disruption Leads to Insulin Resistance and Obesity. Current
biology : CB. 2013;23:372-81.
759. Stubblefield JJ, Terrien J, Green CB. Nocturnin: at the crossroads of clocks
and metabolism. Trends Endocrinol Metab. 2012;23:326-33.
760. Douris N, Green CB. NOC out the fat: a short review of the circadian
deadenylase Nocturnin. Annals of Medicine. 2008;40:622-6.
761. Garbarino-Pico E, Niu S, Rollag MD, Strayer CA, Besharse JC, Green CB.
Immediate early response of the circadian polyA ribonuclease nocturnin to two
extracellular stimuli. RNA. 2007;13:745-55.
762. Baggs JE, Green CB. Functional analysis of nocturnin: a circadian clockregulated gene identified by differential display. Methods Mol Biol. 2006;317:243-54.
763. Green CB, Douris N, Kojima S, Strayer CA, Fogerty J, Lourim D, et al. Loss
of Nocturnin, a circadian deadenylase, confers resistance to hepatic steatosis and dietinduced obesity. Proceedings of the National Academy of Sciences of the United
764. Hee SW, Tsai SH, Chang YC, Chang CJ, Yu IS, Lee PC, et al. The role of
nocturnin in early adipogenesis and modulation of systemic insulin resistance in
human. Obesity (Silver Spring). 2012;20:1558-65.
765. Li R, Yue J, Zhang Y, Zhou L, Hao W, Yuan J, et al. CLOCK/BMAL1
regulates human nocturnin transcription through binding to the E-box of nocturnin
promoter. Molecular and Cellular Biochemistry. 2008;317:169-77.
272
766. Reppert SM, Weaver DR. Coordination of circadian timing in mammals.
Nature. 2002;418:935-41.
767. Shimba S, Ogawa T, Hitosugi S, Ichihashi Y, Nakadaira Y, Kobayashi M, et
al. Deficient of a clock gene, brain and muscle Arnt-like protein-1 (BMAL1), induces
dyslipidemia and ectopic fat formation. PLoS One. 2011;6:e25231.
768. Mathers CD, Loncar D. Projections of global mortality and burden of disease
from 2002 to 2030. PLoS Med. 2006;3:e442.
769. Bhattacharyya S, I OS, Katyal S, Unterman T, Tobacman JK. Exposure to the
common food additive carrageenan leads to glucose intolerance, insulin resistance and
inhibition of insulin signalling in HepG2 cells and C57BL/6J mice. Diab tologia.
2012;55:194-203.
770. Andreozzi F, Procopio C, Greco A, Mannino GC, Miele C, Raciti GA, et al.
Increased levels of the Akt-specific phosphatase PH domain leucine-rich repeat
protein phosphatase (PHLPP)-1 in obese participants are associated with insulin
resistance. Diabetologia. 2011;54:1879-87.
771. Pandey AK, Bhardwaj V, Datta M. Tumour necrosis factor-alpha attenuates
insulin action on phosphoenolpyruvate carboxykinase gene expression and
gluconeogenesis by altering the cellular localization of Foxa2 in HepG2 cells. FEBS
J. 2009;276:3757-69.
772. Ha do T, Tuan DT, Thu NB, Nhiem NX, Ngoc TM, Yim N, et al. Palbinone
and triterpenes from Moutan Cortex (Paeonia suffruticosa, Paeoniaceae) stimulate
glucose uptake and glycogen synthesis via activation of AMPK in insulin-resistant
human HepG2 Cells. Bioorganic and Medicinal Chemistry Letters. 2009;19:5556-9.
773. Gupta D, Varma S, Khandelwal RL. Long-term effects of tumor necrosis
factor-alpha treatment on insulin signaling pathway in HepG2 cells and HepG2 cells
overexpressing constitutively active Akt/PKB. Journal of Cellular Biochemistry.
2007;100:593-607.
774. Chaussade C, Rewcastle GW, Kendall JD, Denny WA, Cho K, Gronning LM,
et al. Evidence for functional redundancy of class IA PI3K isoforms in insulin
signalling. Biochemical Journal. 2007;404:449-58.
775. Lin CL, Lin JK. Epigallocatechin gallate (EGCG) attenuates high glucoseinduced insulin signaling blockade in human hepG2 hepatoma cells. Mol Nutr Food
Res. 2008;52:930-9.
273
274
APPENDIX
To avoid gel artefacts all samples to be compared were loaded onto the minimum
amount of gels. Two different gels were run per antibody. For NGT vs T2DM
comparisons, individuals without type 2 diabetes were loaded adjacent to individuals
with type 2 diabetes. This alternating sequence was followed through the gel.
For the pre and post RYGB comparison there were six pairs of liver samples from
individuals with normal glucose tolerance (sNGT) while there were seven pairs of
livers from individuals with type 2 diabetes (sT2DM). Pre and post RYGB samples
were loaded adjacent to each other. sNGT and sT2DM samples were alternated.
9.1 Western Blots for Chapter 3
Figure 9.1-1. Figure 9.1-2. ENPP1 (110-130kda) western blot using protein extracted from liver tissue
biopsied from individuals at the time of RYGB surgery.
275
Figure 9.1-3. IDE (118kda) western blot using protein extracted from liver tissue biopsied from
individuals at the time of RYGB surgery.
Figure 9.1-4. CEACAM-1 (160-180kda) western blot using protein extracted from liver tissue biopsied
from individuals at the time of RYGB surgery.
276
Figure 9.1-5. Actin (43kda) western blot using protein extracted from liver tissue biopsied from
Figure 9.1-6. ENPP1 (110-130kda) western blot using protein extracted from liver tissue biopsied from
individuals both at the time of RYGB surgery and again at a subsequent operation.
277
Figure 9.1-7. IDE (118kd) western blot using protein extracted from liver tissue biopsied from
Figure 9.1-8 CEACAM-1 (160-180) western blot using protein extracted from liver tissue biopsied
from individuals both at the time of RYGB surgery and again at a subsequent operation
278
Figure 9.1-9 Actin-1 (43kda) western blot using protein extracted from liver tissue biopsied from
279
Figure 9.2-1. Insulin Receptor β subunit (95kda) western blot using protein extracted from liver tissue
biopsied from individuals at the time of RYGB surgery.
Figure 9.2-2. Insulin Receptor β subunit (95kda) western blot using protein extracted from liver tissue
biopsied from individuals both at the time of RYGB surgery and again at a subsequent operation.
280
9.2.1 AKT Phosphorylation Western Blots
Cell lysate from Hep G2 cells overexpressing IR-A or IR-B was resolved on 10%
SDS-PAGE gels and transferred to PVDF membrane. After transfer the membrane
was stained with Ponceau S stain and cut according to molecular weight to allow for
simultaneous probing of Insulin receptor β-subunit, Actin and phosphorylated AKT.
The portion of the membrane probed for p-AKT was stripped and re-probed for total
AKT. Figure 6 presents all 9 biological repeats that were used to estimate p-AKT
levels.
Figure 9.2-3. AKT phosphorylation in HepG2 cells overexpressing IR-A (Hep G2 IR-A) or IR-B (Hep
G2 IR-B). Empty vector was used as control (Hep G2 EV).
281
Figure 9.3-1. LGALS4 (36kda) western blot using protein extracted from liver tissue biopsied from
individuals in the sNGT and sT2DM group at the at the time of RYGB surgery
Figure 9.3-2. LGALS4 (36kda) western blot using protein extracted from liver tissue biopsied from
282

Liver Molecular Mechanisms Involved in Type 2 Diabetes

Transcription

Similar documents

Diabetes, Type 1

Education

Tim McGraw has a challenge for you.

The Economic Benefit of Public Funding of Insulin Pumps in

ARC factsheet Special Research Initiative for Type 1 Juvenile

July - CityMasala

AstraZeneca Mölndal Presentation PDF 13683KB

Sponsor a Child and Save a Life - International Diabetes Federation

NIDDK Medical Student Research Program in Diabetes and Obesity