Role of basic fibroblast growth factor (FGF-2)

Transcription

Role of basic fibroblast growth factor (FGF-2)
University of Veterinary Medicine
Center for Systemic Neurosciences
Department of Neuroanatomy, Hannover Medical School
Role of basic fibroblast growth factor (FGF-2) during
development of mesencephalic dopaminergic neurons
of substantia nigra in mice
THESIS
submitted in partial fulfillment of the requirements for the degree of
- Doctor rerum naturalium (Dr. rer. nat.)
by
Olga Baron
born in Shdanowka (Russia)
Hannover 2011
II
Supervisor:
Prof. Dr. rer. nat. Claudia Grothe
Supervison group:
Prof. Dr. rer. nat. Claudia Grothe
Prof. Dr. med. Reinhard Dengler
Prof. Dr. med. Lanfermann
1st Evaluation:
Prof. Dr. rer. nat. Claudia Grothe
Institute of Neuroanatomy
Hannover Medical School
Carl-Neuberg-Straße 1
30625 Hannover
Prof. Dr. med. Reinhard Dengler
Institute of Neurology
Hannover Medical School
Carl-Neuberg-Straße 1
30625 Hannover
Prof. Dr. med. Lanfermann
Institute of Neuroradiology
Hannover Medical School
Carl-Neuberg-Straße 1
30625 Hannover
2nd Evaluation:
Prof. Dr. rer. nat. Beate Brandt-Saberi
Institute of Anatomy and Molecular Embryology
Ruhr-University Bochum
Universitätsstr. 150
44801 Bochum
Date of final exam:
7th of October, 2011
This work was supported by the Deutsche Forschungsgemeinschaft (DFG) with a grand
to Prof. Dr. rer. nat. Claudia Grothe
III
Publication list:
Ratzka A.*, Kalve I.*, Özer M., Nobre A., Wesemann M, Jungnickel J, Köster-Patzlaff C., Baron O.,
Grothe C. “The co-layer method as an efficient way to genetically modify mesencephalic
progenitor cellstransplanted into 6-OHDA rat model of Parkinson´s disease”, Cell Transplantation,
2011;20:7. Epub 2011, Jul 1. DOI: 10.3727/096368911X586774, *authors contributed equally
Ratzka A., Baron O., Grothe C. “FGF-2 deficiency does not influence FGF ligand and receptor expression
during development of the nigrostriatal system”. PLOS one, 2011;6(8):e23564. Epub 2011 Aug
18.
Ratzka A.*, Baron O.*, Stachowiak M.K., Grothe C. “FGF-2 regulates dopaminergic neuron development
in vivo”. [submitted July, 2011] *authors contributed equally
Baron O., Narhla S., Terranova C., Förthmann B., Stachowiak E., Kuhlemann K., Ratzka A., Claus P.,
Grothe C., Stachowiak M. “Intergrative nuclear FGFR signaling (INFS) in dopaminergic neuron
development: interaction of nuclear FGFR1 and NURR1”. [submitted October, 2011]
(Baron O., Ratzka A., Grothe C. “FGF-2 regulates axonal outgrowth and target innervation durin
nigrostriatal pathway formation”. [in preparation])
Oral presentations
Baron O., Ratzka A., Grothe C. “Role of FGF-2 during development of substantia nigra”. ZSN-Colloquium
2009, Braunlage, Germany, 8.-10.9.2010
Baron O., Ratzka A., Grothe C. “FGF-2 influences late embryonic development of substantia nigra in
mice”. GfE School on “Common mechanisms of development and regeneration”. Günzburg,
Germany, 23.-25.9.2010
Baron O., Ratzka A., Grothe C. “FGF-2 influences late stages of embryonic development of nigral
dopaminergic neurons in mice”. NECTAR-annual meeting, Freiburg, Germany, 25.-27.11.2010
Poster presentations
Baron O., Ratzka A., Grothe C. “Role of FGF-2 during development of substantia nigra”. Gordons
Research Conference on „Fibroblast growth factors in development and disease“, Ventura, USA,
14. -19.03.2010
Baron O., Ratzka A., Grothe C. FGF-2 influences late stages of embryonic development of nigral
dopaminergic neurons in mice”. ZSN-Colloquium 2010, Hannover, Germany, 8.-10.10.2010
Baron O., Ratzka A., Grothe C. “FGF-2 regulates dopaminergic neuron development in substantia nigra”.
The annual meeting of the German Neuroscience Society (NWG), Göttingen, Germany 2327.03.2011
IV
V
Dedicated to my dear mother Maria and my dear brother Max
VI
VII
Table of contents
Summary ______________________________________________________________ 1
Zusammenfassung______________________________________________________ 3
1. Introduction__________________________________________________________ 5
1.1 Degeneration of mesencephalic dopaminergic neurons in Parkinson’s disease ___ 5
1.2 Development of dopaminergic neurons in the ventral midbrain _________________ 6
1.2.1 Dopaminergic neurons of the ventral midbrain _____________________________________ 6
1.2.2 Development of the neural tube ________________________________________________ 7
1.2.3 Regionalization of mDA field ___________________________________________________ 8
1.2.4 Specification and mDA induction ________________________________________________ 9
1.2.5 Terminal differentiation ______________________________________________________ 10
1.2.6 Target innervation and maturation______________________________________________ 11
1.2.7 Nurr1 ____________________________________________________________________ 12
1.3 FGF-2 ________________________________________________________________ 13
1.3.1 FGFs in neural development __________________________________________________ 13
1.3.2 FGF-2 signaling ____________________________________________________________ 14
1.3.3 INFS_____________________________________________________________________ 17
1.4 FGF-2 in nigrostriatal system ____________________________________________ 18
1.5 FGF-2 deficient mice ___________________________________________________ 19
1.5 Aims of the thesis _____________________________________________________ 20
2. Materials and methods________________________________________________ 22
2.1 Animals and breeding __________________________________________________ 22
2.2 Cell culture ___________________________________________________________ 23
2.2.1 Culturing of cell lines ________________________________________________________
2.2.2 Dissection of ventral mesencephalon ___________________________________________
2.2.3 Primary dissociated ventral mesencephalic progenitor cells __________________________
2.2.4 Organotypic tissue culture ____________________________________________________
23
24
24
25
2.3 Biochemistry and molecular biology ______________________________________ 28
2.3.1 Transient gene delivery ______________________________________________________
2.3.2 Fluorescence immunocytochemistry ____________________________________________
2.3.3 Cell-ELISA ________________________________________________________________
2.3.4 Nuclear and cytoplasmic fractionation___________________________________________
2.3.5 BCA-assay ________________________________________________________________
2.3.6 Co-Immunoprecipitation______________________________________________________
2.3.7 SDS-PAGE _______________________________________________________________
2.3.8 Western blot assay _________________________________________________________
2.3.9 Genotyping________________________________________________________________
2.3.10 Quantitative RT-PCR _______________________________________________________
28
29
30
31
32
32
33
33
34
35
2.4 Histology _____________________________________________________________ 36
2.4.1 Tissue processing __________________________________________________________ 36
2.4.2 In vivo BrdU-labeling ________________________________________________________ 37
2.4.3 Tyrosine hydroxylase immunohistochemistry _____________________________________ 37
VIII
2.4.4 Antigen retrieval ____________________________________________________________ 38
2.4.5 Fluorescence immunohistochemistry ___________________________________________ 38
2.5 Microscopic analysis and morphometry ___________________________________ 39
2.5.1 Microscopic imaging ________________________________________________________
2.5.2 Evaluation of proliferation ____________________________________________________
2.5.3 Evaluation of apoptosis ______________________________________________________
2.5.4 Stereological cell counts _____________________________________________________
2.5.5 Nigrostriatal fiber outgrowth___________________________________________________
39
39
40
40
44
2.6 Statistics _____________________________________________________________ 45
2.7 Materials _____________________________________________________________ 45
2.7.1 Chemicals ________________________________________________________________
2.7.2 Cell culture media __________________________________________________________
2.7.3 Buffers and gels ____________________________________________________________
2.7.4 Plasmids _________________________________________________________________
45
47
48
51
3. Results_____________________________________________________________ 55
3.1 Phenotype onset of the supernumerary TH-ir cells in SNpc ___________________ 55
3.2 Unchanged expression of mDA marker genes on mRNA level _________________ 57
3.3 Unchanged expression of FGFs on mRNA level _____________________________ 59
3.4 Reproduction of the phenotype of supernumerary TH-ir cells in vitro ___________ 61
3.5 Overexpression of selected FGF-ligands___________________________________ 63
3.6 Loss of FGF-2 increases the number of proliferative mDA precursors in SVZ ____ 63
3.7 β-catenin and extracellular FGF signaling in E14.5 FGF-2 deficent mice _________ 66
3.8 Nuclear accumulation of FGFR1 in FGF-2 deficient mice _____________________ 68
3.9 Nuclear FGFR1 interacts with Nurr1_______________________________________ 69
3.9.1 FGFR1 and Nurr1 are co-expressed in the nuclei of mDA neurons ____________________ 69
3.9.2 FGFR1 and Nurr1 co-exist in the same nuclear complexes __________________________ 71
3.10 Analysis of postnatal natural cell death___________________________________ 74
3.11 FGF-2 in axonal outgrowth of mDA neurons _______________________________ 75
4. Discussion _________________________________________________________ 79
4.1 FGF-2 is required during development of SNpc _____________________________ 80
4.2 mDA marker genes are unchanged in FGF-2 deficient mice on mRNA level ______ 81
4.3 FGF-2 does not affect the FGF system on RNA level _________________________ 82
4.4 FGF-2 modulates proliferation of nigral mDA progenitors in vivo ______________ 84
4.5 Unchanged activation of cytosolic signal transduction pathways ______________ 86
4.6 FGF signaling is involved in terminal differentiation _________________________ 87
4.7 Crosstalk of mDA and FGF signaling______________________________________ 88
4.8 FGF-2 is involved in maturation and target innervation _______________________ 92
IX
4.9 Multimodal role of FGF-2 in developing VM_________________________________ 95
4.10 Concluding remarks___________________________________________________ 97
5. References _________________________________________________________ 98
6. Acknowledgement/ Danksagung ______________________________________ 115
7. Curriculum vitae ____________________________________________________ 116
X
Abbreviations
6-OHDA - 6-hydroxydopamine
ABC – avidin biotin complex
ABTS – 2,2’-azino-bis(3-ethylbenzthiazoline-6-sulphonic acid
AEC - Aminoethylcarbazol
BDNF – brain-derived neurotrophic factor
BMP – bone morphogenic protein
BCA – bicinchoninic acid
BSA – bovine serum albumine
cAMP - cyclic adenosine monophosphate
CBP – CREB binding protein
CNS – central nervous system
CREB - cAMP response element-binding
DAT – dopamine transporter
DAPI – 4’,6-diamidino-2-phenylindole
DIV – days in vitro
DMF -dimethylformamide
E14.5 – embryonic day 14.5
En-1/2 – engrailed 1/2
ERK1/2 - extracellular regulated kinase ½
ELISA – enzyme linked immunosorbent assay
FB – forebrain
FCS – fetal calf serum
FGF – fibroblast growth factor
FGFR – FGF-receptor
FRS – FGF receptor substrate
GDNF – glial cell-line derived neurotrophic factor
GSK-3 - glycogen synthase kinase 3
HMW – high molecular weight
HSPG - heparan sulfate proteoglucan
INFS – Integrative nuclear FGFR1 signaling
ko – knock out
LMW – low molecular weight
MAPK - mitogen-activated protein kinases
mDA – mesencephalic dopaminergic
MHB – midbrain-hindbrain boundary
MIA – multiple image alignment
MZ – mantel zone
NGF – nerve growth factor
NLS – nuclear localization sequence
OTC – organotypic culture
P0 – postnatal day 0
PBS –phosphate buffered saline
PFA - paraformaldehyde
PD – Parkinson’s disease
qRT-PCR – quantitative real-time polymerase chain reaction
RrF – retrorubral field
RSK - ribosomal s6 kinase
RXR – retinoid X receptor
SC – spinal cord
SHH – sonic hedgehog
XI
SDS-PAGE – sodium dodecylsulfat polyacrylamid gel electrophoresis
SNpc – substantia nigra pars compacta
STR – striatum
SVZ – subventricular zone
TH – tyrosine hydroxylase
TH-ir – TH-immunoreactive
VM – ventral mesencephalon
Vmat2 - vesicular monoamine transporter 2
VTA – ventral tegmental area
VZ – ventricular zone
wt – wild type
XII
List of figures
Figure 1. Neurulation
7
Figure 2. The development of ventral midbrain is subdivided in four stages
8
Figure 3. Wnt1 – Lmx1a regulatory loop cooperates with the SHH-Foxa2 pathway
10
Figure 4. Intracellular FGF signaling via activation of transmembrane FGFRs
15
Figure 5. Organotypic tissue culture according to Stoppini et al. 1991
27
Figure 6. Substantia nigra of wild type (wt) and FGF-2 deficient (ko) neonatal mice
43
Figure 7. FGF-2 deficiency increases mDA neuron number in the SNpc
56
Figure 8. Transcript levels of mDA marker genes are unchanged in VM of FGF-2 deficient
animals at E14.5 and P0.
58
Figure 9. Transcript levels of selected FGFs and FGFRs in ventral midbrain (VM),
striatum (STR) and spinal cord (SC)
18kDa
Figure 10. Overexpression of FGF-2
60
, FGF-8b, FGF-15 or FGF-17a does not influence
differentiation of mDA neurons in vitro
62
Figure 11. Increased proliferation of mDA precursor cells in the rostral VM
65
Figure 12. Increased levels of nuclear FgfR1 in VM of FGF-2 deficient mice
67
Figure 13. Nurr1 and FGFR1 co-exist in the cell nuclei of the ventral midbrain
70
Figure 14. Presence of nuclear FGFR1 and Nurr1 in the same nuclear protein complexes
of mDA progenitors
72
Figure 15. Nuclear FGFR1 and Nurr1 interaction after overexpression in neuroblastoma cells
73
Figure 16. Reduced number of late apoptotic cells in P0 FGF-2 deficient mice
75
Figure 17. FGF-2 participates in mDA axonal outgrowth
78
Figure 18. Putative sources of FGF-2 during nigrostriatal pathway formation
95
1
Summary
Olga Baron: “Role of basic fibroblast growth factor (FGF-2) during development
of mesencephalic dopaminergic neurons of substantia nigra in mice”
Progredient loss of mesencephalic dopaminergic neurons (mDA) in the substantia nigra
pars compacta (SNpc) is the main cause for characteristic symptoms in Parkinson’s
disease. Insight in the regulation of the SNpc development may benefit to the
understanding of disease pathophysiology and improvement of therapeutic approaches.
Previous studies revealed in addition to a protecting function of FGF-2 in mature mDA
neurons, a regulatory role of FGF-2 for proper development of substantia nigra pars
compacta (SNpc). The increased numbers of tyrosine hydroxylase immunoreactive
(THir) neurons in SNpc of adult FGF-2 deficient mice correlated with decreased
numbers in FGF-2 overexpressing mice. However, with regard to the mitogenic and
neuroprotective function of FGF-2 on dopaminergic precursors and adult neurons,
respectively, the opposed outcome was anticipated. To elucidate the physiological
function of FGF-2 in the nigrostriatal development, the present study concentrated on
embryonic (E14.5), newborn (P0), and juvenile (P28) FGF-2 deficient mice.
Stereological analysis on the content of TH-ir cells, in the SNpc of FGF-2 depleted mice
could delineate the onset of the phenotype between E14.5 and P0. Additionally, the
separate examination of the SNpc and ventral tegmental area (VTA) in juvenile mice
(P28) showed a specific increase in number of TH-ir cells in SNpc. Examination of the
mDA marker gene expression by quantitative RT-PCR and in situ hybridization revealed
an unchanged patterning of the embryonic VM of FGF-2 deficient. Moreover, a
comprehensive analysis of transcript levels of the complete FGF system revealed no
compensation of FGF-2 loss by upregulation of any other FGF family member. To
unravel the underlying mechanism, immunohistochemical analysis of proliferation rates
in E14.5 animals and of apoptosis rates in P0 animals was performed. However,
increase of proliferating Lmx1a-ir mDA progenitors in the subventricular zone of FGF-2
deficient embryonic VM indicated an imbalance in FGF signaling resulting in altered cell
2
cycle progression of neuronal progenitors and enhanced differentiation of mDA neurons.
In SNpc of newborn FGF-2 deficient mice a decrease of apoptotic cells negative for
cleaved caspase-3 was detected. Additionally, enhanced mDA fiber outgrowth in VM
and forebrain explant co-cultures of FGF-2 deficient mice was observed compared to
heterogenous or wild type cultures. The decreased apoptosis and enhanced fiber
outgrowth suggest an incoherent wiring control during innervation of the forebrain
structures under FGF-2 deficiency.
Altogether, presumably both physiological alterations determine the phenotype in the
ventral mesencephalon of the adult FGF-2 depleted mice: longer or enhanced
neurogenic divisions of progenitors may cause an increased generation of mDA
neurons, while altered wiring control during the maturation may lead to reduced
ontogenic death of mDA neurons. Investigation of molecular signaling mechanisms
revealed that activation of intracellular signaling cascades, like ERK1/2, Akt and also
Wnt/β-catenin signaling, were maintained in absence of FGF-2 in the VM of FGF-2
deficient E14.5 embryos. On the other hand, nuclear FGFR1 was found to be increased
in the nucleus of FGF-2 deficient mice. Additionally, a novel INFS / Nurr1 interactive
mechanism for gene activation during neuronal development has been identified, as
shown by co-localization and co-imunoprecipitation of Nurr1 and FGFR-1 and
underlined by functional assays of collaborators. Exemplary, this study revealed a novel
INFS / Nurr1 interactive mechanism for gene activation during neuronal development,
which may offer a new therapeutic target to increase production of mDA neurons for
restorative approaches in Parkinson’s disease.
3
Zusammenfassung
Olga Baron: „Rolle des basischen Fibroblastenwachstumsfaktors (FGF-2) bei der
Entwicklung mesencephaler dopaminerger Neurone der Substantia nigra in der
Maus“
Parkinson’sche Erkrankung ist durch den Untergang der dopaminergen Neurone in der
Substantia
nigra
pars
compacta
(SNpc)
charakterisiert.
Die
Entschlüsselung
physiologischer und molekularer Mechanismen während der Entwicklung der SNpc
würde zum besseren Verständnis der Ätiologie der Parkinson’schen Erkrankung sowie
zur weiteren Entwicklung Zell-basierter therapeutischer Ansätze beitragen. Frühere
Studien haben gezeigt, dass FGF-2 möglicherweise eine Rolle bei der Entwicklung
mesenzephaler dopaminerger Neurone spielt. So hat die Untersuchung adulter FGF-2defizienter Mausmutanten ergeben, dass im Vergleich zu den Wildtyp-Tieren in der
SNpc dieser Mutanten signifikant mehr Tyrosinhydroxylase (TH)-positive Neurone
vorhanden sind, welches als Marker für die dopaminegen Neurone dient, da es
limitierend bei der Dopaminsynthese ist. Entsprechend zu diesem Befund wurden in den
FGF-2 überexpremierenden Mutanten signifikant weniger dopaminerge Neurone
vorgefunden. Da FGF-2 als mitogener und neurotropher Faktor für dopaminerge
Neurone bekannt ist, wurde ein genau entgegengesetzter Phänotyp erwartet. Diese
Befunde wurden mit einer möglichen Überkompensation des FGF-2-Verlustetes durch
andere Faktoren diskutiert. Um die Rolle des FGF-2 bei der Entwicklung der
dopaminergen Neurone eingehender zu untersuchen, wurden in dieser Arbeit
embryonale (E14,5), neugeborene (P0) und juvenile (P28) FGF-2 defiziente
Mausmutanten untersucht.
Die stereologische Erhebung der TH-positiven Zellen hat ergeben, dass der Phänotyp
mit erhöhter TH-positiver Zellzahl sich zwischen E14,5 und P0 entwickelt. Ferner wurde
in den juvenilen Stadien festgestellt, dass der Phänotyp SNpc-spezifisch ist. Die
quantitative
RT-PCR
zeigte
keine
Unterscheide
bezüglich
der
dopaminergen
Markergene bei der Entwicklung der FGF-2 defizienten Mäuse und Wildtyp-Tiere, was
4
auf eine normale Spezifikation der dopaminergen Domäne in der frühen embryonalen
Entwicklung der FGF-2-defizienten Mäuse hindeutet. Auch die mRNA-Level anderer
Mitglieder der FGF-Familie wurden in den FGF-2 defizienten Mutanten unverändert
vorgefunden, was eher eine Kompensation auf Proteinebene beziehungsweise durch
andere Faktoren vermuten lässt. Zwei physiologische Prozesse wurden mit Hilfe
immunzytochemischer Analysen verändert vorgefunden, die beide zusammen die
Entwicklung des beobachteten Phänotypes begründen könnten. Einerseits wurde eine
erhöhte Proliferation Lmx1a-positiver dopaminerger Vorläufer in der subventrikulären
Zone jedoch nicht in der ventrikulären Zone der FGF-2-defizienten Embryonen
vorgefunden. Dies deutet auf ein Ungleichgewicht im FGF-Signalweg hin, wobei mehr
Vorläufer
zu
dopaminergen
Neuronen
differenzieren.
Der
andere
verändert
vorgefundene entwicklungsphysiologische Prozess ist der verringerte ontogenetische
Zelltod in neugeborenen FGF-2-Mutanten, der mit verändertem dopaminergem
Faserwachstum zu telenzephalen Zielstrukturen in explantierten Co-Kulturen aus
ventralem Mittelhirn sowie Vorderhirn korreliert werden konnte. Wobei die längeren
Fasern der reinen FGF-defizienten Co-Kulturen sowie breiteren Trakte der heterogenen
Co-Kulturen im Vergleich zu kürzeren in einem schmaleren Trakt gelegenen Fasern der
reinen Wildtyp-Co-Kulturen auf eine reduzierte Kontrolle der adäquaten Verdrahtung der
nigrostriatalen dopaminergen Fasern hindeuten. Die biochemische Analyse der ERK1/2,
Akt und Wnt/β-catenin Signalwege im embryonalen VM zeigte keine Unterschiede in
ihrer Aktivierung in FGF-2 defizienten Mäusen im Vergleich zu Wildtyp-Tieren. Allerdings
wurde in embryonalen Stadien eine Akkumulation des FGF-Rezeptors-1 in nukleären
Extrakten aus ventralem Mittelhirn der FGF-2-defizienten Mäuse vorgefunden, was auf
eine verstärkte Aktivierung des Intergrativen Nukleären FGFR-1 Signalweges (INFS)
hindeutet. INFS wurde mit neuronaler Differenzierung und Induktion der TH-expression
in Verbindung gebracht. Zusätzlich, wurde ein neuer interaktiver Mechanismus zwischen
INFS und Nurr1, einem zentralen integrativen Transkriptionsfaktor in der dopaminergen
Entwicklung, identifiziert und charakterisiert, der bei der Entwicklung zukünftiger
regenerativer Strategien für Parkinson-Therapie eine Rolle spielen könnte.
5
1. Introduction
1.1 Degeneration of mesencephalic dopaminergic neurons in
Parkinson’s disease
The mesencephalic dopaminergic (mDA) neurons of the substantia nigra pars compacta
(SNpc) project to the forebrain and innervate the GABAergic neurons of the caudateputamen complex in the dorsolateral striatum. The so called nigrostriatal system
functions as a modulatory feature in the movement control which is executed by the
basal ganglia circuit (Kreitzer and Malenka, 2008). Progredient loss of mDA neurons in
the substantia nigra causes the cardinal symptoms in Parkinson’s disease (PD):
bradykinesia, tremor, rigor and postural instability. The symptomatic stage emerges
when more than 60 % of the mDA cells died. So far no cure is evident up to now.
Besides a few familiar and trauma caused cases (approximately 20%), the etiology of
sporadic (idiopathic) Parkinson’s disease remains unresolved (Shulman et al., 2011).
Among several well discussed rationales like oxidative stress, environmental
neurotoxins, and/or multigenetic determination (Thomas and Beal, 2007, Hardy, 2010,
Morley and Hurtig, 2010, Shulman et al., 2011), the age-related depletion of
neurotrophic factors has been also considered as the matter of mDA neuron
degeneration (Siegel and Chauhan, 2000, Krieglstein, 2004). In this context several
factors were studied, among which brain-derived neurotrophic factor (BDNF), glial cellline derived neurotrophic factor (GDNF) and basic fibroblast growth factor (FGF-2)
seemed to be the most promising candidates. Current therapeutic strategies target
mainly the improvement of motoric symptoms. Up to now the central objective is the
dopamine replacement, which mainly includes classical oral pharmacotherapy with LDOPA, the blood-brain barrier crossing dopamine precursor, which is also combined
with inhibitors of dopamine-degrading enzymes. Invasive approaches ensure a constant
L-DOPA delivery in form of pumps to omit dyskinetic phases, which occur as a side
effect of a long term treatment (Poewe et al., 2010). Deep brain stimulation was shown
to improve significantly motor symptoms and the quality of live, accompanied by
6
reduction of L-DOPA requirement (Kleiner-Fisman et al., 2006, Morley and Hurtig,
2010). Clinical studies were performed using regenerative approaches like gene therapy
(Bjorklund et al., 2010, Bjorklund and Kordower, 2010) and intrastriatal cell replacement
strategies (Arenas, 2010, Lindvall and Kokaia, 2009). Other efforts concentrate on
presymptomatic recognition of Parkinson’s disease and development of causal
approaches concerning antioxidative pharmacotherapy and neurotrophic factors, which
could prevent the progredient degeneration (Savica et al., 2010, Schapira, 2008,
Schapira, 2009, Wu et al., 2011). Another future perspective in the regeneration of the
midbrain mDA system is the induction of inherent programs which target the adult
neurogenesis (Borta and Hoglinger, 2007, Okano et al., 2007, O'Keeffe et al., 2009).
However, better insights into the molecular signaling networks during development of
the mDA neurons is crucial for understanding of PD pathophysiology and might
contribute to the regenerative approaches which require an extensive production of
mDA neurons, either in cell-replacement strategies as well as in induced endogenous
adult neurogenesis.
1.2 Development of dopaminergic neurons in the ventral midbrain
1.2.1 Dopaminergic neurons of the ventral midbrain
mDA neurons of the ventral midbrain are arranged in three different cell groups: ventral
tegmental area = VTA (A10), substantia nigra pars compacta = SNpc (A9) and
retrorubral field = RrF (A8). The axons of the SNpc project to the nucleus caudatus and
putamen forming the nigrostriatal pathway, while the projections of VTA and RrF
assemble the mesolimbic and mesocortical system, innervating nucleus accumbens,
amygdala, olfactory tubercle and prefrontal cortex. The development of ventral mDA
domain includes four main stages: regionalization, specification, differentiation, and
maturation (Prakash and Wurst, 2006). The complex signaling network responsible for
mDA neuron development comprises of multiple extrinsic and intrinsic players, which
are involved at least in two or three different cross-talking pathways. A consistent
7
understanding of this intricate regulatory network remains to be achieved (Abeliovich
and Hammond, 2007, Smidt and Burbach, 2007, Gale and Li, 2008).
1.2.2 Development of the neural tube
Development of the nervous system starts during the gastrulation. The notochord, a
transient dorsal mesodermal structure organized by the “node” (known as the “Hensen’s
node” in avians), produces factors, which induce the neuroectoderm in the ectodermic
germ layer. In terms, the bone morphogenic protein (BMP) signaling, which promotes
the epidermal fate, is antagonized by inhibitory signals. The further process called
neurulation continues with the formation of a thick flat bulge, the neural plate, which
starts to roll up forming neural folds with specialized cells of the neural crest at their
boarders (Figure 1). Finally, the fusion of neural fold at the dorsal midline results in the
closure of the neural tube, whereas the neural crest cells delaminate and migrate out
(Vieira et al., 2010).
8
Figure 1. Neurulation. During neurulation the neural plate forms a neural tube by fusion of the neural
folds at the dorsal midline. The regionalization of the neural plate forced by ventralizing action of SHH and
dorsalizing effect of BMP generates columnar dorsoventral organization of the neural tube: FP – floor
plate, BP – basal plate, AP-alar plate and RP – roof plate (modified according to Vieira et al., 2010).
1.2.3 Regionalization of mDA field
The regionalization of the neuroectoderm occurs progressively, starting at the neural
plate stage, continuing during and after the neurulation. The developing neural tube,
becomes subdivided in longitudinal, the non overlapping regions parallel to the anteriorposterior axis, and transverse domains as single segments, which are characterized by
specific gene expression patterns forming specific boarders. According to the
prosomeric model the transverse subdivision encloses (from rostral to caudal) the
telencephalic prosomers 4-6, prosomers 1-3 of rostral diencephalon, midbrain, and
finishing with rhombomers 1-4 of the rhombencephalon. The Figure1 illustrates the
longitudinal subdivision, which contains most-ventrally the floor plate, followed by basal
plate, alar plate and finally terminated dorsally with the roof plate (Rubenstein et al.,
1994).
Figure 2. The development of ventral midbrain is subdivided in four stages. During regionalization
the Otx2 and Gbx2 set the boarder between hind- and midbrain, the MHB (blue belt, hb = hind brain, mb =
midbrain). The combined action of SHH at the floor plate and MHB-derived FGF8 defines the origin of the
mDA field. The specification of mDA fate requires a complex signaling cascade which is mainly regulated
by cross-talk between Wnt1-Lmx1a and SHH-Foxa2 pathway. The activation of Msx1 and Ngn2 leads to
neuronal commitment and suppression of alternate fates. The mDA progenitors exit the cell cycle, leave
the ventricular zone and start to express the postmitotic marker Nurr1. The mDA precursors differentiate
to TH expressing mDA-neurons, while migrating towards the pial surface already starting to extent
processes. Combinatorial action of En1/2, Lmx1b, Nurr1, Pitx3 as well as other factors is necessary for
the final differentiation events and projection establishment to the forebrain targets. The maturation
endures until the 3d postnatal week, including the migration of young mDA neurons to the final destination
as well as selection of adequately wired neurons during postnatal ontogenic cell death phases (modified
according to Gale and Li, 2008).
9
The positioning of the mesencephalic mDA field is initiated by formation of the floor
plate and the midbrain-hindbrain boundary (MHB) between midbrain and rhombomer 1,
which is also called the isthmic organizer or isthmus. The anterior Otx2 expression and
Gbx2 in the hindbrain have been identified as the key signaling cues in the isthmus
formation between embryonic day E7 and E9. Importantly, the MHB instructs the
expression of molecules involved in mDA fate induction, FGF8 and Wnt1, in the
adjacent transverse bands (Liu and Joyner, 2001). Another factor participating in mDA
fate induction is sonic hedgehog (SHH), which is a ventralizing factor, secreted by the
floor plate (Hynes et al., 1995). In vivo and vitro experiments indicate that TGF-β is
required for SHH inductive potential (Farkas et al., 2003). However both inductive
signaling cascades a required for induction of mDA fate, since the origin of the mDA
field is determined at the intersection of FGF8 and Wnt1 with SHH driven signaling (Ye
et al., 1998).
1.2.4 Specification and mDA induction
The specification and determination of the mDA fate occurs between E9 and E14, and
widely overlaps with the differentiation (E11.5-15), which follows immediately the
induction of early mDA progenitors. During the early specification, the organizer regions
instruct the fate of the adjacent self-renewing progenitors, which in terms leave the
ventricular zone and exit the cell-cycle. mDA precursors migrate towards the pial surface
and undergo a gradual commitment to mDA fate, starting to express the postmitotic
mDA marker genes.
MHB instructs the expression of a number of transcription factors, like Engrailed 1/2 (En1/2), Lmx1a and Lmx1b, Pax-2, -5, -8 (Liu and Joyner, 2001). SHH directly induce
Foxa2 (former HNF-3ß) expression via Gli transcription factors (Sasaki et al., 1997).
Recent findings unraveled a novel Wnt1-Lmx1a autoregulatory loop, which explains the
converged regulation of mDA differentiation by Lmx1a and Lmx1b and shows a
synergistic communication with the SHH-Foxa2 pathway (Figure 3) (Chung et al., 2009).
10
Lmx1a is an intrinsic key determinant of mDA neurons, whose mDA inducing function
depends on ventralizing activity of SHH (Andersson et al., 2006). Induction of Otx2 by
Wint1 and Msx1 by Lmx1a results in inhibition of alternate fate by negative regulation of
Nkx2.2 and Nkx6.1 (Andersson et al., 2006, Prakash et al., 2006). Foxa2 participates in
repression of Nkx2.2 in SHH dependent manner (Ferri et al., 2007) and cooperates with
Lmx1a independently of SHH (Lin et al., 2009), which reflects the multiply collateralized
signaling events. Besides the inhibition of the alternate fate, Otx2, Msx1, and Foxa2
induce Ngn2 expression promoting the neural commitment (Andersson et al., 2006, Kele
et al., 2006, Ferri et al., 2007, Chung et al., 2009).
Figure 3. Wnt1 – Lmx1a regulatory loop cooperates with the SHH-Foxa2 pathway in induction of mDA
neuronal cell fate, regulating the inhibition of alternate fate, induction of neurogenesis and postmitotic
mDA specific phenotype determinants (modified according to Chung et al 2009).
1.2.5 Terminal differentiation
The postmitotic mDA neurons immerge very soon adjacent to the ventricular zone, and
are identified starting at E10.5 by expression of Nurr1 (Wallen et al., 1999), the key
regulator in specification and maintenance of mDA transmitter phenotype (Wallen and
Perlmann, 2003, Jacobs et al., 2009a, Jacobs et al., 2009b). Not yet having achieved
their final destination, the early mDA neurons start already to express tyrosine
hydroxylase (TH), the rate limiting enzyme of the dopamine synthesis, which is widely
11
used as a marker to identify mDA neurons. Another transcription factor, which serves as
an intrinsic determinant of mDA fate is Pitx3. The Pitx3 deficiency results in a loss of a
specific subgroup of mDA neurons in SNpc, which can be counteracted by retinoic acid
(Jacobs et al., 2007). The expression of Aldh1a1, which is required for conversion of
retinol to retinoic acid, actually precedes Pitx3 in mDA neurons (Wallen et al., 1999).
However, both Pitx3 and Nurr1 cooperate in transcriptional activation of mDA genes,
including the regulation of Aldh1a1 expression (Jacobs et al., 2009a). Further studies
are necessary to resolve the implication of retinoic acid, among others, in signaling
network during terminal mDA differentiation. Secreted factors, like Wnt5a, FGF-20, and
Tgf-β, were shown to promote differentiation of mDA neurons (Castelo-Branco et al.,
2003, Grothe et al., 2004, Roussa et al., 2004), although the precise mechanisms
remain unresolved.
1.2.6 Target innervation and maturation
The maturation phase immediately pursues the differentiation. The first rostrally
directed projections start to emerge as soon as TH-immunoreactive (TH-ir) cells can be
detected in the ventral midbrain (Nakamura et al., 2000). The migratory processes as
well as innervation of the respective forebrain areas endure throughout the late
embryonic development until the third postnatal week (Voorn et al., 1988). The
ontogenic cell death occurs during postnatal development in the nigrostriatal system
(Burke, 2003, Burke, 2004). The inadequately wired mDA neurons undergo a natural
cell death, which is of apoptotic nature (Jackson-Lewis et al., 2000). The progression of
the so called programmed cell death in the SNpc follows a biphasic manner with the
main peak after birth and the more moderate second peak after the second postnatal
weeks (Jackson-Lewis et al., 2000, Burke, 2003).
The axonal outgrowth and target innervation are inextricably linked with the final
differentiation of newly born mDA neurons. Several factors were identified to regulate
these processes synergistically. Following the classical neurotrophic factor concept
12
(Oppenheim, 1991), GDNF serve as the main target derived neurotrophic factors,
supporting the innervating fibers, stimulating arborization, and promoting survival of the
midbrain cells (Kholodilov et al., 2004, Burke, 2006). Nurr1 seems to be implicated in the
axonal outgrowth and target innervation, as it regulates the intrinsic expression of Ret
(Wallen and Perlmann, 2003), a GDNF receptor, as well as Neuropilin-1 (Hermanson et
al., 2006), a co-receptor for semaphorins. Semaphorins 3A, 3C and 3F were identified
as putative cues in mDA path finding (Hernandez-Montiel et al., 2008, Kolk et al., 2009,
Yamauchi et al., 2009). The transcription factors En-1/2, which are also engaged in
generation of MHB, regulate the ontogenic cell death, preventing intrinsically the
adequately wired neurons to undergo apoptosis (Alberi et al., 2004). A recent publication
indicate, that Wnt5a promotes mDA axon elongation, retraction, and repulsion in a timedependent manner, as well as maturation and fasciculation of the medial forebrain
boundary (MFB) (Blakely et al., 2011). BDNF is crucially involved in survival of mDA
neurons (Baquet et al., 2005), including its cognate tyrosine kinase receptors TrkB and
TrkC (von Bohlen und Halbach et al., 2005), which are expressed by mDA neurons
(Nishio et al., 1998). It still not yet resolved, whether BDNF acts in an autocrine manner
or as a target derived neurotrophic factor.
1.2.7 Nurr1
Nurr1 is essential as a key transcription factor in terminal differentiation and maturation
of mDA neurons. Several interaction partners have been shown to regulate Nurr1
transcriptional activity, suggesting that Nurr1 may act to converge many cellular signals
(Smidt and Burbach, 2009). Nurr1 (NR4A2) belongs together with Nur77 (NR4A1) and
Nor1 (NR4A3) to the subfamily of orphan nuclear receptors, consisting of functionally
distinct domains including the evolutionary conserved DNA-binding domain in the central
region and ligand-binding domain in the C-terminal region of the protein (Law et al.,
1992, Zetterstrom et al., 1997). A unique feature of Nurr1 is the lack of the ligandbinding capacity, due to hydrophobic amino acid side chains, which entirely fill the space
normally occupied by ligands (Wang et al., 2003). Nurr1 binds to specific DNA binding
13
sites as monomer, dimer or after heterodimerization with retinoid X receptor (RXR) (Law
et al., 1992, Zetterstrom et al., 1996). The transcriptional activation and repression of
Nurr1 is regulated by at least two sumoylation sites (Galleguillos et al., 2004). Nurr1 is
required for expression of various genes (TH, DAT, Vmat2, Dlk1, Aldh1a1, c-ret)
essential for dopamine synthesis and function (Wallen and Perlmann, 2003, Jacobs et
al., 2009a). However, the mechanism by which Nurr1 subsequently activates the
transcription machinery is not completely resolved. The disruption of Nurr1 gene in mice
results in insufficient development of mDA neurons and their prenatal death (Zetterstrom
et al., 1997, Saucedo-Cardenas et al., 1998). Although the upstream regulation of Nurr1
is not fully understood, there are evidences for direct induction of Nurr1 expression by
Lmx1a and Foxa 2 (Chung et al., 2009). Nurr1 can be induced by hormones and growth
factors and several interaction partners have been identified (Perlmann and WallenMackenzie, 2004, Smidt and Burbach, 2009).
1.3 FGF-2
1.3.1 FGFs in neural development
FGF-2 is a member of the FGF family, which comprises 22 members in mammals. FGFs
have multiple roles in the central nervous system (Reuss and von Bohlen und Halbach,
2003). Besides the well known neurotrophic properties for a wide range of adult neurons
of central and peripheral nervous system, FGFs are also implicated in development of
the central nervous system (Ford-Perriss et al., 2001, Dono, 2003, Mason, 2007). FGF
signaling is proposed to mediate neural induction by antagonizing BMP signaling (Akai
and Storey, 2003), further, to regulate patterning processes, proliferation of neuronal
progenitors (Bouvier and Mytilineou, 1995, Raballo et al., 2000) and differentiation
(Reuss et al., 2003, Martinez-Morales et al., 2005). The importance of FGF signaling in
the regulation of balance between self-renewing properties of neural stem cells and
differentiation has been illustrated in ventral midbrain and cortical system of complex
FGF receptor (FGFR) mutants (Maric et al., 2007, Saarimaki-Vire et al., 2007, Partanen,
2007, Kang et al., 2009, Lahti et al., 2011). With regard to their neurotrophic function,
14
during development FGFs are involved in formation of functional neural networks
(Umemori, 2009), including axonal outgrowth (Shanmugalingam et al., 2000, Yamauchi
et al., 2009) and guidance (McFarlane et al., 1995), and synapse differentiation (Dai and
Peng, 1995, Li et al., 2002).
1.3.2 FGF-2 signaling
FGFs can be classified as intracrine (intracellular), paracrine (canonical) and endocrine
(hormone-like) FGFs by their mechanisms of action (Itoh and Ornitz, 2011). FGF-2
belongs to the paracrine FGFs. Atypically to other paracrine FGFs, FGF2 as well as
FGF1 lacks the N-terminal hydrophobic sequence (Itoh and Ornitz, 2011), which
normally directs newly synthetisized proteins to or through the membrane of the
endoplasmatic reticulum (ER). FGF-2 was shown to be released by an unconventional,
ER/Golgi-independent pathway via direct translocation across the plasma membrane
(Nickel, 2011, Nickel and Rabouille, 2009, Nickel and Seedorf, 2008). This mechanism
requires a posttranslational modification of FGF-2, which is mediated by Tec-kinase via
phosphorylation at tyrosine 82 (Ebert et al., 2010). All paracrine FGFs mediate biological
responses as extracellular proteins by building ternary complexes with cell surface
FGFRs and heparin/heparan sulphate as a cofactor (Klagsbrun and Baird, 1991, Yayon
et al., 1991). In humans and mice the four identified FGFR genes (FGFR1-FGFR4)
encode receptor tyrosine kinases consisting of extracellular ligand-binding domain with
three immunoglobulin-like domains (I, II and III), a transmembrane domain and an
intracellular tyrosine kinase domain. The ligand binding specificity of FGFRs is
determined by the immunoglobulin-like domain III, which occurs in two major isoforms
(IIIb and IIIc) due to alternative splicing of FGFR1-FGFR3 mRNA (Zhang et al., 2006,
Itoh and Ornitz, 2011). FGF-2 binds with different affinities to FGFRs, with high affinity to
the IIIc isoforms and only low to IIIb isoforms (Ornitz et al., 1996).
15
Figure 4. Intracellular FGF signaling via activation of transmembrane FGFRs. Extracellular activation
of FGFRs by FGFs stimulates the PI3-AKT pathway (yellow highlight),and the Ras-raf-MAPK pathway
(grey highlight). The activated MAPKs (ERKs, p38, or JNKs) translocate to the nucleus, where they
regulate target genes associated with growth and differentiation (adapted from Dailey et al., 2005).
Paracrine FGFs function in development by influencing the intracellular signaling events
of neighboring cells without a requirement of cell-cell contact (Itoh and Ornitz, 2011).
The ligand binding by FGFR leads to receptor dimerization, transphosphorylation of the
intracellular tyrosine kinase domain and activation of downstream signaling pathways
(Fig. 4). Besides the HSPGs as FGFR cofactors several other co-receptors/-factors have
been identified to interact with FGFRs and influence the signaling transduction, which in
part belong to the cell adhesion molecules. Intracellular responses due to FGFR
activation are transduced through multiple second messenger systems. Thereby, the
16
main signaling pathways are the Ras-raf-MAPK and PI3-Akt/PKB pathways. Activation
of extracellular regulated kinase 1/2 (ERK1/2) mitogen activated protein kinases (MAPK)
seems to resemble a common response after transmembrane FGFR1 activation, while
p38 and Jun MAPKs may be activated in a cell type specific manner. While ERK1/2
pathway is most widely implicated in developmental functions regulating growth and
differentiation, the phospatidylinositol 3 (PI3)-Akt/protein kinase B(PKB) pathway
mediates anti-apoptotic effects and survival. Both second messenger pathways require
the recruitment of FGF receptor substrate (FRS) adaptors. ERK activation stimulates
expression of transcription factors of the Ets family including cAMP response elementbinding (CREB) (Dailey et al., 2005, Eswarakumar et al., 2005, Mason, 2007, Itoh and
Ornitz, 2011).
Further, of high interest is the crosstalk between the canonical Wnt and FGF signaling,
which depends on Akt and ERK1/2 pathway activation resulting in inhibition of β-catenin
degradation by glycogen synthase kinase 3β (GSK-3β) (Frodin and Gammeltoft, 1999,
Torres et al., 1999, Dailey et al., 2005, Katoh, 2006). Shortly, the canonical Wnt/ βcatenin pathway includes the stabilization of β-catenin in the cytosol by inactivation of βcatenin degrading machinery, which requires activation of Wnt receptors. In turn, βcatenin translocates to the nucleus and serves as transcriptional co-activator
(MacDonald et al., 2009).
FGF-2 shares with FGF-1, FGF-3 and their receptor FGFR1 a unique function among
the FGFs: the nuclear localization (Florkiewicz et al., 1991, Stachowiak et al., 1996,
Antoine et al., 1997, Reilly and Maher, 2001, Itoh and Ornitz, 2011). Due to alternative
translation initiation at canonical AUG or downstream localized CUG codons the
endogenous FGF-2 protein appears in rodents as 21 kDa and 23 kDa high molecular
weight (HMW) and 18 kDa low molecular weight (LMW) FGF-2 isoforms (Florkiewicz et
al., 1991). The main structural difference of the HMW isoforms is the presence of
additional nuclear localization sequence (NLS) at the N-terminus, which directs those to
the nucleus by a direct route (Quarto et al., 1991). The 18 kDa FGF-2 isoform, which
17
also posses as other isoforms the c-teminal NLS sequence, also localizes in the
nucleus, but is considered to be mainly cytosolic (Amalric et al., 1991, Bugler et al.,
1991, Florkiewicz et al., 1991, Claus et al., 2004a). FGFR1 bound extracellular FGFs
can be internalized and translocated to the nucleus (Reilly et al., 2004). FGFR1 does not
posses any NLS and requires ribosomal s6 kinase (RSK), importin-β and/or other
unknown factors to be released from ER and to translocate to the nucleus, which can
occur either directly following the translation or after internalization (Reilly and Maher,
2001, Hu et al., 2004, Dunham-Ems et al., 2006, Stachowiak et al., 2007). In the
nucleus FGFR1 seems to participate in a very complex transcription activating program,
the so called integrative nuclear FGFR1 signaling (INFS) (Stachowiak et al., 2007,
Stachowiak et al., 2011).
1.3.3 INFS
Beside the classical function of transmembrane FGFR to transmit extracellular signals
into the cytoplasm, FgfR1 plays a central role in INFS. INFS is characterized by
translocation of FGFR1 to the nucleus, in response to diverse stimuli including HMW
FGF2, BMP7, hormonal receptors, NGF, neurotransmitters and retinoic acid
(Stachowiak et al., 2007, Stachowiak et al., 2011). In contrast to the mitogenic effects of
extracellular FGFs the activated INFS stimulates cell-cycle exit of proliferative
progenitors and induces cell differentiation (Stachowiak et al., 2003, Stachowiak et al.,
2007). The nuclear FGFR1 appears to constitute a universal “feed-forward-and-gate”
network module that directs toward postmitotic development. Nuclear FGFR1 releases
CBP and RSK from inactive complexes. This coupled activation of CREB signaling plus
sequence specific transcription factors (ssTFs) enables a coordinated regulation of a
multi-gene program involved in differentiation (Stachowiak et al., 2007, Stachowiak et
al., 2011). A direct interaction of intranuclear FGFR1 with CBP was shown to activate
CREB (Fang et al., 2005, Stachowiak et al., 2007), which leads to induction of TH gene
expression in bovine adrenal medullary cells (Peng et al., 2002).
18
1.4 FGF-2 in nigrostriatal system
FGF-2 is a relevant neurotrophic factor within the nigrostriatal dopaminergic system. In
general FGF-2 is widely distributed in the developing and mature central nervous
system (Grothe et al., 1991, Weise et al., 1993, Grothe and Wewetzer, 1996, Ozawa et
al., 1996). FGF-2 expression was shown in the developing and adult SNpc (Gonzalez et
al., 1995, Gonzalez et al., 1990, Bean et al., 1991, Bean et al., 1992, Cintra et al., 1991),
in glial cells of the neostriatum and ventral midbrain (Cintra et al., 1991, Chadi et al.,
1993), as well as in mDA neurons (Bean et al., 1991, Cintra et al., 1991, Tooyama et al.,
1992). Expression of the multiple protein forms of FGF-2 is developmentally regulated in
the rodent brain. While the 21-kDa and 18-kDa forms peak in embryonic brain, the
expression of the 23 kDa form was first detected postnatally in whole brain extracts
(Giordano et al., 1992). In the SNpc of adult rats mainly HMW and to a minor extend 18
kDa FGF-2 isoforms are expressed (Claus et al., 2004b, Tooyama et al., 1992).
According to the expression pattern and binding affinity, FGF-2 can act via four receptor
isoforms within SNpc: FGFR-1IIIb, -1IIIc, - 2IIIc, and -3IIIc (Ornitz et al., 1996, Claus et
al., 2004b, Grothe and Timmer, 2007).
First in vitro experiments reported that FGF-2 promotes survival and development of
mDA neurons in culture (Ferrari et al., 1989). In addition, FGF-2 was proved to protect
mDA neurons from neurotoxin induced death in vitro (Park and Mytilineou, 1992), later
also in vivo, in animal model of PD, showing a more pronounced lesion of mDA neurons
after 6-OHDA administration in FGF-2 deficient mice and better survival in FGF-2
overexpressing mice (Timmer et al., 2007). Investigation of the post mortem tissue of
Parkinson’s patients revealed a striking link between FGF-2 and Parkinson’s disease, as
the damaged SNpc showed significantly lower number of FGF-2-ir pigmented mDA
neurons in PD patients compared to non Parkinsonian brains (Tooyama et al., 1993,
Tooyama et al., 1994).
19
Application of FGF-2 in restorative rat models of PD showed, that pretreatment of the
ventral mesencephalic progenitors with FGF-2 before grafting as well as administration
of FGF-2 after grafting via intracerebral infusion or via co-grafting of FGF-2 producing
fibroblasts or Schwann cell, respectively, enhances mDA neuron number and improves
graft survival (Mayer et al., 1993, Takayama et al., 1995, Timmer et al., 2004).
Interestingly, co-grafting of Schwann cells over-expressing HMW FGF-2 isoforms leads
to a significantly higher yield of surviving mDA neurons within the graft than cells
overexpressing 18 kDa FGF-2 (Timmer et al., 2004). In vitro sudies showed that
commercially available 18 kDa isoform mediates expansion of mDA progenitors (Studer
et al., 1998, Timmer et al., 2006, Pruszak et al., 2009). However, the mitogenic FGF-2
effect seems to be time-dependent, since prolonged FGF-2 treatment for 8 days in vitro
(DIV) results in decreased number of mDA neurons compared to 4 DIV (Jensen et al.,
2008). The aforementioned decreased neurotoxin induced survival of mDA neurons in
adult FGF-2 deficient mice coincides with increased number of tyrosine hydroxylase
immunoreactive (TH-ir) mDA neurons in unlesoned animals. Consistent with this the
FGF-2 overexpressing animals show additionally to increased survival a decreased
number of mDA neurons in unlesoned SNpc (Timmer et al., 2007). These findings
indicate that FGF-2 may regulate different physiological processes in adult and
developing nigrostriatal system (Grothe and Timmer, 2007).
1.5 FGF-2 deficient mice
Unfortunately, the deletion of a single FGF ligand gene often results in moderate
phenotypes (Itoh, 2007), reflecting a redundancy among the FGF family of ligands.
Likewise, FGF-2 deficient mice display a moderate phenotype including a reduced
number of specific neuron subtypes in the cerebral cortex, hippocampal formation and
cervical spinal cord, also hypertension, impaired regulation of blood pressure (Dono et
al., 1998, Vaccarino et al., 1999, Zechel et al., 2009), as well as increased number of
dopaminergic neurons within SNpc (Timmer et al., 2007), as already mentioned in
chapter 1.4. The increased number of mDA neurons is accompanied with increase of
20
SNpc volume in adult FGF-2 deficient animals, while the density of TH-ir cells remains
unchanged (Timmer et al., 2007). This correlates with another study, which analyzed
only densities of TH-ir cells and TH-ir fibers, revealing no differences between FGF-2
deficient and wild-type animals (Zechel et al., 2006).
However, with regard to the mitogenic and neuroprotective function of FGF-2 on
dopaminergic precursors and adult neurons (Ferrari et al., 1989, Bouvier and Mytilineou,
1995, Grothe et al., 2000, Timmer et al., 2007), respectively, the opposed outcome has
been anticipated. Therefore, it has been proposed that FGF-2 deficiency might be overcompensated by other FGF-family ligands. Although FGF-1 and FGF-2 widely overlap in
function and are both highly abundant in developing and adult CNS, the study on
double-deficient mice showed that FGF-1 is not responsible for the mild phenotype in
FGF-2 deficient mice (Miller et al., 2000). So far, none comprehensive analysis of the
dynamic changes in expression of the FGF system was conducted throughout the CNS
development in FGF-2 deficient mice.
1.5 Aims of the thesis
Massive loss of dopaminergic neurons in the SNpc leads to the characteristic symptoms
of Parkinson’s disease. Understanding the regulation of the SNpc development may
contribute to an improvement of therapeutic approaches. The endogenous FGF-2 is a
physiologically relevant neurotrophic factor in the developing and adult nigrostriatal
system (Grothe and Timmer, 2007). Previous studies in our lab indicated, additionally to
a protective function of FGF-2 in mature mDA neurons, involvement of FGF-2 mediated
signaling in developing SNpc. The increased numbers of TH-ir neurons in SNpc of adult
FGF-2 deficient mice, as well as decreased numbers in FGF-2 overexpressing mice
suggested a regulatory role of FGF-2 for proper development of SNpc (Timmer et al.,
2007).
The multifunctionality of endogenous FGF-2 provides several options of contribution to
SNpc development. FGF-2 is known to regulate proliferation of progenitors (Palmer et
21
al., 1995, Eiselleova et al., 2009, Vaccarino et al., 1999), their cell differentiation
(Vicario-Abejon et al., 1995) and migration (Dono et al., 1998) or cell death (Ma et al.,
2007, Yagami et al., 2010). The consequent challenge of this work was to define the
time window, which is responsible for the phenotype onset in the FGF-2 deficient mice.
Therefore subsequent analysis of stages representing the main milestones in VMdevelopment was of highest interest. Embryonic day 14.5 represented the stage of
terminal differentiation, postnatal day 0 (P0) assigns the first wave of ontogenic cell
death in maturating SNpc, and the juvenile animals (P28) represented the fully
developed SNpc. Molecular biological characterization of several marker genes for
dopaminergic neurons as well as of the FGF system was supposed to depict altered
physiological events and compensatory mechanisms, respectively, during SNpc
development. Of central interest were the physiological processes, which might be
affected under FGF-2 deficiency at certain time points, which includes progenitor
proliferation and/or cell cycle exit as well as postnatal apoptosis and axonal outgrowth.
Subsequent investigation of signaling pathway activation associated with FGF mediated
signal transduction, for example in growth and survival, should depict the mechanism
responsible for the assumed altered physiological processes. Identification of functional
interaction of molecular key regulators of mDA and FGF signaling pathways should
provide an insight in signaling integration of molecular crosstalk during basic
developmental processes.
22
2. Materials and methods
2.1 Animals and breeding
All experimental protocols were done in accordance with the German law for the
protection of animals with a permission of the local authority (Bezirksregierung
Hannover; guidelines of the Tierschutzgesetz i.d.F.v.). For timed pregnancy, the day of
the vaginal plug was assigned as embryonic day 0.5 (E0.5).
FGF-2-deleted mice. FGF-2 deficient mouse strain (FGF-2tm1ZIIr) (Dono et al., 1998)
maintained on C57Bl6/J background were received from The Jackson Laboratory (Bar
Harbor, ME). This strain was generated utilizing homologues recombination to insert a
neomycin expression cassette into the FGF-2 gene replacing the 1st exon. This mutation
resulted in deletion of all FGF-2 isoforms, since CUG and AUG start-codons for
translational initiation of all and molecular weight FGF-2 isoforms were removed. The
wild type and homozygous knock-out (FGF-2 ko) mice were obtained by crossbreeding
of heterozygous genetically altered animals, and genotypes were determined by PCR of
the tail DNA. The analyzed wild-type and knock-out animals were chosen from the same
litters.
EGFP-transgenic mouse strain Tg(ACTB-EGFP)B5 Nagy/J (obtained from A. Vortkamp,
University
Duisburg-Essen)
was
backcrossed
on
C57Bl6/J
background.
The
maintenance of the EGFP-transgenic mouse strain (TgEGFP) was accomplished by
breeding of homozygous animals.
Double mutants. The EGFP;FGF-2 ko mouse strain was achieved by crossbreeding of
the FGF-2 ko and TgEGFP mouse strains, both on C57Bl6/J background. Maintenance
of the double mutants occurred by breeding of homozygous animals.
23
2.2 Cell culture
2.2.1 Culturing of cell lines
Human neuroblastoma cell line (NB cells): The SK-N-BE(2) neuroblastoma cell line was
established in November of 1972 from a marrow biopsy taken from child with
disseminated neuroblastoma after repeated courses of chemotherapy and radiotherapy
(Lee and Kim, 2004). The cells exhibit moderate levels of dopamine beta hydroxylase
activity. The doubling time is 30 h.
SV40-immortalized rat ventral mesencephalic neuronal progenitor cells: The SV40i-VMNPCs were generated by introduction of the Simian Virus 40 (SV40) into the primary rat
embryonic neuronal progenitor cells, which were dissected from the ventral midbrain.
The resulting transformed cell clones displayed a two- to three-fold higher proliferation
rate compared to the primary cells. Under differentiation conditions, the cell clones
expressed mRNAs of transcription factors and other proteins essential for dopaminergic
development (Nobre et al., 2010). The clones were seeded in 75 cm2 in N2-medium
containing 3% fetal calf serum (FCS), afterwards cultivated in serum-free N2-medium or
in differentiation medium containing 1 mM dbcAMP and 20 ng GDNF.
Cell culture conditions: The standard culture conditions for the NB cells were 8 % CO2
and 37 °C, for SV40i-VM-NPCs 5% CO2, in a humidified incubator under normal
atmospheric oxygen content with nutritional support of serum containing media.
Cultivation was performed in 25 cm² flasks and 75 cm² flasks (Nuclon, Nunc) filled with 5
ml and 10 ml medium, respectively. Medium was changed every 2-3 days. A passage
was performed, when the cells achieved a confluence of 90-100 %.
Cell passage: For passage the cells were rinsed twice with warm sterile phosphate
buffered saline (PBS) and incubated in 0.05% Trypsin-EDTA for 3 min. The enzymatic
reaction was stopped by addition of a double amount of FCS containing medium. After
trituration the cells were pelletized in the centrifuge, the cell pellet was re-suspended in
medium, and the cells were seeded in an appropriate dilution in new cell culture flasks.
24
Freezing and thawing of cells: For freezing, the cells (6 Mio cells for SV-40-VM-NPCs,
and 10 Mio for NB cells) were re-suspended in 1 ml medium supplemented with 5 %
dimethyl sulfoxide (DMSO). The cell suspension was slowly frozen in cryovials, first for
30 min at -20 °C and finally at -80 °C. The thawing of cells was performed very fast in a
37 °C warm water bath. The defrosted cells were taken up in 6 ml preheated medium,
centrifuged, re-suspended and initially seeded in a 25 cm² flask.
2.2.2 Dissection of ventral mesencephalon
The dissociated VM progenitors were obtained from foeti at embryonic day 11.5. The
VM tissue was dissected as previously described (Bjorklund et al., 1983, Nikkhah et al.,
1994, Timmer et al., 2006, Pruszak et al., 2009). Briefly, the mice were sacrificed by
cervical dislocation and the gravid uteruses were harvested and placed in a Petri dish
with sterile ice cold PBS. Embryos were collected and transferred into a sterile Petri dish
filled with sterile ice cold Hank’s buffered saline for further manipulations. The crown
rump length (CRL) was measured under a stereomicroscope. The embryos were
decapitated with the forceps and the bodies disposed. Using the scalpel the midbrain
was separated from the neural tube and the ventral part was dissected. The butterfly
shaped dissected pieces of the VM were collected separately for each genotype on ice
in collection medium.
2.2.3 Primary dissociated ventral mesencephalic progenitor cells
To destroy free DNA dissected pieces of the VM were incubated in the collection
medium supplemented with 0.05 % DNase at 37° C for 15 min. The enzymatic reaction
was stopped by addition of 10 % FCS to the medium. To remove DNase the tissue was
pelletized at 1000 rpm for 5 min. The pellet was re-suspended in 1 ml of attachment
medium and triturated first through a blue 1 ml pipette tip and then a yellow 200 µl
pipette tip - 5-10 times each - in order to dissociate the cells mechanically. The viability
was nearly 100 % as determined by trypan blue dye exclusion.
25
The cell culture was performed according to Timmer et al. (2006) with required
modifications for the mouse tissue. The 96-well microplates were pre-coated with 0.1
mg/ml polyornithine (solved in 15 mM boric acid buffer, pH 8.4) and 6 µg/ml laminin in
distilled water for 24 hours on the day before dissection. After preparation the precoated microplates were washed twice with distilled water, pre-filled with 100 µl of warm
attachment medium and incubated in the incubator until seeding. The cell density in the
dissociated
tissue
suspension
was
assessed
with
a
cell-counting
chamber
(hemocytometer). The suspension was diluted if the assessed cell density was higher
then 600 cells/µl. 30 000 cells/well in a calculated suspension volume were seeded on
the pre-filled 96-well microplates.
The pilot experiments showed that for comparative evaluation of VM cultures from FGF2 deficient and wild type mice the highest yield of TH-ir cells was achieved, when the
attachment medium was replaced after 24 hours in vitro by differentiation medium. For
transfection experiments the attachment medium was replaced by serum free
proliferation medium containing FGF-2 as a mitogen. However, after attachment or
proliferation the cells were allowed to differentiate for six days in vitro (DIV) in
differentiation medium containing B27-supplement and ascorbic acid. The first day in
differentiation medium was considered as DIV 1. Cultures were maintained at 37°C in
humidified incubator supplied with 5% CO2 under normal oxygen conditions (~20%).
2.2.4 Organotypic tissue culture
Organotypic tissue culture (OTC) enables an easier experimental manipulation of the
tissue while the cells maintain an organotypic organization receiving the adequate in
vivo-like environmental input from surrounding cells. The ventral mesencephalon of
E14.5 mice cultured in collagen gels or in cell culture inserts according to the method
described previously (Stoppini et al., 1991). The ventral mesencephalon of E14.5 mice
was dissected as described above alterating the size and orientation of the explant,
depending on experimental layout.
26
Collagen gels
To form the collagen gels 8 parts of the rat tail collagen (4 mg/ml solved in 0,1 M acetic
acid) were mixed on ice with one part of 10x DMEM/Ham’s F12 and 1 part of
reconstitution buffer for neutralization of the acid in the collagen. After well mixing a drop
of the collagen mixture was applied into 6 well plates, and the tissue was transferred
and arranged in the collagen drop. 5-6 tissue cultures were cultured in one well in
parallel. The gels were allowed to form for 30 min at 37°C in the humid incubator.
Afterwards, the dishes were filled with 1 ml OTC-medium supplemented either with or
without human 18 kDa FGF-2 (20 ng/ml) allowing the top of the gel to have air contact.
Nigrostriatal co-cultures
After the optimal conditions for organotypic cultures were carried out the nigrostriatal cocultures of the forebrain and ventral midbrain tissue were performed with a modified
method according to Stoppini (1991). Dissection of the forebrain (FB) and the VM for
explant co-cultures occurred in ice-cold Hank’s BSS supplemented with 10 mM HEPES
(4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid). First, the FB and VM were
intersected through the diencephalon between prosomere 1 and prosomere 2: the
transecting line passed the mamillothlamic body and pretectum. The forebrain and
ventral midbrain were divided in two parts by a midsaggital cut, which resulted in two
explants per embryo. The forebrain explants were trimmed to remove the cortex and
bulbus olfactorius, mainly leaving the whole ganglionic eminence and parts of the
diencephalon to maintain the cues for the dopaminergic pathway. The ventral
mesencephalic explants were cut to remove the isthmus and the roof. The trimmed
explants were harvested on ice in the 6-well multidish containing membrane inserts filled
with 1 ml OTC medium supplemented with 10 mM HEPES. After all tissue was collected,
the forebrain and ventral mesencephalic explants were arranged under the
stereomicroscope putting the explants with the medial side (side of the ventricle) down
on the insert - exposed to the medium, and with the lateral side up - exposed to the air.
27
The diencephalic transsection sides of the VM and FB explants were attached to each
other meeting the dorsal and ventral orientation of the both explants. The OTC medium
in the insert was aspirated carefully with the pipette and 1 ml of fresh warm HEPES-free
OTC medium was applied under the insert (Fig. 5). The cultures were cultivated for 5
DIV in standard incubator with 5% CO2 at 37 °C. Every 2 days a complete medium
change was performed. If necessary, one drop of medium was applied on the culture in
the insert to prevent drying of the culture. After 5 days, the co-cultures were rinsed once
with warm PBS and fixed in 4% paraformaldehyde (PFA) in PBS over night at 4°C, and
were afterwards extensively washed with PBS using a shaking machine. The membrane
peaces with attached co-cultures were cut out of the insert and the tissue was
processed on the membranes for fluorescence immunocytochemistry as described in
2.3.2. All steps were applied using a shaking machine. The blocking buffer and carrier
solution contained an increased amount of Triton-X 100 of 0.5%. The washing steps
were extended to minimum ½ hour per step.
Figure 5. Organotypic tissue culture according to Stoppini et al. 1991. The scheme shows one tissue
culture placed on the membrane of the insert in one well of a multi-dish. For cultivation the medium should
not cover the membrane.
28
2.3 Biochemistry and molecular biology
2.3.1 Transient gene delivery
Lipofection is a lipid-based transfection technology which allows to deliver nucleic acids
in a wide range of mitotically active cells and cell lines, which incorporate the exogenous
DNA in the nucleus during the mitosis. For large-scale transfections of NB cells in 75
cm² culture flasks the Metafectene Pro reagent was used. Lipofectamine 2000 reagent
was applied for more efficient minor transfections of primary VM cells in 96-well
microtiter plates. The transfection procedures were performed according to the
manufacturer’s protocol.
Transfection of primary cells with Lipofectamine 2000: The primary VM were seeded in a
concentration of 4x104 cells/well in a 96-well microtiter plate and cultured for one day in
attachment and two days in proliferation medium. On the day of transfection, DNA and
Lipofectamine 2000 reagent stocks were mixed in Opti-MEM I in appropriate dilutions
(Table 1) and incubated for maximum 5 min. The forming of DNA-lipid-complexes was
performed for 20 min at room temperature by carefully mixing the previously generated
DNA and Lipofectamine stocks in ratio of 1:2. Meanwhile, the penicillin/streptomycin
containing medium was aspirated, the cells were washed with warm sterile PBS and
fresh proliferation medium was administered, without serum and antibiotics. After
complex formation the DNA containing liposomes were added to the medium and the
cells were incubated for 4-6 hours in culture conditions. Then, the solution was removed
and replaced with differentiation medium. The cells were allowed to differentiate for
further 5 days.
Table 1: Medium, DNA and Lipofection reagent amounts per transfection.
Dish
96-well plate
Area
Medium volume
Lipofectamine
DNA
0,3 cm³
100 µl
0,5 µl/25 µl
0,2 µg/ 25 µl
29
Transfection of NB cells with Metafectene: First, cells were expanded in culture medium
until they reached a confluence of 60 %. On the day of transfection plasmid-DNA
encoding for pCAGGGS-Nurr-1-flag protein (0.5 µg/ml, A. Ratzka) and plasmid-DNA
encoding for hFGFR1 (1.01 µg/µl, M. Stachowiak) were diluted in Opti-MEM I in a
relative amount of 10 µg DNA/300 µl medium (5 µg of each plasmid). Metafectene
reagent was diluted in a ratio 1:15 in 300 µl Opti-MEM. After maximum 5 min, the
previously generated DNA and Metafectene suspensions were carefully mixed and
DNA-Lipid-complexes was formed for 15 min at room temperature. Meanwhile, the
penicillin/streptomycin containing medium was aspirated, the cells were washed with
warm sterile PBS and fresh antibiotic free medium was administered. After complex
formation the DNA containing liposomes were added to the medium and the cells were
incubated over night in culture conditions. On the next morning the medium was
removed, the cells were washed once with warm medium and incubated for 24 more
hours in a medium containing 1 µM retinoic acid.
2.3.2 Fluorescence immunocytochemistry
For fluorescence immunocytochemistry (F-ICC) the indirect antigen detection method
was performed using primary antibodies recognizing the specific antigen and
fluorophore-labeled secondary antibodies raised against the IgGs from the same
species as the primary antibody. F-ICC was performed in 96 microtiter plates. Cells were
washed with 37 °C warm PBS, fixed with warm 4% PFA in PBS for 20 min at room
temperature, and washed three times with PBS. Afterwards cells were incubated in ICC
blocking buffer for one hour at room temperature. The primary antibodies were diluted in
carrier solution and incubated overnight at 4 °C. On the next day the remaining unbound
antibodies were removed by three washing steps with PBS and the fluorophoreconjugated secondary antibodies, diluted in carrier solution, were applied to the cultures
for one hour. Cells were washed again (three times with PBS). For staining of cell nuclei
4',6-diamidino-2-phenylindole (DAPI) was applied in a dilution of 1:1000 in PBS for 5
30
min. Finally after repeated washing, the staining was analyzed directly in the plates
using an inverted microscope (Olympus, IX70) with an UV lamp for fluorescent imaging.
2.3.3 Cell-ELISA
Cell-ELISA of primary ventral mesencephalic cultures was performed as previously
described (Grothe et al., 2000) in 96-well microtiter plates. Cells were fixed with precooled methanol for 10 min at -20 °C, followed by three times washing with PBS. Then,
the cultures were incubated for 10-15 min with 1% H2O2 in water to suppress
endogenous peroxidase activity. Nonspecific bindings were blocked by incubation with
PBS containing 10% horse serum, 1% goat serum and 0.3 % Triton X-100. After one
washing step with PBS cells were incubated with the primary monoclonal antibodies in
appropriate dilution in PBS containing 1% goat serum and 0.3% Triton X-100 overnight
at 4°C. Specificity controls with omission of primary antibodies were included. On next
morning the cultures were rinsed four times with PBS and a second blocking step was
applied for 20 min at room temperature using the same buffer as above. For detection of
bound primary antibodies a peroxidase-based Vectastain avidin-biotin complex (ABC) kit
was used according to the manufacturer’s instructions (Elite ABC kit). Briefly, the
secondary biotinylated anti-mouse IgG antibody was applied for 45 min at room
temperature in a dilution of 1:200 in PBS containing 1 % normal horse serum, 1%
normal goat serum and 0.3 % Triton X-100. The ABC complexes were formed for one
hour at room temperature. The solutions A and B were diluted 1:100 in PBS 30 min
prior to the complex formation. After three washing steps with PBS, 2,2'-azino-bis(3ethylbenzthiazoline-6-sulphonic acid) (ABTS) was added as a peroxidase substrate for
color development. The relative absorbance was measured at 405 nm with universal
microplate reader ELX800 (Bio-Tek Instruments, Inc.). Data were corrected for unspecific
reaction obtained in the absence of the first antibody.
31
2.3.4 Nuclear and cytoplasmic fractionation
The nuclear and cytoplasmic fractions of cultured cells and fresh or frozen tissue were
performed as described previously (Stachowiak et al., 1996, Reilly et al., 2004, Fang et
al., 2005).
In order to produce nuclear and cytoplasmic fractions from the cultured NB cells or SV40i-VM-NPCs, the cells were washed in ice cold PBS, and the PBS was thoroughly
drained away. A minimal volume (1 ml) of homogenization buffer was added to the 75
cm2 cell culture flask. Cells were gently scraped, transferred to an Eppendorf tube, and
dissociated with a 1000 µl pipette tip. In order to produce fractions from dissected tissue,
the tissue was homogenized in homogenization buffer using a green plastic pestle fitting
in an Eppendorf tube. The rest of the tissue was washed down from the pestle with
homogenization buffer. Following steps were identical for cells and tissue fractionation.
The cells were allowed to swell on ice for 15 min, after which 0.6% IGEPAL CA-630 was
added to the cell suspension. The cells were then vortexed for 1 min and the resolved
nuclei were pelletized by centrifugation in a microcentrifuge at 8000 rpm (Hermle, Z233
MK-2) for one minute at 4 °C. The supernatant representing the cytoplasmic fraction
with cytoplasmic membranes was removed and saved in a separate tube. The resulting
nuclear pellet was washed two times in homogenization buffer containing 0.6% IGEPAL
CA-630. The nuclei were repeatedly pelletized by centrifugation in a microcentrifuge at
8000 rpm for one minute at 4 °C and the supernatant was added to the cytoplasmic
fraction. For immediately following western blot assays, the nuclear pellet was dissolved
in 50-200 µl nuclear extraction buffer, sonicated two times for 10 sec with 57% power.
The nuclear proteins were allowed to dissolve on ice for 15 min, after vortexing 1-3 µl
were taken for BCA-protein assay. To denaturize the proteins 2x Laemmli buffer was
added, heated at 95°C for 5-10 min, vortexed and loaded on SDS-PAGE gel or stored
for subsequent electrophoresis. The nuclear extraction for immunoprecipitation assays
was performed as following. The nuclear pellet was resuspended in nuclear extraction
32
buffer containing 50 mM NaCl (~ same volume as the pellet: 50 – 100 µl) and allowed to
solve on ice for 30 min. The same volume of nuclear extraction buffer containing 250
mM NaCl was added, and the suspension was then briefly sonicated on ice.
2.3.5 BCA-assay
BCA (bicinchoninic acid) assay was performed according to manufactures protocol. The
concept of the assay includes in the first step the biuret reaction, where the bivalent
copper ions (Cu2+) are chelated by peptides containing more than 3 amino acids, and
reduced to monovalent Cu+. In the second step, the reduced copper ions are chelated
by BCA resulting in purple colored products, which can be measured quantitatively with
a photometer. The protein concentrations of the cell and tissue lysates were determined
using a BSA protein standard. The BSA standard was applied in 25 µl triplicates each
spanning concentrations 1.25 – 0.04 mg/ml. The probes were diluted 1:10 or 1:30 in 25
µl and applied next to BCA standard on the same 96-well microtiter plate. The BCA
assay reagent B and BCA reagent A were mixed 1:50, and 200 µl/well were immediately
applied to the plate. After 15 min incubation the relative absorbance was measured with
universal microplate reader ELX800 (Bio-Tek Instruments, Inc.) at 480 nm wave length.
All values were normalized to the absorbance measured by zero amount of protein.
Calibration curve was calculated for BSA standard and the relative protein amount was
estimated using the regression formula.
2.3.6 Co-Immunoprecipitation
The prepared nuclear and cytoplasmic fractions were first adjusted to an overall protein
concentration of 1 – 2 µg/µl in nuclear extraction buffer containing 150 mM NaCl and
finally diluted 1:2 with RIPA buffer. Equal amounts of protein extracts (0.5 – 1 mg) were
incubated with appropriate antibody overnight at 4°C in rotating 1.5 ml Eppendorf tubes.
To precipitate the rabbit IgG antibodies magnetic beads fused to protein A (Dynabeads,
Invitrogen) were used, since protein A has a high affinity to the Fc-regions of the rabbit
IgGs. Prior to the precipitation, the magnetic beads were washed once and than re-
33
suspended in RIPA buffer. The immunocomplexes were precipitated with 0.75 mg of
Dynabeads for 1 h at 4 °C in rotating tubes. The Dynabeads associated protein
complexes were pulled down with magnet, washed twice with RIPA buffer and twice with
PBS for 5 min on the rotator at room temperature, re-suspended in an appropriate
amount of 2x Laemmli buffer, and denaturized at 95 °C for 5 min in a water bath. The
free magnetic beads were pulled down and the protein lysates were further processed
with SDS-PAGE.
2.3.7 SDS-PAGE
Sodiumdodecylsulfat polyacrylamid gel electrophoresis (SDS-PAGE) was used to
separate the denaturized proteins surrounded by negatively charged SDS ions
according to their molecular weight. The polymerized polyacrylamid gels build a porous
matrix in which the proteins of different sizes show assorted migration patterns following
the electromagnetic field gradient. A tris-glycin-SDS system (Laemmli buffer) was used,
which includes 2-mercaptoethanol as a reduction reagent for disulfide bonds (Laemmli,
1970). Using the Bio-Rad mini Protean 3 system, first, the separating gel was
polymerized according to the manufacturer’s instructions. The proteins were cumulated
in the first phase, the stacking gel with higher pores and lower pH (6.8). The separation
occurred in the second phase, the separation gel with higher pH (8.8), while the
normally smaller pore size was adjusted by different acrylamide concentrations
according to the size of expected protein. For proteins over 50 kD the 8% acrylamid
separation gels were used, if co-detection of smaller proteins (e.g. histones as loading
control) was necessary 10% gels were used. The SpectraTM multicolor broad range
protein ladder was applied to identify the molecular weight of the particular protein band.
2.3.8 Western blot assay
In western blot assay the separated protein bands were transferred to a Highbond ECL
nitrocellulose membrane (Ammersham, # RPN68D) using methanol containing transfer
buffer in Bio-Rad mini Protean 3 system for blotting at 120 V for 1h. After protein transfer
34
the nitrocellulose membranes were blocked with 5 % low fat milk powder in TBS-T buffer
for 1 h at room temperature. For detection of specific protein bands the appropriate
primary antibodies were applied in blocking solution at the rotating shaker in a 50 ml
centrifuge tube for 1 h at room temperature followed by incubation over night at 4 °C.
After washing in TBS-T the membrane was incubated with secondary antibodies
coupled to horse radish peroxidase in blocking solution for 1 h at room temperature and
finally washed. For detection, the chemiluminescent detection reagent was applied on
the blotting membrane according to manufacturer’s protocol and the specific protein
bands on the blotting membrane were visualized with the chemiluminescence imager
(INTAS Science imaging systems GmbH, Göttingen, Germany). The relative
densitometric evaluation of single protein bands from the blot image was performed with
the LabImage 1D.
2.3.9 Genotyping
The extraction of genomic DNA was performed with alcohol precipitation method. This
method basically utilizes the circumstance of spontaneous precipitation of DNA after
application of alcohol in the presence of monovalent salt. The sampled tissue was predigested in 100 µg/ml proteinase K in NaCl-buffer over night at 56° C in a water bath.
The alcohol precipitation was performed by addition of the same volume of isopropanol
to the DNA-salt solution. After centrifugation for 5 min at 13 000 rpm the pellet was
washed with 70% ethanol, dried on air, and solved in distilled water.
For amplification, 1 µl of extracted DNA was applied to 24 µl of the master mix including
Taq DNA polymerase I and amplified under following conditions in the PCR
thermocycler: first denaturizing at 95°C for 3 min followed by 31 cycles composed of
denaturizing:
95°C for 30 sec,
annealing:
58°C for 30 sec,
synthesis:
72°C for 1 min,
35
finishing with a final prolonged synthesis for 5 min at 72°C.
Following specific primers were used to generate products of 470 bp for wild type and
820 bp for mutant allele:
FGF-2_GT2_wtF: 5’-CTCCTGGCCTTAACCCTTTCT-3’;
FGF-2_GT2_wtR: 5’-GAGGGATCAAGTCAGGCTTTG-3’
FGF-2_GT2_NeoR: 5’-CCCGTGATATTGCTGAAGAGC-3’
The products were diluted 1:6 in the 6x Loading Dye Solution, loaded on the 1%
agarose gels supplied with 0.002% ethidium bromide, and separated via electrophoresis
in TBE buffered system. The ethidium bromide-DNA complexes were visualized using
UV light. 1 kB DNA ladder was used to control the sizes of the separated DNAfragments.
The genotyping for the TgEGFP mice was performed via comparative analysis of the ear
skin fragments under the UV light (using FITC filter λ = 488 nm).
2.3.10 Quantitative RT-PCR
The quantitative real time polymerase chain reaction (qRT-PCR) is a method of DNA
amplification, which enables a quantitative measurement of the amplification products.
The method utilizes fluorescent dyes, which intercalate with the double stranded DNA.
The proportional increase of the fluorescence signal is quantified at every cycle and the
relative quantification of the amplified products is evaluated at the exponential phase of
the amplification plot.
To quantify the relative expression levels of certain genes, the total RNA was extracted
from ventral mesencephali of E14.5 and P0 FGF-2+/+ and FGF-2-/- littermates. The tissue
was dissected in ice cold PBS, snap frozen in liquid nitrogen and stored at -80°C.
Animals were genotyped and for each genotype individual tissue samples were pooled,
36
resulting for E14.5 stage in 6 and for P0 stage in 5 animals per pooled sample. For RNA
extraction tissue was homogenized in Trizol reagent and total RNA was extracted as
recommended by the manufacturer. To eliminate any genomic DNA contamination a
DNase digest was performed. Total RNA (1 µg) was converted into cDNA with the
iScript cDNA synthesis kit using random hexamers (BioRad). Quantitative RT-PCR was
performed in duplicates per sample and gene in 96-well plate format. The 14 µl reaction
mix contained 5 µl cDNA (corresponding to 2.5 ng RNA), 3 µl primer mix (5.25 pmol of F
and R primers) and 7 µl Power SYBR-Green PCR Master Mix (Applied Biosystems).
The
amplification
of
specific
gene
fragments
and
simultaneous
quantitative
measurement were done in StepOnePlus thermocycler (Applied Biosystems) applying
the standard program (95°C for 10 min, followed by 40 cycles of 95°C for 15 sec and
60°C for 1 min). Finally, a dissociation curve was calculated for each well, to ensure
specificity of the PCR product. The measurements for investigated genes were
normalized to the housekeeping gene Hprt, and the 2(-∆∆Ct) method was applied to
calculate mRNA levels (Livak and Schmittgen, 2001), which were given as the fold
change compared to wild type littermates.
2.4 Histology
2.4.1 Tissue processing
Pregnant mice were sacrificed by cervical dislocation and the embryos (E14.5) were
extracted from the uterus. Newborn mice (P0) were sacrificed by decapitation. The
brains of the embryos and newborn mice were fixed by immersion over night at 4°C in 4
% PFA in PBS. The juvenile mice (P28) were transcardially perfused with 5 ml ice-cold
0.9 % NaCl, afterwards with 20 ml 4 % PFA, in PBS. The tissue was postfixed in 4 %
PFA in PBS over night at 4°C and cryoprotected in 30 % sucrose in PBS. The fixed and
cryoprotected brains of embryonic and newborn mice were embedded in Tissue Tec
OTC compound. Serial coronar sections were sampled on cryostat; 40 µm thick free
floating sections of juvenile and newborn mice were harvested in PBS, 20 µm thick
sections of embryonic mice were sampled on the Superfrost plus object slides. For
37
storage the free floating slices were transferred to anti-freeze solution, while the
adherent slices were dried and stored at -20 °C in slide boxes.
2.4.2 In vivo BrdU-labeling
Bromodeoxyuridine (BrdU) is a synthetic nucleoside, which can be incorporated into
newly synthesized DNA in living cells substituting thymidine during replication. Detection
of BrdU with specific antibodies in fixed tissue allows the recognition of proliferating
cells. A single BrdU pulse (pulse chase experiment) several hours before sacrifice
allows detection of cells, which exited the cell cycle, and monitoring their fate (Lahti et
al., 2011). Therefore, time pregnant mice received intraperitoneal BrdU injections (100
µg/mg body weight, diluted if necessary in PBS) applied in 1 ml solution with a 25 G
cannula. The pregnant mice were sacrificed 20 h after BrdU-injection, embryos were
harvested at E14.5, and the proliferation of Lmx1a-ir DA precursor cells was quantified.
2.4.3 Tyrosine hydroxylase immunohistochemistry
For stereological cell counts of TH-ir cells tissue was processed as previously described
(Nikkhah et al., 1994) with minor modifications. Briefly, the free floating sections of
newborn mice were postfixed for 20 min in 4% PFA in PBS. For handling of the free
floating slices sections, cell strainer (100 µm nylon, BD Falcon, #8343719) were used as
trays in 6-well multi-dishes to minimize the destruction and loss of the tissue. All sections
were pre-cleared for 5 min in 1% hydrogen peroxide (H2O2) and 10% methanol in PBS.
The unspecific bindings were blocked for 1.5 h in ICC blocking buffer containing 0.5 %
Triton X-100. Incubation with primary rabbit anti-TH antibody was done over night by 4
°C, followed by incubation with secondary biotinylated anti-rabbit antibody for 1.5 hours
at room temperature. For enhancement, the avidin-biotin-complex formation was
performed with Vectastain ABC Kit according to the manufactures protocol. The
antibodies were visualized for 3-5 min with aminoethylcarbazol (AEC): one tablet of AEC
was solved in 2.5 ml N,N-dimethylformamide (DMF), diluted in 50 ml of 80 mM acetate
38
buffer (pH 5) and activated with 0.05 % H2O2. The slices were rinsed in PBS and finally
mounted in water based mounting medium to prevent excessive tissue shrinkage.
2.4.4 Antigen retrieval
Often occurring artifact after aldehyde fixation is the masking of the antigen epitopes by
excessive cross-linking of the tissue by paraformaldehyde, which disables the antibodies
to recognize the specific antigen binding sides. The retrieval of the antigen epitopes was
achieved by heating of PFA-fixed slices in citrate buffer (pH 6). The dried adherent
slices were rehydrated in PBS prior to the incubation in citrate buffer. Then, the slices
were transferred to the citrate buffer, preheated in the microwave avoiding bubbles,
incubated for 10 min in 95 °C water bath, and cooled down for 10 min in the open
cuvette at the room temperature before transferring to the PBS.
2.4.5 Fluorescence immunohistochemistry
For fluorescence immunohistochemistry the slices were blocked and permeabilized for 1
h in blocking buffer. Incubation with primary antibodies was done over night by 4 °C in
carrier solution. The incubation with secondary antibodies labeled with fluorophores was
performed for 1 h at room temperature. Due to cross-reaction of the secondary antimouse antibodies with the endogenous IgGs in not perfused tissue a direct labeling of
primary mouse antibodies was necessary before applying to the embryonic and
postnatal mouse tissue. The direct labeling was performed with fluorophore-conjugated
Fab fragments (Zenon Kit), the antigen binding parts of the antibody, according to the
manufacturer’s protocol. The fluorophore conjugated anti-mouse Fab fragments were
first pre-adsorbed and then applied to the primary antibodies. The labeled primary
antibodies were diluted in carrier solution buffer and applied to the tissue for 2 h at room
temperature. Some primary antibodies worked also in not perfused tissue if detected
with specific anti-mouse IgG1 secondary antibodies. For the detection of BrdU labeled
cells, tissue was first denaturized with 2 M HCl for 1 h at room temperature; followed by
39
5 min neutralization in 0.1 % borate buffer (pH 8.5). Antigen retrieval procedure was
performed for cleaved caspase-3 and the direct labeling method.
2.5 Microscopic analysis and morphometry
2.5.1 Microscopic imaging
The epifluorescence imaging was performed with an Olympus BX60 fluorescence
microscope equipped with digital CCD ColorView-III camera (Olympus) and CellP
software (Olympus). If necessary Multiple Image Alignments (MIA) were performed for
fluorescent signals and for bright field microscopy on BX60 microscope using the CellP
software. The evaluation of F-ICC in 96-well plates was performed with an inverse
Olympus IX70 fluorescence microscope equipped with digital CCD Colorview-II camera.
The multichannel fluorescence images were merged with the CellP software. For detail
images a Leica TCS SP2 confocal microscopy set up was used with oil immersion
objectives HCX PL APO BL (63x, numerical aperture 1.4). If necessary the brightness
and contrast were adjusted with Adobe Photoshop software uniformly for all images
taken for one experiment.
2.5.2 Evaluation of proliferation
Proliferation was investigated in every 6th section (20 µm thickness) of E14.5 brains by
counting the number of cells positive for mitotic marker Ki67 within the TH-ir region of
the ventral mesencephalon. Multiple Image Alignment of the ventral mesencephalon
acquired at 10x objective magnification was performed with the CellP software
(Olympus). The TH-ir profile of each section was divided in three independent zones,
defined as: ventricular zone (VZ), subventricular zone (SVZ) and mantle zone (MZ). The
Ki67-ir cells within each zone were quantified using ImageJ software at four VM sections
per animal, covering the prospective SNpc and VTA region, excluding the dorsocaudal
part of the VTA.
40
The quantification of cells positive for BrdU and Lmx1a was carried out on four rostral
and four caudal coronal brain sections (20 µm thickness, 60 µm distance), separately for
VZ and SVZ using ImageJ software.
2.5.3 Evaluation of apoptosis
Apoptotic cells were counted in every 4th section (40 µm thickness) of P0 midbrains,
which were processed for TH, cleaved caspase-3 and DAPI staining. The complete THir region of the SNpc was screened systematically with the Olympus BX60 fluorescence
microscope (40x, numerical aperture 0.75) using the ocular grid. Cells positive for
cleaved caspase-3 (cC3+), for cC3+ and apoptotic bodies (ApB/cC3+), or only for ApB
were counted separately.
2.5.4 Stereological cell counts
Stereology is an interdisciplinary field which allows a quantitative interpretation of three
dimensional materials on two dimensional planar sections. Among several other
methods, the two dimensional (2D) or model based technique is widely used to estimate
the number of cells in specific brain regions. Basically, the planar sections are evaluated
within the full volume of the sampling side at predetermined intervals using a 2D
counting frame. This method is influenced by errors like over-counting and “lost caps”
due to evaluation of the entire volume of the z-axis within the sampling side (Guillery,
2002, Baryshnikova et al., 2006). Over-counting is simply the circumstance, that split
cells are being counted twice in both consecutive slices, if an extensive evaluation of all
slices is performed across the entire z-axis without exclusion of top and bottom guard
zones as described later for the 3-d method. This problem remains when random
fractions or series of the volume are evaluated (Guillery, 2002). Therefore a correction
factor was introduced (Abercrombie and Johnson, 1946, Guillery, 2002). In contrast, the
loss of cell fragments at the tissue surface due to extensive washing procedures, the so
called “lost caps” problem, remains an unpredictable influence. Optical fractionator is a
more precise 3D design based approach to count cells, which remains unbiased by
41
errors like over-counting or “lost caps” fragments. Since a cube shaped 3D counting
frame allows an additional 3rd exclusion boarders in the z-axis (top and bottom guard
zones) (West, 1993, West et al., 1991, Keuker et al., 2001). However, other biasing
errors like tissue shrinkage and z-axis distortion should be avoided by adequate tissue
processing. The cryosectioned tissue is suitable for the optical fractionator method
(Dorph-Petersen et al., 2001).
Depending on the needed accuracy the decision between 2D and 3D should fall
according to the expected differences between the probes, in case two individual groups
are compared with each other. If the hypothetical differences are small, the more
accurate method like optical fractionator should be used. The relatively high errors of
less accurate methods like uncorrected 2D counts would hide the minor variations
(Guillery and Herrup, 1997, Hedreen, 1998).
All morphological procedures were performed using an Olympus Optical (Tokyo, Japan)
microscope (BX 60) with a motorized stage. The stereological cell counting was
performed with the Olympus optical CAST-grid system. The measurements in the z-axis
were done using the electric microcator (ND 281; Heidenhain, Traunreut, Germany).
The ventral mesencephalon of E14.5, SNpc of P0 as well as SNpc and VTA of P28 mice
was quantified with regard to the number of mDA neurons and area of TH-ir profile.
During all morphometric measurements the experimentator was blinded, using coded
sections.
Model based cell counting (2-D method)
Due to small tissue size of E14.5 midbrains, sections were directly mounted on glass
slides before processing for TH immunohistochemistry, which required the handling of
thinner sections (20 µm) ought to assure a continuous staining throughout the z-axis.
After immunohistochemical processing the final tissue thickness was insufficient for the
3D counting method. So, the model based (2D) counting method was used for stage
E14.5 to quantify the number of TH-ir cells in the whole ventral mesencephalon, as the
42
discrimination of the VTA and SNpc mDA neuron subtypes is not yet possible in
embryonic brain. The region of interest was outlined in every third section (60 µm apart)
at 10x objective magnification. All TH-ir cells within the section profile were counted at
100x objective magnification (oil immersion, numerical aperture 1.25). The total number
of THir cells per hemisphere was corrected for split nuclei using the formula of
Abercrombie (Abercrombie and Johnson, 1946). For calculation of the correction factor,
the measurements of the cell diameter and tissue thickness were done using the electric
microcator.
Design based cell counting (3-D method)
Total numbers of TH-ir cells were estimated using fractionator-sampling design
(Gundersen et al., 1988, West et al., 1991). Sections used for counting (in newborn mice
every 2nd and in juvenile every 3th) covered the entire substantia nigra and ventral
tegmental area, respectively, starting with the first appearance of TH-ir neurons,
extending to the most caudal parts of VTA. The borders of the SNpc at all levels in the
rostrocaudal axis were delineated at 10x magnification. The medial border was defined
by a vertical line passing through the medial tip of the cerebral peduncle and by the
nerve of the accessory oculomotor nucleus, when present in sections (Fig. 6), thereby
excluding the TH-positive cells in the VTA. Photomicrographs of consecutive sections
double processed for TH and Calbindin proved the VTA/SNpc boarder delineation, as
more mDA neurons positive for Calbindin are located in the VTA (Fig.3, C and D). The
ventral border followed the dorsal border of the cerebral peduncle, thereby including the
TH-positive cells in pars reticularis (SNpr), and the area extended laterally to include the
pars lateralis in addition to pars compacta (Kirik et al., 2001).
43
Figure 6. Substantia nigra of wild type (wt) and FGF-2 deficient (ko) neonatal mice. The overview
shows TH-immunostained sections (120 µm apart) from four P0 animals, beginning with rostral part of VM
(top) and ending with caudal sections (bottom). The transverse line defines the medial border between
VTA and SNpc. Scale = 1 mm.
44
In each section TH-ir neuron somas and not the nuclei, as in other histochemical
approaches, were used as the counting unit according to optical dissector rules
(Gundersen et al., 1988). The counting frame was placed randomly on the first counting
area and systematically moved through all fields. Cell counts were made under 100x
magnification (objective: numerical aperture, 1.25) at regular predetermined intervals (x
and y step length of 81 µm for newborn brains; and 121 µm for postnatal day 28) within
an unbiased counting frame of known area (for newborn 50 µm x 39 µm, and 80 µm x
64 µm for juvenile). Only the profiles that came into focus within the counting volume
(dissector height for newborn brains 12 µm, for juvenile 17 µm) were counted. The total
number of neurons in one hemisphere was estimated according to the optical
fractionator formula (West et al., 1991).
2.5.5 Nigrostriatal fiber outgrowth
The nigrostriatal organotypic co-cultures were evaluated for the outgrowth of the TH-ir
fibers from the VM explant to the FB explant. The evaluation parameters were the length
and the distance of the fiber outgrowth, as well as the width of the TH-ir bundle growing
into the forebrain explant. The quotient of the distance and length was a measure for a
directed outgrowth. The measurements were carried out on original partially merged
multichannel MIAs of the co-cultures acquired with the Olympus BX60 fluorescence
microscope using the CellP software (Olympus). For measuring the length of outgrowth
the fiber bundle was tracked with the polygon line tool (mean measurements/co-culture
n = 12) at regular intervals beginning at the boarder of ventral mesencephalic and
forebrain explant, following the outgrowth direction of the fibers until the longest fiber
ended. The border was defined as the edge of the EGFP-positive VM explant. The
distances (mean measurements/co-culture n = 12) of the tracked fibers were measured
by putting perpendicular lines to the VM/FB boarder at the very same intervals. The
width of the tract was measured by putting straight lines crossing the direction of the
fiber outgrowth (mean measurements/co-culture n = 7) at the basis of the TH-ir tract in
the FB explant before the fibers started to spread.
45
2.6 Statistics
Statistical analysis was performed in Microsoft Excel or with GraphPad Prism 4
software. The data is expressed as means ± SEM. After the data passed the tests for
normality (Kolmogorov-Smirnov-test) and equal variances (F-test), the unpaired
Student’s t-test was applied to test for statistically significant differences between mutant
and wild type groups. In case the data did not pass the normality test the Mann-Whitneytest was applied. The significance niveau was defined as following: * - p ≤ 0.05; ** - p ≤
0.01 and *** - p ≤ 0.001. TH-ir cell numbers of the differently transfected groups were
tested by Kruskal-Wallis-test (including Dunns post hoc test).
2.7 Materials
2.7.1 Chemicals
Name
Agarose, UltraPure
2-Mercaptoethanol
1 kB DNA Ladder
3-Amino-9-ethyl-carbazole
5-Bromo-2'-deoxy-uridine (BrdU) Labeling and Detection
Kit I
6x Loading Dye Solution
Acetic acid
Acrylamide/Bis, 30% solution, 29:1
Amonium Perulfate (APS)
B-27 supplement
BCA Protein Assay Reagent A
BCA Protein Assay Reagent B
Boric acid
Bovine Serum Albumine (BSA)
Bromphenol blue
Clean-Blot IP Detection Reagent (HRP)
collagen, rat tail
Comlete, EDTA-free – Proteinase inhibitor coctail
D(+)-Sucrose
Dako Fluorescent mounting medium
DAPI, dilactate
dbcAMP
Dimethylsulfoxide (DMSO)
Dithiothreitol (DTT)
DMEM high glucose (4,5 g/l)
Distributor
Invitrogen, Spain
SIGMA, Steinheim, Germany
Fermentas, Thermo Scientific, USA
SIGMA, Steinheim, Germany
Catolog #
15510-027
63689
SM0311
A6926
Boehringer Mannheim, Germany
Fermentas, Thermo Scientific, USA
J.T.Baker, Deventer, NL
BIO-RAD, Hercules, CA
SIGMA, Steinheim, Germany
GIBCO/ Invitrogen, Auckland, NZ
Thermo Scientific, Rockford, USA
Thermo Scientific, Rockford, USA
Carl Roth GmbH, Karlsruhe
PAA, Pasching, Austria
SIGMA, Steinheim, Germany
Thermo Scientific, Rockford, USA
SIGMA, Steinheim, Germany
Roche, Indianapolis, USA
Carl Roth GmbH, Karlsruhe
DAKO, Glostrup, Denmark
SIGMA, Steinheim, Germany
SIGMA, Steinheim, Germany
SIGMA, Steinheim, Germany
SIGMA, Steinheim, Germany
PAA, Pasching, Austria
1296736
R0611
6052
161-0156
A-3678
17504-044
23228
23224
5935,1
K31-011
B8026
21230
C7661
11873580001
4661,1
S3023
D96564
D0627
D2650
D-9779
E15-810
46
DMEM/Ham's F12
DMEM/Ham's F12 powder
DNase I
Dynabeads, Protein A
Ethylene glycol tetraacetic acid (EGTA)
Ethidium bromid (10 mg/ml)
Ethylenediaminetetracetic acid (EDTA)
Ethylenglycol
Fetal Calf Serum (FCS)
Glial cell line-derived Neurotrophic Factor (GDNF)
Gelatin
Glycerol ReagentPlus
Glycin
Goat serum
Hank's BSS
HEPES 1M
Hydrogen peroxide
IGEPAL CA-630
Immobilon Western Chemiluminiscent HRP Substrate
iScript cDNA synthesis kit
Laminin, mouse
L-Ascorbic acid 2-phosphate sesquimagnesium salt
hydrate
L-Glutamine 200 mM
Lipoofectamine 2000 Reagent
Metafectene Pro
Methanol
N,N,N',N'-Tetramethylenediamine (TEMED)
N,N-Dimethyformamide
N2-supplement
OPTI-MEM I
Paraformaldehyde (PFA)
PBS Dulbecco Instamed (9.55 g/l)
Penicillin/streptomycin 100x
Peroxidase Substrate Kit ABTS
PhosStop - Phosphatase inhibitor coctail
Polyornithin-hydrobromide
Potassium chloride
PowerSYBR-Green
Proteinase K
Recombinant human FGF-2
Sodium dodecyl sulfate (SDS), ultra pure
Sodium acetate
Sodium bicarbonate 7,5 % solution
Sodium chloride (NaCl)
Sodium citrate tribasic dihydrate, ACS reagent
Sodium flourid (NaF)
PAA, Pasching, Austria
GIBCO, Auckland, NZ
Roche, Indianapolis, USA
Invitrogen, Oslo, Norway
Carl Roth GmbH, Karlsruhe
Carl Roth GmbH, Karlsruhe
Carl Roth GmbH, Karlsruhe
Merk, Darmstadt, Germany
SIGMA, Steinheim, Germany
PeproTech, London, UK
Merk, Darmstadt, Germany
SIGMA, Steinheim, Germany
Carl Roth GmbH, Karlsruhe
GIBCO, Auckland, NZ
PAA, Pasching, Austria
GIBCO, Auckland, NZ
SIGMA, Steinheim, Germany
SIGMA, Steinheim, Germany
Millipore, Bilerica, USA
BioRad, München, Germany
BD Biosciences, Bedford, USA
E15-012
42400
11284932
100.01D
3054.1
2218,1
8043.1
1.09621.1000
F7524
450-10
1.04078.1000
G7757-1L
3908,2
16210-064
H15-010
15630-049
H1009
I8896
WBKSLS0100
170-8891
354232
SIGMA, Steinheim, Germany
PAA, Pasching, Austria
Invitrogen, Carlsbad, USA
Biontex, Martinsried, Germany
J.T.Baker, Deventer, NL
SIGMA, Steinheim, Germany
SIGMA, Steinheim, Germany
GIBCO, Auckland, NZ
GIBCO, Auckland, NZ
SIGMA, Steinheim, Germany
Biochrom, Berlin, Germany
PAA, Pasching, Austria
Vector Lab. Inc., Burlingham, CA
Roche, Mannheim, Germany
SIGMA, Steinheim, Germany
Merk, Darmstadt, Germany
Applied Biosciences, Warrington, UK
Merk, Darmstadt, Germany
PeproTechInc., USA
Carl Roth GmbH, Karlsruhe
SIGMA, Steinheim, Germany
GIBCO, Auckland, NZ
J.T.Baker, Deventer, NL
SIGMA, Steinheim, Germany
SIGMA, Steinheim, Germany
A8960
M11-004
11668-027
05.2012
8045
T9281
D-4551
17502-048
31985-047
15,812-7
L1882-50
P11-010
SK-4500
04906837001
P3655
1465560
4367659
1.24568
100-18B
2326,2
71183
25080
0278
S4641
S-1504
47
Sodium pyrovate solution 100 mM
Sodium-orthovanadate (NaVO3)
Spectra Multicolor Broad Range Protein Ladder
SuperSignal West Femto Maximum Sensitivity Substrate
Taq DNA Polymerase I
Tissue Tec OTC compound
Tris(hydroxymethyl)-aminomethane (TRIS)
Triton X-100
Trizol reagent
Trypan blue
Trypsin-EDTA (0,05% in PBS)
Tween 20
PAA, Pasching, Austria
S11-003
SIGMA, Steinheim, Germany
S-6508
Fermentas, Thermo Scientific, USA
SM1841
ThermoScientific, Rockford, USA
34095
Purified in our lab from DH5a bacteria transformed
with pTAQ vector (Desai and Pfaffle, 1995)
Sakura, Zoeterwoude, NL
4583
Carl Roth GmbH, Karlsruhe
4855,2
Roche, Mannheim, Germany
12134300
GibcoBRL, Auckland, NZ
15596-018
SIGMA, Steinheim, Germany
T8154
PAA, Pasching, Austria
L11-004
Roth Gmbh, Karlsruhe, Germany
9127,1
2.7.2 Cell culture media
DMEM
10 % FCS
NB cell medium
1 mM sodium-pyruvate
0.1 mg/ml penicillin/ streptomycin
DMEM/Ham´s F12
N2-supplement (1 :100)
0.25% BSA
N2-Medium
2 mM glutamine
1 mM sodium-pyruvate
0.1 mg/ml penicillin/ streptomycin
DMEM/Ham´s F12
2 mM glutamine
Collection medium
B27-supplement (1:50)
1 mM sodium-pyruvate
DMEM/Ham´s F12
3% FCS
20 ng/ml human basic FGF
B27-supplement (1:50)
Attachment medium
N2-supplement (1 :100)
0.25% BSA
2 mM glutamine
1 mM sodium-pyruvate
0.1 mg/ml penicillin/ streptomycin
48
DMEM/Ham´s F12
Human FGF
N2-supplement (1:100)
Proliferation medium
0.25% BSA
2 mM glutamine
1 mM sodium-pyruvate
0.1 mg/ml penicillin/ streptomycin.
DMEM/Ham´s F12
0.25% BSA
B27-supplement (1:50)
Differentiation medium
3% FCS
2 mM glutamine
100 µM ascorbic acid
0.1 mg/ml penicillin/ streptomycin.
DMEM/Ham´s F12
0.25% BSA
OTC-medium
2 mM glutamine
0.1 mg/ml penicillin/ streptomycin
2.7.3 Buffers and gels
In A.dest
Reconstitution buffer
250 mM HEPES
1.125 M NaOH
Sterile filtered
PBS
ICC blocking buffer
0.3% Triton X-100
5% normal goat serum
1% BSA
PBS
ICC carrier solution
0.3% Triton X-100
1% normal goat serum
1% BSA
19% of Solution A: 0.2 M NaH2PO4·H2O
Soerensen phosphate buffer
81% of Solution B: 0.2 M Na2HPO4·H2O
49
137 mN NaCl
20 mM Tris-HCl pH 7.5
25 mM sodium glycerophosphate
2 mM EDTA
RIPA buffer
1 mM sodium-orthovanadate
1% Triton-X 100
1% sodium-deoxycholate
Complete protease inhibitor coctail (1:50)
10 mM HEPES
10 mM KCl
0.1 mM EDTA
0.1 mM EGTA
2 mM dithiothreitol (DTT)
Homogenization buffer pH 7.9
25 mM NaF
1 mM NaVO3
Complete protease inhibitor coctail (1:50)
PhosStop (1:20) (for detection of protein
phosphorylation)
1% Triton-X 100
50 mM Hepes
50 mM NaCl
Nuclear extraction buffer pH 7.5
5 mM EDTA
1mM DTT
Complete protease inhibitor coctail (1:50)
137 mM NaCl
TBS-T buffer
20 mM Tris-HCl pH7,6
0.1 % (v/v) Tween-20
125mM Tris-HCl (pH 8,8)
4% SDS
20% Glycerol
Laemmli buffer
2 mM EDTA
Bromphenol blue
10% 2-Mercaptoethanol (added straight before)
50
25 mM Tris
Electrophoresis buffer
192 mM Glycin
3.5 mM SDS
25 mM Tris
Transfer buffer
192 mM Glycin
20 % Methanol
100 mM Tris
5 mM EDTA (pH 8.0)
NaCl-buffer
(Proteinase K buffer)
0.2% SDS
200 mM NaCl
100 µg/ml Proteinase K
10% of 10x Taq Pol I buffer
0.2 mM dNTPs
1x master mix for genotyping
0.01 µg/µl Primer
1% Taq DNA Polymerase I
500 mM KCl
10x PCR buffer
100 mM Tris (pH 8.3)
15 mM MgCl
89 mM Tris
TBE buffer
89 mM Boric acid
2 mM EDTA
40% PBS
30% Glycerol
Anti-freeze solution
30% Ethylenglycol
15 mM Acetic acid
Acetate buffer (80 mM), pH5
80 mM Sodium acetate
1,7 g sodium citrate tribasic dehydrate
Citrate buffer
in 500 ml A. dest.
pH 6, adjusted with HCl
51
10% Acrylamid/Bis (1:29)
375 mM Tris (pH 8,8)
Separating gel, 10%
0.1% SDS
0.1% APS
0.04% TEMED
8% Acrylamide/Bis (1:29)
375 mM Tris (pH 8,8)
Separating gel, 8%
0.1% SDS
0.1% APS
0.06 µl TEMED
5% Acrylamide/Bis (1:29)
125 mM Tris (pH 8,8)
Stacking gel
0.1% SDS
0.1% APS
0.1% TEMED
2.7.4 Plasmids
Expression plasmids were derived from pCAGGS plasmid, which contained the CAG-promoter,
provided by Dr. Hitoshi Niwa, RIKEN Center for Developmental Biology, Japan (Niwa et al.,
1991). Cloning of the c-terminal 3xFLAG tagged enhanced green fluorescence protein (EGFP)
expression plasmid pCAGGS-EGFP-FLAG (R412) has been described previously (Ratzka et al.,
2011 accepted). The coding sequence of 18 kDa rat FGF-2 (NM_019305.2, 533 – 994 bp), rat
FGF-8b (corresponds to rat FGF-8a NM_133286.1, 1 – 612 bp, with an 33 bp insertion of
GTAACTGTTCAGTCCTCACCTAATTTTACACAG
between
69
and
70bp),
rat
FGF-15
(NM_130753.1, 1 – 654 bp) and rat FGF-17a (corresponds to rat FGF-17b NM_019198.1, 1 –
648, without 33 bp CAGGGGGAGAATCACCCGTCTCCTAATTTTAAC between 69 – 103 bp)
was amplified by PCR from rat E12 embryonic cDNA (PCR primer sequences are available upon
request). EcoRI- or MfeI-sites followed by a kozak sequence were introduced by the forward
primer and the stop-codon was replaced by XbaI-site by the reverse primer, which allowed in
frame cloning to the 3xFLAG tag of the EcoRI/XbaI digested pCAGGS-FLAG plasmid backbone.
Thereby, the FGF expression plasmids pCAGGS-FGF2-18kDa-FLAG (R417), pCAGGS-FGF8bFLAG (R421), pCAGGS-FGF15-FLAG (R423) and pCAGGS-FGF17a-FLAG (R424) were
generated.
52
2.7.5 Primer
The primer sequences were designed with the primer3 software if not assigned. The
oligonucleotides were ordered from Eurofins MWG Operon (Ebersberg, Germany).
Gen
Sequence 5' 3'
CTTGTCGGATTTAGGAGGCTG
Aldh1a1 387-523
BDNF 869-973
DAT 880-957
Ddc 1027-1123
Drd1a 1149-1249
Drd2 575-685
En1 1322-1461
En2 1469-1597
Foxa1 415-519
Foxa2 118-201
Hprt1 256-442
Lmx1a 815-939
Ngn2 254-377
Nurr1 464-538
Pax2 761-867
Pitx3 121-200
TH 1326-1409
Vmat2 1074-1180
Source
f
r
f
r
f
r
f
r
f
r
f
r
f
r
f
r
GGCCACACACTCCAATAGGTT
GAAGAGCTGCTGGATGAGGAC
CAAAGGCCACTTGACTGCTGAG
CTGCCTGGTGCTGGTCATT
ATCCACACCACCTTCCCTGA
CCAGACACCATGAACGCAAGT
GTCCCTCAATGCCTTCCATGT
GTCCTGTCCTTCACCATCTCTTG
CGAGACGATGGAGGAGTAGACC
ATCGTCACTTACACCAGTATCTACAGGA
GTGGTCTGGCAGTTCTTGGC
TCCTACTCATGGGTTCGGCTA
CTTCTTTAGCTTCCTGGTGCG
ATTCTGACCGGCCTTCTTCAG
TCTGAAACTCAGCCTTGAGCC
ATGAGAGCAACGACTGGAACA
TCATGGAGTTCATAGAGCCCA
f
GGAGCCGTGAAGATGGAAGG
r
f
r
f
r
f
r
f
r
f
r
f
r
f
r
GTTGCTCACGGAAGAGTAGCC
TTCCTCATGGACTGATTATGGACA
AGAGGGCCACAATGTGATGG
ACCATCCTGACCACTCAGCAG
ACCTGAACCACACGGACACTC
GACATTCCCGGACACACACC
GCCCAGCAGCATCAGTACCT
CAGTATGGGTCCTCGCCTCA
GCTGTATTCTCCCGAAGAGTGG
GTCTTCCAGGCATCAGAGCAC
GCCGATGCAGATAGACTGGAC
GCACTAGACCTCCCTCCATGG
TGCGTCCGATAACGACAGC
GCCAAGGACAAGCTCAGGAAC
ATCAATGGCCAGGGTGTACG
f
GAGACCATGTGTTCCCGAAAG
r
CCCATTTTGTGTGCAAGTATC
Schiff et al. 2009
Dr. A. Ratzka
Colebrooke et al 2007 (Brain Res,
1152)
Colebrooke et al 2007 (Brain Res,
1152)
Schiff et al. 2009
Dr. A. Ratzka
Schiff et al. 2009
Schiff et al. 2009
Schiff et al. 2009
53
2.7.6 Antibodies
Antigen
ß- Tubulin III
α-DAT
Origin
mouse
rat
Flg (C15) (FGFR-1)
FLAG M2
GFAP
GIRK 2
GFP (clone 7.1 & 13.1)
Histon H3
Ki67
rabbit
mouse
mouse
rabbit
mouse
mouse
rabbit
Nurr1/Nur77(E-20)
TH
TH
TH
cleaved Caspase-3
rabbit
mouse
rabbit
mouse
rabbit
Distributor
Upstate Biotechnology
Millipore
Santa Cruz Biotechnology,
Inc.
Sigma-Aldrich
Sigma-Aldrich
Alomone labs
Roche
Cell Signaling
abcam
Santa Cruz Biotechnology,
Inc.
Chemicon
Millipore
Sigma-Aldrich
Cell Signaling
FGFR-1
mouse
M. Stachowiak
FGFR-1
mouse
abcam
M19B2
1:1000
Calbindin D-28k
rabbit
Swant
CB-38a
1:4000
a-BrdU
mouse
Roche
11170376001
Lmx1a
Phospho-Erk1/2 Pathway
Sampler Kit
rabbit
Millipore
AB10533
rabbit
Cell Signaling
9911
1:2000/
1:1000
Phospho-Akt Kit
rabbit
Cell Signaling
9280
1:1000
β-Catenin
rabbit
95815
1:1000
RXR
rabbit
Cell Signaling
Santa Cruz Biotechnology,
Inc.
DAKO
M737
Molecular Probes, Invitrogen
Chemicon
Molecular Probes, Invitrogen
Molecular Probes, Invitrogen
Z25005
MAB374
A11001
A21422
rabbit IgG
Alexa 555 Zenon mouse
IgG1 labelling kit
GAPDH
a-mouse IgG-Alexa 488
a-mouse IgG-Alexa 555
rabbit
Goat, Fab
fragments
mouse
goat
goat
Catolog #
05-559
MAB369
Western
ICC/IHC
1:400
1:100
sc-121
F1804
G3893
APC-006
1 814 460
9715
ab15580
1:500
1:3000
1:1000
1:600
1:100
1:200
ELISA
1:140
Co-IP
Notes
5 ng/µl
For ICC: Blocking and carrier solution without
BSA;10% NGS in Soerensen buffer
1:600
Antigene retrieval
1:1000
1:500
sc-990
MAB 5280
AB 152
T 1299
9661
5 ng/µl
10% NGS in blocking
1:200
1:500
1:1000
1:200
1:400
2. AB: a-mouse IgG1
Antigene retrieval
1:40
For ICC: Blocking and carrier solution without
BSA;10% NGS in Soerensen buffer
1:100
2. AB: a-mouse IgG1
1:10000
10% NGS in blocking, washed in 0.1% BSA in PBS
Incubation of first antibodies in 5% BSA in TBS-T
1:500
5 ng/µl
1:500
1:4000
1:500
1:500
Antigene retrieval
54
Antigen
a-rabbit IgG-Alexa 488
a-rabbit IgG-Alexa 555
a-rat IgG Alexa-488
a-mouse IgG-HRP
a-rabbit IgG-HRP
a-mouse IgG1-Alexa 568
Origin
goat
goat
goat
sheep
donkey
Distributor
Molecular Probes (Invitrogen)
Molecular Probes (Invitrogen)
Molecular Probes, MoBiTec
Amersham
Amersham
Catolog #
A11008
A-21428
A11006
NA 931
NA 934
goat
Western
ICC/IHC
1:500
1:500
1:500
ELISA
1:4000
1:5000
Molecular Probes (Invitrogen)
A21124
1:200
a-rabbit ABC-Kit
VECTOR Laboratories, Inc.
PK-6101
1:200
a-mouse ABC-Kit
VECTOR Laboratories, Inc.
PK-6101
1:200
Co-IP
Notes
55
3. Results
3.1 Phenotype onset of the supernumerary TH-ir cells in SNpc
Our group previously showed that SNpc of adult FGF-2 deficient mice contains
significantly more TH-ir neurons than the wild type littermates (Timmer et al., 2007). To
find the time window where the supernumerary TH-ir cells develop within the SNpc, the
stereological counting of TH-ir cells at defined developmental stages (E14.5, P0 and
P28) was performed. Additionally to morphological parameters, which were described in
chapter 2.5.4, immunohistological double-labeling of SNpc mDA neurons by Girk2/TH or
VTA mDA neurons by Calbindin/TH stainings can be applied to discriminate SNpc and
VTA mDA neuron populations (Fig. 7 A) (Thompson et al., 2005).
Table 2. TH-ir cell number in VM of developing wild type and FGF-2 deficient mice.
TH cell number
P28 SNpc
P28 VTA
P0 SNpc
E14.5 VM
wt
ko
3654.7
5968.3
3806.1
5824.2
4997.3
6155.4
5104.8
5890.6
%
p
136
103
134
101
0,001
0,647
0,0005
0,986
The evaluation of SNpc and VTA in the P28 mice revealed an increase of 16% in TH-Ir
cells in FGF-2 deficient animals compared to wild type littermates (p < 0.05, Fig. 7 C,
right bars). When both mDA neuron populations were analyzed separately (Table 2), no
differences were found in the VTA region (Fig. 7 C, middle bars), whereas the SNpc of
FGF-2 deficient animals displayed a 36% increased number of TH-ir neurons compared
to wild type mice (Fig. 7 C, left bars). Thereby, a subtype specific effect of FGF-2
deficiency on mDA neurons located in the SNpc but not in the VTA has been
demonstrated.
56
Figure 7. FGF-2 deficiency increases mDA neuron number in the SNpc. Ventral mesencephalic TH-ir
neuron number was quantified at three developmental stages (A-C) juvenile mice (P28), (D-F) newborn
mice (P0) and (G-I) E14.5 embryos. Two regions containing mDA neurons, the lateral located SNpc and
the medial located VTA, can be distinguished either by (A) co-localization of Calbindin (red) and TH
(green) as yellow signal in the VTA or by (B) histological definition. (C) The SNpc of FGF-2 deficient
animals (FGF-2 ko) contains 36% more mDA neurons compared to wild-type littermates (*** p < 0.001),
whereas the VTA contains similar numbers of mDA neurons in both genotypes. The total number of mDA
neurons (SNpc and VTA) is increased by 16% in FGF-2 deficient animals compared to wild-type
littermates (p < 0.05). (D) Rostrocaudal distribution of TH-ir cells, which was quantified on 12 VM cross
sections (levels a - l, each 80 µm apart), was significantly increased at rostral level d and caudal level j in
FGF-2 deficient mice compared to wild-type littermates. Positions of both sections are indicated by solid
lines on P0 sagittal sections stained for TH-ir. (E) The dotted line surrounds the TH-ir cells of the caudal
SNpc of a newborn mouse. (F) Stereological quantification revealed 34% increase in the number of SNpc
mDA neurons in FGF-2 ko animals (*** p < 0.001). (G) At stage E14.5 the localization of the future VTA
region can not be defined by Calbindin (green) and TH (red) immunohistochemistry, as double labeled
cells (yellow) are randomly distributed throughout the TH-ir profile (arrow, inset G). (H) The dotted line
surrounds the VM TH-ir cells included in the analysis, which resulted in similar numbers of TH-ir neurons
for wild-type and FGF-2 ko animals (I). Scale bars: 200 µm.
Consistent with the increased number in P28 animals, newborn FGF-2 deficient mice
displayed an increase of 34% in SNpc mDA neuron number (p < 0.001, Fig. 7 D-F). This
57
indicated a prenatal onset of these differences between the genotypes. Moreover
rostrocaudal plotting of mDA neuron number showed a uniform increase at almost every
rostrocaudal level (Fig. 7 D). Calbindin/TH double-staining at embryonic stage E14.5,
revealed that mDA neurons of the VM could not be distinguished with regard to their
prospective identity as SNpc or VTA neurons by morphometric or immunohistological
methods, as the Calbindin-positve cells were widely distributed within the TH-ir domain
(Fig. 7 G). Therefore, the total number of TH-ir neurons was quantified in the VM of
E14.5 embryos (area surrounded by dotted line, Fig. 7 H). This analysis revealed similar
numbers of VM TH-ir neurons in FGF-2 deficient and wild-type E14.5 embryos (Fig. 7 I).
Thus, the increased number of SNpc mDA neurons occurred during late stages of
embryonic development of FGF-2 deficient mice between E14.5 and P0.
3.2 Unchanged expression of mDA marker genes on mRNA level
Several intrinsic and extrinsic factors participate in mDA fate specification, differentiation
and maintenance (Abeliovich and Hammond, 2007, Gale and Li, 2008, Chung et al.,
2009). A possible alteration of the expression of some of these factors was determined
in VM preparations of FGF-2 deficient mice and compared to the relative expression in
wild type mice during the relevant developmental stages: before the onset of the
phenotype at E14.5 and thereafter at P0.
First, a screening of pooled cDNA from 5-6 animals/ genotype and stage was performed
with qRT-PCR using specific primer pairs for quantitative detection of specific mRNA
transcripts (Fig. 8 A, B). At E14.5, only expression of Aldh1a1, which participates in
retinoic acid metabolism (chapter 1.2.5), was increased over 50 % (to 70%). The other
genes, early (Foxa1/2, Lmx1a, Pax2, Ngn2, Mash1, En1/2), and postmitotic markers
(Pitx3, Nurr1, TH, DAT, Vmat2, Ddc and both dopamine receptors Drd1a/2) as well as
BDNF, remained unchanged. Any regulation of the same genes was evident at P0 (the
screening of P0 animals was performed by Kerstin Kuhlemann and Dr. Andreas
Ratzka).
58
Figure 8. Transcript levels of mDA marker genes are unchanged in VM of FGF-2 deficient animals
at E14.5 and P0. (A,B) The cDNA pools of 5-6 VMs from FGF-2 deficient were screened for expression
changes in early genes associated with mDA phenotype and postmitotic markers and compared to wild
type mice (=0). Only expression of Aldh1a1 in E14.5 VM exceeded the 50% boarder (A), while other
genes remained unchanged in E14.5 (A) and P0 (B) VM. (C): Gene expression levels in VM tissue
samples isolated from E14.5 wild-type (white bars), E14.5 FGF-2 ko (dark gray bars), P0 wild-type (light
gray bars) and FGF-2 ko (black bars) were determined by quantitative RT-PCR and are given as fold
changes in relation to E14.5 wild-type sample, which was set to 1. Expression levels of nine mDA marker
genes were unchanged between wild-type and FGF-2 deficient animals at E14.5 and P0 stage.
59
To test for statistical significance, the expression of Aldh1a1 as well as some of the
central mDA marker genes (Dat, Dlk1, Foxa2, Lmx1a, Pitx3, Nurr1, TH, and Vmat2) was
analyzed in cDNA samples acquired from single animals (Fig. 8 C, this experiments
were performed and evaluated by Kerstin Kuhlemann and Dr. Andreas Ratzka).
However, no significant differences (p > 0.05) were evident, neither in E14.5 nor in
newborn FGF-2 deficient animals compared to the wild type, including Aldh1a1
expression in E14.5 VM. The relative variation within single group was very high for 8
out of 9 genes. Since the whole VM was used for RNA extraction, SNpc (A9) specific
differences might be masked by mDA neurons located in adjacent regions of the
retrorubral field (A8) and VTA (A10) or even by other cell types expressing at least some
of the tested genes.
3.3 Unchanged expression of FGFs on mRNA level
Despite the mitogenic role of FGF-2 in vitro, the loss-of function mutant (FGF- deficient
mice) develops a significantly increased number of TH-ir cells in SNpc prenatally
(chapter 3.1). An over-compensation of FGF-2 loss by other FGF family members was
proposed previously, which might be reflected in the upregulation of their expression
(Grothe and Timmer 2007). To identify possible differences within the FGF system
between the genotypes on a molecular level we quantified comparatively their mRNA
transcripts in tissue of VM, striatum (STR) and as reference spinal cord (SC) by
quantitative RT-PCR. The respective tissues were sampled from E14.5, P0, P28 and
adult FGF-2 deficient and wild type littermates in cooperation with Dr. A. Ratzka. The
RNA extraction and conversion in cDNA was performed by Kerstin Kuhlemann. The
specific primer pairs for all 22 ligands of the FGF-Family as well as their receptors were
designed and tested by Dr. Andreas Ratzka. The qRT-PCR experiments were
performed by Kerstin Kuhlemann and Dr. Andreas Ratzka.
60
Figure 9. Transcript levels of selected FGFs and FGFRs in ventral midbrain (VM), striatum (STR)
and spinal cord (SC). All values are given as relative measure to the reference marked with bold letter despite some cases the reference was SC at P0. FGF-1 was low expressed at embryonic stages and
became upregulated postnatally (A). FGF-2 expression increases gradually during embryonic and
postnatal development (B). Highest expression of FGF-3 was detected in STR and also in embryonic VM.
FGF-8, -13, -15 and -17 showed their highest expression in embryonic tissue, including VM (D-G)
Expression of FGF-20 was detected specifically in the VM at all stages. While the expression of FGFR1c
remained more or less stable throughout the development, FGFR2c and -3c became gradually
upregulated (I-K).
As expected, the expression of FGF-2 was not detected in FGF-2 deficient animals at
any investigated developmental time point. The mRNA expression levels of all FGFs
61
and their receptors were unchanged between FGF-2 deficient mice and wild type
littermates at any investigated time point. Thus, a compensation of FGF-2 loss by other
FGFs is unlikely on mRNA level. However, the extensive quantitative analysis of FGF
expression within the developing nigrostriatal system revealed some novel insights in
the dynamic regulation of some family members (Fig. 9) as well as the relevant receptor
isoforms FGFR1c, FGFR2c and FGFR3c, which were reported to be expressed in the
VM (compare chapter 1.4). In wild type animals FGF-2 expression in VM doubled
gradually from stage E14.5 to P0 and from P0 to P28. FGF-1 and FGF-3, which share
the nuclear function with FGF-2 (chapter 1.3.2) were differently regulated. While FGF-1
becomes first substantially expressed in postnatal VM, STR and SC, FGF-3 showed
highest expression at all stages in STR but also in embryonic VM, which became
downregulated during development. Another interesting dynamic regulation was the
expression of FGF-8, -13, -15 and -17 showing the highest expression levels in
embryonic tissue, including VM, where their expression decreased with increasing age.
Also, the exclusive expression of FGF-20 only in VM tissue was verified. The expression
of FGFRs was differentially regulated during the CNS development. While the
expression of FGFR1c remained more or less stable throughout the development,
FGFR2c and -3c became gradually upregulated.
3.4 Reproduction of the phenotype of supernumerary TH-ir cells in
vitro
As the substantia nigra of adult FGF-2 deficient mice contains more mDA neurons
compared to wild-type animals (chapter 3.1), the question was raised whether such
differences could be reproduced during differentiation of mDA progenitor cells in vitro.
Therefore, E11.5 VM cells from either wild-type or FGF-2 deficient mice were cultured
for 6 days under differentiation condition in vitro. However, comparative evaluation of
TH-ir cells revealed no significant difference between the genotypes by cell ELISA
technique for neuronal marker β-Tubulin III (Fig. 10 A) and TH (Fig. 10 B, C). While the
results for the β-Tubulin III-ir cell ELISA measurements remained stable across the
62
experiments, the results for TH-ir measurement varied, probably due to subtle
differences in the age and maturation stage of mDA precursor cells of the dissected
brains. In addition, counting of TH-ir cells revealed similar numbers of mDA neurons of
FGF-2 deficient and wild-type derived cells confirming the cell ELISA data (Fig. 10 D).
18kDa
Figure 10. Overexpression of FGF-2
, FGF-8b, FGF-15 or FGF-17a does not influence
differentiation of mDA neurons in vitro. (A-D) Comparative evaluation of E11.5 derived VM cultures
from wild-type and FGF-2 deficient mice revealed similar numbers of neurons (β-tubulin III-ir, A) and mDA
neurons (TH-ir, B), quantified either by cell-ELISA (C) or immunocytochemistry (D). (E-N) The transient
18kDa
overexpression of FGF-2
(G), FGF-8b (H), FGF-15 (I) or FGF-17a (J), did not significantly increase
the yield of TH-ir cells (K-N) neither in wild-type (E) nor in FGF-2 deficient VM cells (F).
63
3.5 Overexpression of selected FGF-ligands
The investigation of expression levels of the complete FGF family in nigrostriatal system
of embryonic, newborn and adult FGF-2 deficient revealed no regulation of their
expression on the mRNA level. Nevertheless, it was observed, that FGF-8, FGF-15 and
FGF-17 showed high expression levels specifically in the embryonic VM and were
down-regulated during development (chapter 3.3). Therefore the effect of continuously
high availability of these FGFs on TH-ir cell differentiation in vitro was analyzed. FGF
expression plasmids encoding either for FGF-218kDa, FGF-8b, FGF-15 or FGF-17a were
transiently delivered in primary cultures of E11.5 VM cells, derived either from FGF-2
deficient mice or wild type mice. Cells transfected with empty-plasmid served as control
and were set to 100%. Immunocytochemical detection of transfected cells by targeting
the FLAG-epitope revealed approximate 10 - 20% FLAG-ir cells 6 days after transfection
(Fig. 10 G-J). Quantification of TH-ir cell number in FGF- and empty-plasmid transfected
controls revealed no significant differences neither in wild-type nor FGF-2 deficient VM
cell preparations (Fig. 10 E-F). In addition, semi-quantitative examination of three
independent transfection experiments, revealed no obvious differences between wildtype and or FGF-2 deficient cells (data not shown).
3.6 Loss of FGF-2 increases the number of proliferative mDA
precursors in SVZ
Studies with complex FGFR mouse mutants showed, that FGF signaling is required for
the balance between self-renewal, cell-cycle exit and neurogenesis in the ventral
midbrain (Saarimaki-Vire et al., 2007, Lahti et al., 2011). Since exogenously applied
FGF-2 is widely used as a mitogen to expand neuronal progenitor cells in vitro (compare
chapter 1.4), it was expected that the loss of endogenous FGF-2 might affect the
proliferation of VM progenitors. Therefore, cells positive for the cell cycle marker Ki67,
which labels proliferating cells in all phases of the active cell cycle (G1, S, G2 and M
phase), were evaluated in 6 FGF-2 deficient and 4 wild-type E14.5 littermates (same
animals as used for TH morphometry). Ki67-ir cells were quantified in the TH-ir profile of
64
the rostral VM which was divided in three zones: ventricular zone (VZ), subventricular
zone (SVZ) and mantle zone (MZ) (Fig. 11 A, B). This analysis revealed an increased
number of proliferating Ki67-ir cells in the SVZ of FGF-2 deficient animals (p < 0.05, Fig.
11 C).
To confirm these findings, a BrdU pulse chase analysis was performed on 10
independent E14.5 embryos (5 FGF-2-/- and 5 wild-type). Double labeling with Lmx1a
and TH antibodies was used to distinguish the VZ / SVZ containing mDA precursor cells
(Lmx1a-ir) from the MZ containing mature mDA neurons (TH-ir) in rostral (Fig. 11 D) and
caudal (Fig. 11 E) sections of the VM. On consecutive sections cells double labeled for
BrdU and Lmx1a were quantified in the VZ and SVZ (Fig. 11 G,H,J,K). The number of
proliferating Lmx1a-ir mDA-precursor cells, which had incorporated BrdU during the last
20 h, was significantly increased in the SVZ of rostral sections of FGF-2 deficient
embryos (p < 0.05; Fig. 11 F), whereas differences in the rostral VZ (Fig. 11 F) or caudal
SVZ / VZ were not significant (Fig. 11 I). These regional differences correlated with the
appearance of additional mDA neurons in postnatal FGF-2 deficient mice specifically in
the SNpc, which normally originate from those rostral VM mDA precursors.
65
Figure 11. Increased proliferation of mDA precursor cells in the rostral VM. Cell proliferation in the
VM of E14.5 embryos was analyzed on coronal sections using either Ki67 (A-C) or BrdU (D-M) staining.
(A, B) The number of cells positive for mitotic marker protein Ki67 (green signal) was evaluated within the
TH-ir profile (red signal), which was divided in three regions: VZ, SVZ and MZ. (C) Quantification of the
Ki67-ir cells, revealed a significant increase number in the SVZ of FGF-2 deficient mice. (D-E) Coronal
section levels of the rostral (D) and caudal (E) VM, showing the distribution of Lmx1a-ir mDA precursor
cells (green signal) mainly in the VZ and SVZ, whereas TH-ir mature mDA neurons (red signal) are
located in the MZ. (F-M) Proliferation of BrdU labeled (red signal, L) Lmx1a-ir mDA progenitor cells (green
signal, M) was quantified in rostral (G,H,J,K) and caudal section levels (not shown) within the VZ and
SVZ. (F,I) Quantification of BrdU/Lmx1a-ir cells revealed a significant increased number of proliferating
cells in SVZ of the rostral VM of FGF-2 deficient mice (F), whereas the rostral VZ (F) and the caudal VZ /
SVZ (I), respectively, contained similar numbers between the genotypes. Scale bars: in A,B,D,E 200 µm,
in F,G,H,I 100 µm, L,M 10 µm.
66
3.7 β-catenin and extracellular FGF signaling in E14.5 FGF-2 deficent
mice
Wnts are key regulators of proliferation and differentiation of mDA precursors during VM
neurogenesis (Castelo-Branco et al., 2003). β-catenin, as key player within the
canonical Wnt signaling cascade, can enhance NSC proliferation in the presence of
FGF-2 and induce neuronal differentiation in its absence (Israsena et al., 2004). FGF
signaling was shown to interact with Wnt signaling via downregulation of GSK-3β activity
mediated both by Akt pathway (Katoh, 2006) and ERK1/2 pathway (Frodin and
Gammeltoft, 1999, Torres et al., 1999), which prevents degradation of β-catenin level
and leads to enrichment of cytoplasmic β-catenin and its translocation to the nucleus
(chapter 1.3.2). Therefore, the β-catenin levels were investigated at the relevant
developmental stage (E14.5) in cytoplasmic and nuclear extracts from VM of FGF-2
deficient and wild type mice using western blot. The cytoplasmic extracts were also
analyzed for alterations of intracellular signaling cascades. The kinase activation was
detected by the application of anti-phosphoprotein antibodies recognizing the certain
phosphorylation sites of the respective kinases. Beta-actin, GAPDH, non phospholylated
Akt protein, or histone H3 were used as loading controls for densitometric normalization
(Fig. 12). No significant decrease in β-catenin levels was evident, neither in cytoplasmic
(Fig. 12 A,C) nor in nuclear (Fig. 12 B,C) extracts of mutant mice. Also the extracellularly
activated FGF signaling cascades were unchanged in the FGF-2 deficient. The
phosphorylation of p42/44-MAPK (ERK1/2) (Fig. 12 D,F) as well as Akt (Fig. 12 E,G)
was not significantly changed in mutants, suggesting that different, maybe intrinsic,
regulatory mechanisms are responsible for enhanced differentiation of TH-ir cells in
FGF-2 deficient mutants.
67
Figure 12. Increased levels of nuclear FGFR1 in VM of FGF-2 deficient mice. (A-C) Western blot
analysis of β-catenin levels in cytoplasmic (A) and nuclear extracts (B) of E14.5 VM revealed no
alterations in FGF-2 deficient mice (C). (D-G) Levels of phosphorylated p42/44 MAPK (D,F) and Akt (E,G)
were unaltered in E14.5 VM of FGF-2 deficient mice. (H,I,L,M) FGFR1-ir cells (green signal) were
detected throughout the VM of wild-type (H,L,O,P) and FGF-2 deficient mice (I,M,Q,R), including VZ and
SVZ (arrows, H,I), and TH-ir region (red signal, L,M). (O-R) High power view (square in L,M) display
nuclear (blue DAPI signal) localization of FGFR1-ir (green signal) in TH-ir cells (red signal) (confocal
microscop Leica DM IRB). (J,K) High magnification confocal images (Zeiss, Axiovert) of DAPI (blue signal,
K) stained E14.5 VM cells display cytoplasmic (arrowheads) and nuclear localization (arrows) of FGFR1
(red signal, J,K). (N) Western blot of E14.5 VM tissue detects FGFR1 in cytoplasmic and nuclear extracts.
Purity of extracts was verified by exclusive cytoplasmic Gapdh and nuclear histone H3 distribution. (S)
Densitometry of nuclear FGFR1 bands revealed two fold accumulation of nuclear FGFR1 in FGF-2
deficient mice. (T) Exogenous application of FGF-2 for 24 hours reduced nuclear FGFR1 levels in E14.5
VM explant cultures of FGF-2 deficient mice compared to untreated controls. Scale bar: in H,I,L,M 100
µm, in O-R 5 µm.
68
3.8 Nuclear accumulation of FGFR1 in FGF-2 deficient mice
Integrative Nuclear FGFR1 Signaling (INFS) is a common mechanism which can be
activated by nuclear HMW isoforms of FGF-2 as well as diverse FGF-unrelated factors
which promote neuronal development (Stachowiak et al., 2007, Stachowiak et al., 2011).
INFS includes the translocation of FGFR1 into the nucleus, its direct interaction with
CBP leads to activation of CREB and induction of gene expression, like TH (Peng et al.,
2002, Stachowiak et al., 2007). To determine whether the INFS mechanism may be
influenced during differentiation phase in FGF-2 deficient mice, the FGFR1 distribution
was analyzed in E14.5 VM. Immunohistochemcal analysis of FGFR1 and TH in E14.5
cryosections showed, that FGFR1 was present in wild type and in knock-out mice
throughout the whole VM (Fig. 12 H,I,L and M), showing nuclear localization also in THir cells of both genotypes (Fig. 12 O-R). By confocal microscopy FGFR1-ir was detected
in both, cytoplasm (arrowheads, Fig. 12 J,K) and nucleus (arrows, Fig. 12 J,K), of DAPI
stained VM cells. The expression pattern was more or less similar, but maybe the
presence of FGFR was less pronounced in the VZ and SVZ of wild type animals (arrow
Fig. 12 H,I). Therefore, a subsequent western blot analysis was performed to clarify,
whether the subcellular distribution of FGFR1 was altered in FGF-2 deficient E14.5 VM.
Interestingly, we found an increased accumulation of FGFR1 in the nucleus of FGF-2
deficient mice, while in the cytoplasm the FGFR1 concentration remained unchanged
(Fig. 12 N.S). The purity of the fractions was verified by absence of GAPDH in nuclear
extracts and absence of histone H3 in cytoplasmic extracts (Fig. 12 N). The in vitro
cultivation of E14.5 VM from FGF-2 deficient mice in presence of recombinant FGF-2
(20 ng/ml) for 24 hours in vitro could reverse the nuclear accumulation of FGFR if
compared to untreated tissue cultures (Fig. 12 T). The enhanced accumulation of FGFR
in the nucleus and the INFS mediated gene regulation might be responsible for the
enhanced differentiation of TH-ir cells.
69
3.9 Nuclear FGFR1 interacts with Nurr1
3.9.1 FGFR1 and Nurr1 are co-expressed in the nuclei of mDA neurons
As mentioned in chapter 1.2.7 of the introduction, Nurr1 is a key transcription factor in
integration of terminal differentiation of mDA neurons. A similar function was described
for nuclear FGFR1 as novel FGF signaling pathway (INFS) being involved in
differentiation processes during development (chapter 1.3.3). Both nuclear components
were shown to induce TH gene expression. Therefore a possible crosstalk of Nurr1 and
nuclear FGFR1 signaling was investigated starting with the expression analysis of both
proteins in the ventral brain of wild type and FGF-2 knock animals.
Nurr1 was highly expressed in the ventral midbrain of mouse embryos at E14.5. In this
region the localization of Nurr1 was restricted to mDA postmitotic precursors and
maturating mDA neurons. At E14.5 the majority of Nurr1-ir cells present in the mantel
zone started to express TH (Figure 13 A,E,I,B,F and J). Since Nurr1 expression is
initiated in the postmitotic mDA precursors, Nurr1-ir cells were first found in the subventricular zone (Fig 13 E, arrow), and were absent in the ventricular zone (Fig 13 E,
arrowhead), the origin of proliferating, self-renewing neural stem cells. The distribution of
Nurr1-ir cells within the ventral midbrain of FGF-2 deficient seemed unaltered if
compared to wild type mice.
In chapter 3.8 the expression of FGFR1 was analyzed with polyclonal FGFR1 antibody
in cryosections showing a broad expression of FGFR1 throughout the embryonic
midbrain (Fig. 12, H, I) and also nuclear localization in TH-ir cells (Fig. 12, O-R). For the
immunohistochemical double staining of Nurr1 and FGFR1 a different anti-FGFR1
antibody (monoclonal, abcam M19B2) was utilized, which showed a slightly different
expression pattern (Fig. 13 C,D), which might be due to recognition of different FGFR1
isoforms. However, the expression patterns of both nuclear receptors Nurr1 and FGFR1
overlapped especially in the ventral mesencephalic mDA domain of wild type (Fig. 13, C,
G, K) and FGF-2 ko (Fig. 13 D, H, L) embryos. In the confocal analysis at higher
70
magnification a spatial co-localization of Nurr1 protein and FGFR1 was observed in the
cell nuclei of immunohistochemically processed slices from E14.5 mouse brains (Fig. 13,
M-P, arrows), representatively shown in wild type tissue. Spatial co-localization of
nuclear FGFR1 and Nurr1 indicated that both proteins might be localized in the same
nuclear complexes.
Figure 13. Nurr1 and FGFR1 co-exist in the cell nuclei of the ventral midbrain. Nurr 1 is localized in
the nuclei of the TH-ir cells of the wild type (A, E, I) and FGF-2 deficient (B,F,J) embryos. Expression
pattern of FGFR1 is distributed throughout the VM of wild type (C) and FGF-2 ko animals (D) overlapping
with Nurr1 expression pattern in the ventral aspects of mDA domain (G, H) showing also nuclear
localization (K,L). The confocal high magnification images show a clear presence of FGFR1 (M) and Nurr1
(N) in the nucleus of VM cells showing a granular distribution (P). The overlapping of the FGFR1 (red) and
Nurr1 (green) channels results in co-localization of both proteins resulting in yellow signal (O, arrows).
71
3.9.2 FGFR1 and Nurr1 co-exist in the same nuclear complexes
For studies of a possible interaction of Nurr1 and FGFR1 in the cell nucleus of
dopaminergic progenitors a SV-40 VM-NPC neuronal progenitor cell line was used. The
SV40i-VM-NPCs were generated in our lab by introduction of the Simian Virus 40
(SV40) to the primary rat embryonic neuronal progenitor cells, which expressed mRNAs
of genes associated with mDA development (Nobre et al., 2010).
First, the expression of Nurr1 and FGFR1 protein was verified in the nucleus of SV40VM-NPCs immunocytochemically (Fig. 14 A-D) and with western blot (Fig. 14 E). The
double labeling of FGFR1 and Nurr1 immunocytochemically was analyzed with confocal
laser-scanning microscope. FGFR1 as well as Nurr1 showed similar to the in vivo
situation (chapter 3.9.1) a granular distribution within the nucleus and additionally in the
cytoplasm. The overlay of single confocal planes with Nurr1 and FGFR1 signal resulted
in yellow assigned co-localization signal (Fig. 14 C, arrows). Interestingly, the colocalization occurred not only in the nucleus of SV40-VM-NPCs, but also in the
cytoplasm (Fig. 14 D, arrowhead). The nuclear presence of the Nurr1 and FGFR1
protein was proven by western blot using the same antibodies as for the
immunoprecipitation (Fig. 14 E).
Co-immunoprecipitation technique was applied in nuclear lysates of SV40i-VM-NPCs.
The endogenous FGFR1 protein was precipitated with a polyclonal anti-FGFR1
antibody. As negative control for co-immunoprecipitation the lysates were incubated with
unspecific rabbit IgGs. The protein-IgG complexes were pulled down with the Protein Acoupled magnetic beads (Dynabeads). The precipitated protein-complexes were
denaturized, resolved with SDS-PAGE and analyzed in western blot for presence of
Nurr1. Immunocytochemistry and immunoprecipitation could show that endogenous
Nurr1 and FGFR1 co-exist in the same nuclear complexes in the mDA progenitors (Fig.
14 F).
72
The precipitation of endogenous FGFR1 with polyclonal anti-FGFR1 antibody from
nuclear lysates of E14.5 ventral mesencephalon from FGF-2 deficient embryos resulted
twice in co-precipitation of Nurr1 showing a weak band in western blot, while in lysates
of wt tissue only unspecific reaction was evident. However, this finding was hardly
reproducible in subsequent experiments, especially in the lysates from the wild type
embryos, maybe due to a low availability of the tissue resulting in low nuclear protein
amounts (maximum 200 µg/precipitation) (Data is therefore not shown, but mentioned
as a failed experiment).
Figure 14. Presence of nuclear FGFR1 and Nurr1 in the same nuclear protein complexes of mDA
progenitors. SV40-VM-NPCs were seeded on polyornithin and laminin coated coverslips (80,000
cells/well in 24 well plate), cultivated for 24 hours in serum-free N2-medium, fixed in 4% PFA in PBS,
processed immunocytochemically and analyzed with confocal laser-scanning microscope (Leica TCS
SP2). Confocal images of FGFR1 (A, C, D) and Nurr1 (B, C, D) showed also granular distribution of the
proteins in the nucleus of mDA progenitors, but also in the cytoplasm (D). The overlay of FGFR1 and
Nurr1 resulted in yellow signal, where Nurr1 and FGFR1 are co-localized in the nucleus (C, arrows) and
also in the cytoplasm (D, arrowhead). (E) Western blot detection of Nurr1 and FGFR1 in nuclear extracts
of SV-40 VM-NPCs with antibodies used for immunoprecipitation. (F) For co-immunoprecipitation the
SV40-VM-NPCs were seeded in coated 75 cm² flasks (2 million cells/flask). The nuclear extracts were
diluted to 1 µm/µl protein. 400 µl of protein lysates were incubated with 2 µg of antibody, precipitated with
Protein A Dynabeads, denaturized in Laemmli buffer, processed with SDS-PAGE and western blot. The
immunoprecipitation with IgGs represents the negative control for precipitation. The input represents the
loading control of 100 µg pure denaturized nuclear protein extract. The precipitation of FGFR1 resulted in
co-precipitation of Nurr1.
73
To prove the interaction of Nurr1 and FGFR1 in principal, the human neuroblastoma cell
line was used to over-express FGFR1 and FLAG-tagged Nurr1. Immunoprecipitation of
Nurr1, FGFR1, and FLAG-tagged protein, respectively, was performed in nuclear
extracts. Indeed, Nurr1 was found to co-precipitate with FGFR1, as shown in FGFR1-IP
after immunodetection of the FLAG-epitope fused to Nurr1-protein (Fig. 15, 1st row) or
with the antibody raised against Nurr1 (Fig. 15, 2nd row). Vice versa, the precipitation of
Nurr1 resulted in co-precipitaiton of FGFR1 as detected with a monoclonal FGFR-1
antibody (Fig. 15, 3rd row). Interestingly, the precipitation of FGFR1 showed an
enrichment of an additional band in Nurr1 immunoblot, which might be due to coprecipitation of sumoylated Nurr1 form and/or endogenous Nur77. Nurr77 is a closely
related family member of orphan nuclear family and is also recognized by this antibody.
However, Nur77 expression is absent in VM (Xiao et al., 1996, Saucedo-Cardenas and
Conneely, 1996).
Figure 15. Nuclear FGFR1 and Nurr1 interaction after overexpression in neuroblastoma cells. The
huma neuroblastoma cells were transfected with plasmids encoding for full length FGFR1 protein as well
as Nurr1-protein fused to a 3xFLAG-tag using Metafectene Pro. After 24 h the transfected cells were
supplemented with 1 µM retinoic acid and cultured for further 24 h in vitro. The nuclear extracts were
precipitated with polyclonal anti-Nurr1, anti-FGFR1 and rabbit IgGs as negative control. The resulting
protein-IgG complexes were denaturized in Laemmli buffer and separated with SDS-PAGE. The detection
of co-precipitated proteins was performed with western blot. Input represents 50 µg protein of the not
precipitated nuclear extract. Precipitation with Nurr1 antibody functioned properly as shown by
st
nd
precipitation of Nurr1 with the anti-FLAG antibody (1 row) and anti-Nurr1 antibody (2 row). The second
band in the input as well as in Nurr1-IP may represent the sumoylated-form of Nurr1 or the closely related
74
endogenous Nur77 protein. Additionally, a precipitation of 70 kDa, 80 kDa as well as glucosylated 130
rd
kDa isoforms of FGFR1 were able to co-precipitate with Nurr1 (3 row). Correspondingly, the FGFR1-IP
st
resulted in co-precipitation of Nurr1, recognized by anti-FLAG-tag antibody (1 row) as well as with antiNurr1/Nurr77 antibody. The negative controls, precipitated with rabbit IgGs, were missing the specific
band representing the corresponding protein, confirming a specificity of the Nurr1- and FGFR1-IPs.
3.10 Analysis of postnatal natural cell death
During the maturation process of mDA neurons, including nigrostriatal wiring, the major
peak of ontogenic cell death in the ventral midbrain occurs perinatally (Burke, 2004).
Therefore the apoptosis within SNpc was analyzed in newborn FGF-2 deficient mice. As
the expression of phenotypic markers, like TH, declined in mDA neurons undergoing
apoptosis, a large number of those cells would not be identifiable by TH
immunohistochemistry (Burke, 2004). To circumvent this, the TH-ir profile was
delineated by a dotted line and all apoptotic cells were counted within this profile (Fig. 14
A), as suggested previously (Burke, 2004).
Nevertheless, some of the apoptotic cells, which retained the TH-immunoreactivity,
showed already cleaved caspase-3 (cC3) staining. These cells represent presumably
early apoptotic stages (Fig. 16 C,D, arrow). Three apoptotic events were counted
separately: cC3-positive cells missing apoptotic bodies (ApB) (Fig. 16 D, arrow), cells
containing ApB but cC3-negative (Fig. 16 E, arrowhead), and cC3-positive cells showing
ApB (Fig. 16 F, arrow). Although the total number of apoptotic cells was unchanged
between FGF-2 deficient and wild type animals, the number of cells displaying ApB
without cC3 staining (ApB only) was reduced by 39% in FGF-2 deficient animals (p <
0.05 U-test, Fig. 16 B). No differences were observed for cC3-positive cells, neither for
ApB/cC3+ nor cC3+ only cell groups, indicating either a later onset of natural cell death
or reduction of caspase-3 independent cell death under FGF-2 deficiency.
75
Figure 16. Reduced number of late apoptotic cells in P0 FGF-2 deficient mice. (A) At stage P0 all
apoptotic cells inside the TH-ir profile (red signal) of the SNpc (surrounded by the dotted line) were
counted. Apoptotic cells were identified either by cleaved caspase 3 (cC3+, green signal) staining (arrow
in D) or by apoptotic bodies visualized by DAPI (blue signal) staining (arrow and arrowhead in E). (B-F)
Three apoptotic events were counted separately, apoptotic bodies without cC3+ staining (ApB only,
arrowhead in E), apoptotic bodies and cC3+ (ApB/cC3+, arrow in F), cC3+ but without apoptotic bodies
(cC3+ only, arrow in D). Cells containing ApB were reduced to 61% in the SNpc TH-ir profile of FGF-2
deficient mice (* p < 0.05, U-test), whereas both cleaved caspase-3 positive cell groups (ApB/cC3+ and
cC3+ only) remained unchanged (B). Images C-F were acquired using confocal Leica DM IRB
microscope. Scale bars: in A 500 µm, in C-F 10 µm.
3.11 FGF-2 in axonal outgrowth of mDA neurons
The apoptotic events in postnatal SNpc occur due to removal of inadequately wired
mDA neurons with the forebrain targets (Burke, 2004, Prakash and Wurst, 2006). Since
the apoptosis in newborn FGF-2 deficient mice was found diminished compared to wild
type mice (chapter 3.10), next question was whether FGF-2 influences the axonal
outgrowth mDA neurons and/or target innervation. Therefore, an explant co-culture
model was established. The embryonic brains were harvested at stage E14.5 from wild
type and FGF-2 knockout animals. Forebrain (FB) and ventral mesencephalic (VM)
76
explants were dissected as described in chapter 2.2.4 and shown in Fig. 15 A. Different
combinations of wild types (wt) and FGF-2 deficient (ko) explants (wt VM/ wt FB; wt VM/
ko FB; ko VM/ wt FB and ko VM/ ko FB) were co-cultured in cell culture inserts
according to model described previously (Stoppini et al., 1991). To distinguish the VM
and FB explants transgenic mouse strains constitutively expressing EGFP were used for
the VM explants. Thereby, the TgEGFP;FGF-2+/+ VM explants represented wild types
for FGF-2 and the TgEGFP;FGF-2-/- double mutants represented FGF-2 deficient VM
explants (Fig. 17, C). Although, the litters were sampled from homozygous breadings, to
prevent the influences due to variability of different litters at least one explant genotype
was used from the same mother for a single experiment. For example VM explants of
certain genotype were used from the same litter for all co-cultures with FB explants from
two different litters representing two different genotypes (Fig. 17, B). The co-cultures
were allowed to establish fibers from VM to FB explant for 5 DIV in OTC-culture medium
free from serum, FGF-2 or other trophic supplements, like B27-supplement. The
organotypic cultures were fixed, processed immunohistochemically for TH and
evaluated at 4x objective magnification with epifluorescence microscope. The length and
distance of the fiber outgrowth, as well as width of the TH-ir tract were measured using
the CellP software (Olympus) (Fig. 17, E). The length was defined by regularly spaced
wavy lines, which were laid according to the progression of the TH-ir fibers from the
boarder of the VM explant through the FB explant, until the terminations of the longest
fibers were reached in this regions. The distance simply represents the linear distance of
the tracked fibers from the boarder of the VM explant to the termination of the TH-ir
fibers in FB explant. The width of the TH-ir tract was measured in the FB entering
segment, where the fibers are still bundled and before they start to spread. A quotient of
the length and distance was calculated to estimate the quality of axonal outgrowth,
which was unchanged between all conditions (0.9 ± 0.1) assigning a directed outgrowth.
The measurements of the fiber outgrowth length revealed that co-cultures which were
missing FGF-2 in the VM and FB (ko VM/ ko FB) resulted in significantly longer TH-ir
77
fibers if compared to the other co-cultures containing VM and/or FB explant from wild
type mice, respectively: 21% to wt VM/ wt FB (p < 0.05); 29% to wt VM/ ko FB (p < 0.01)
and ko VM/ wt FB (p < 0.001) (Fig. 17, F). Similarly, the distance of TH-ir fibers was
significantly longer in ko VM/ ko FB co-cultures compared to co-cultures where either
VM or FB are from wild type embryos (p < 0.01 and 0.001). Interestingly, concerning the
distance, the difference between the co-cultures of wt VM/ wt FB and ko VM/ ko FB was
not significant. The width of the TH-ir tract was significantly increased in the co-cultures
which contained one explant from ko embryo compared to wt VM/ wt FB co-cultures: wt
VM/ ko FB 26% (p < 0.05) and ko VM/ wt FB 22% (p < 0.05) wider than wt VM/ wt FB
co-cultures. The 15% increased width of ko VM/ ko FB co-cultures was not significantly
different to wt VM/ wt FB co-cultures.
The longer fiber outgrowth in co-cultures completely missing FGF-2 compared to wild
type co-cultures indicates that FGF-2 is somehow involved in mDA fiber outgrowth and
target innervation. The fact that FGF-2 deficient cultures have also longer fibers than the
co-cultures which contain either the VM or FB explant from wild type mice, suggests that
hereby a complex interplay of striatal and mesencephalic FGF-2 is required for
adequate regulation of multiple processes like path finding and target innervation. The
thinnest width of the TH-ir fiber tract in wt VM/ wt FB cultures underlines the role of FGF2 in path finding.
Figure 17. FGF-2 participates in mDA axonal outgrowth. (A, C): Forebrain (FB) and ventral
mesencephalic (VM) explants were dissected from E14.5 embryonic brains). To distinguish the VM and
FB explants double transgenic mouse strains constitutively expressing EGFP were used for the VM
explants. (B, D): Different combinations of wild types (wt) and FGF-2 deficient (ko) explants (wt VM/ wt FB;
wt VM/ ko FB; ko VM/ wt FB and ko VM/ ko FB) were co-cultured for 5 DIV in cell culture inserts. To
prevent the variability influenced by different litters at least one explant genotype was used from the same
mother for a single experiment.
(E): The organotypic cultures were fixed, processed
immunohistochemically for TH and evaluated at 4x objective magnification with epifluorescence
microscope. The length and distance of the fiber outgrowth, as well as width of the TH-ir tract were
measured. (F): Co-cultures which were lacking FGF-2 in the VM and FB (ko VM/ ko FB) result in
significantly longer TH-ir fibers if compared to all other co-cultures, which is also reflected in distance
measurements if compared to heterogeneous co-cultures. The thinnest width of the TH-ir fiber tract was
found in wt VM/ wt FB cultures. (* - p < 0.05; ** - p < 0.01; ***- p < 0.001)
78
79
4. Discussion
In the mature mouse brain, FGF-2 is involved in regulation of dopamine turnover (Forget
et al., 2006) and survival of mDA neurons (Grothe and Timmer, 2007, Timmer et al.,
2007). Previously, our group studied the physiological function of the endogenous FGF2 system by evaluation of the adult nigrostriatal system of FGF-2 deficient and
overexpressing mice. The loss of endogenous FGF-2 revealed an increased volume of
SNpc and higher number of tyrosine hydroxylase immunoreactive dopaminergic
neurons, whereas the overexpression of FGF-2 showed an opposite effect (Timmer et
al., 2007). The hypothesis was raised that FGF-2 should participate in development of
the nigrostriatal system, ensuring establishment of an adequate number of mDA
neurons within the substantia nigra pars compacta (Grothe and Timmer, 2007). Since
FGF-2 is widely used as a mitogen to expand mDA neurons in vitro (Studer et al., 1998,
Timmer et al., 2006, Jensen et al., 2008, Pruszak et al., 2009), an opposed outcome of
phenotypes was expected in the loss-of-function as well as gain-of-function mutants.
Therefore a model of over-compensation of FGF-2 loss by up-regulation of other FGFs
has been proposed.
The present study concentrated on the confirmation that the increased number of mDA
neurons in FGF-2 deficient mice arises during SNpc development. Further, the alteration
of physiological processes and molecular mechanisms causative for the phenotype in
FGF-2 deficient mice were also a point of interest. Identification of putative
compensatory mechanisms in the present loss-of function animal model was supposed
to provide insights in the regulation of the complex signaling, which allows development
of functional mDA neurons. Better understanding of mDA neuron development
contributes to the improvement of therapeutic approaches in PD, like intrastriatal cell
replacement strategies or induction of endogenous restorative mechanisms.
80
4.1 FGF-2 is required during development of SNpc
Development of SNpc within the VM starts at embryonic day 7 with regionalization of the
neural tube and terminates with fully mature mDA neurons around postnatal day 28
(Prakash and Wurst, 2006). In fact the stereological analysis of P28 old FGF-2 deficient
mice confirmed the onset of the phenotype during development of SNpc, as an increase
of TH-ir neurons was already evident in juvenile mice.
Moreover, in juvenile mice, only SNpc (A9) mDA neuron number was increased while
the VTA (A10) remained unaffected, suggesting a specific involvement of FGF-2 in
development of A9 mDA neurons. Similar reports concerning the A9 and A10 subtype
specific regulation exist for other neurotrophic factors, which participate in mDA
development. Conditional BDNF deficient mutants showed a specific reduction of SNpc
mDA neuron subtypes, which were characterized by the absence of calcium-binding
proteins (Baquet et al., 2005). In this context, the role of differential expression of
calcium-binding proteins was discussed previously, which are predominantly present in
A10 mDA neurons. Certain other molecules, like GIRK2 in SNpc, were identified, which
altogether may play key roles in the relative vulnerability of A9 and A10 mDA neurons to
toxins and in PD (Kim et al., 2000, German et al., 1992, Chung et al., 2005).
Interestingly, although the loss of TH-ir in GDNF deficient mice is detected in both
regions, in SNpc and VTA (Pascual et al., 2008), in vitro GDNF affects the A9 and A10
mDA neurons subtypes differently depending on time and degree of exposure (Borgal et
al., 2007). While a single exposure enhanced A9 mDA neuron number, repeated
exposure to GDNF doubled the number of A10 neurons.
Further, the 36% more TH-ir cells in SNpc of juvenile FGF-2 deficient mice (present
study) in contrast to 18% more neurons in adult mice (Timmer et al., 2007), could be
explained by several reasons. One possibility would be a result of different strain
backgrounds in both studies (C57BL6 in present study versus 129P2/OlaHsd:Black
Swiss), since strain dependent variations of mDA neuron number were described
81
previously (Ross et al., 1976, Timmer et al., 2007, Vadasz et al., 2007). However, as
both strains displayed same significant mDA neuron phenotype, influence of the genetic
background can be excluded. The other possibility for the decline of the increased mDA
number from juvenile to the adult FGF-2 deficient SNpc might be a result of higher
vulnerability due to FGF-2 absence, which becomes more prominent with age. In fact,
there are some striking arguments why FGF-2 absence could lead to higher vulnerability
of mDA neurons: 1.) FGF-2 is a very important neurotrophic factor, as the deficient mice
show a significantly decreased survival of mDA neurons after neurotoxin application
(Timmer et al., 2007); 2.) one of the numerous genetic single nucleotide polymorphism
(SNP) screen studies showed a correlation between PD and FGF-2 gene mutation
(Mizuta et al., 2008), which moreover underlines the connection between FGF-2
availability and prevalence for PD as described in chapter 1.4.
Further analysis of younger stages delineated the onset of the phenotype between P0
and E14.5, as the number of mDA neurons was unchanged in E14.5 but already 35%
higher in newborn FGF-2 deficient mice. This finding assigns the relevance of FGF-2 in
VM during late embryogenesis. However, the reproduction of the phenotype in vitro
failed, as the dissociated cultures of VM progenitors from E11.5 wild type and FGF-2
deficient mice showed no significant difference in TH-ir cell number as determined by
immunocytochemistry and Cell-ELISA assay. The reason for this failure might be the
loss of organotypic organization resulting in insufficient environmental input or a
masking effect of FGF-2 supplemented medium for the first day of attachment, since the
dissociated cultures fail to survive if FGF-2 support is missing (personal observation, see
also (Fawcett et al., 1995)).
4.2 mDA marker genes are unchanged in FGF-2 deficient mice on
mRNA level
The complex signaling network during mDA specification and differentiation is
characterized by an intricate crosstalk between secreted factors, like FGF-8, Wnts and
Shh, and transcriptional activation by several known transcription factors, like Pax2/5,
82
Foxa1/2, En1/2, Lmx1a/b, Ngn2, Msx1, Aldh1a1, Pitx3, Nurr1 etc (Abeliovich and
Hammond, 2007, Gale and Li, 2008). To determine whether compensatory processes
resulted in upregulation of the expression of one of those genes in FGF-2 deficient mice,
the qRT-PCR analysis of the mRNA levels was performed. No differences of the early
genes associated with mDA development were detected in VM of E14.5 FGF-2 deficient
embryos. The unchanged expression of the early mDA markers suggests a normal
patterning of mDA field in FGF-2 deficient mice, which was confirmed by normal
expression pattern of Foxa1, Lmx1a and Nurr1 in ISH analysis, which was performed by
Dr. Andreas Ratzka and Kerstin Kuhlemann in our lab (Ratzka A.*, 2011 submitted).
This finding together with the late onset of the phenotype after E14.5 (chapter 4.1)
indicates that FGF-2 is dispensable for the regionalization and specification of the mDA
field.
Also the mRNA level of the adult markers, like TH, DAT, Vmat2, Ddc and the dopamine
receptors (Drd1a and Drd2), remained unchanged in the embryonic, postnatal and adult
FGF-2 deficient mice. This is consistent with previous findings, which showed no
increase in TH protein or intrastriatal dopamine levels in the adult FGF-2 deficient mice
(Zechel et al., 2006, Timmer et al., 2007). A possible explanation for this might be the
involvement of FGF-2 is in regulation dopamine turnover (Forget et al., 2006), thus,
resulting in reduced relative amount of enzymes required for mDA synthesis per mDA
neuron, which is reflected in unchanged levels of adult mDA markers although the
number of mDA neurons is increased.
4.3 FGF-2 does not affect the FGF system on RNA level
The discrepancy between mitogenic function of FGF-2 on mDA neurons in vitro and the
opposed phenotypes of more TH-ir cells in FGF-2 deficient mice and less TH-ir cells in
the overexpressing mice, respectively, raised the hypothesis that FGF-2 loss might be
(over)-compensated by upregulation of other FGF family members (Grothe and Timmer,
2007, Timmer et al., 2007). The milder phenotypes of single FGF gene deletions in mice
83
were discussed to result due to redundancy among the FGF family members (Itoh,
2007). Similar redundancy effects seem to occur in mutants of the growth/differentiation
factor (GDF) system, where for example an upregulation of Gdf5 mRNA was observed
in GDF-7 deficient tail tendons (Mikic et al., 2008). However, our present comprehensive
qRT-PCR analysis of the FGF-family of ligands and their high affinity receptors revealed
no significant changes in mRNA expression neither in the nigrostriatal system nor in the
spinal cord (as reference tissue) of the FGF-2 deficient mice at any developmental
stage. Anyway, an alternative physiological regulation on other level, like regulation of
translation, was not completely excluded. However, it was interesting to find that FGF-8,
FGF-15 and FGF-17 were expressed in E14.5 VM and down-regulated during
subsequent developmental stages as determined by qRT-PCR. In addition, in wild type
VM we detected also a reciprocal behavior in FGF-8 and FGF-2 expression within the
critical window of the phenotype onset: strong reduction FGF-8 expression and
contrasting increase of FGF-2 expression. Similar behavior was observed for FGF-3 and
FGF-17, while FGF-15 and FGF-13 became downregulated after P0. This switch in
FGF-expression should be considered as putative compensatory mechanism during the
differentiation phase of mDA neurons, as it will be discussed below (chapter 4.4). FGF8a and FGF17b were shown to promote expansion of mDA progenitors in explant
cultures in vitro. However, in vivo the role in patterning, specification and progenitor
maintenance was shown for FGF-8b and also FGF-17b (Guo et al., 2010). The role of
FGF-17a and FGF-15 was not investigated in VM so far. As far is known, FGF-15
promotes neural differentiation in the dorsal mesencephalon and is regulated by SHH
and FGF-8 (Gimeno and Martinez, 2007, Fischer et al., 2011). FGF-13 is proposed to
be the ancestor of the subfamily of the intracellular FGFs (Itoh and Ornitz, 2008), and is
reported to be involved in neural development and differentiation (Nishimoto and
Nishida, 2007). The regulation of FGF-3 on the other hand was less prominent, than by
other here mentioned FGFs, although it’s nuclear localization (Antoine et al., 1997),
which it shares with FGF-1 and FGF-2 (Itoh and Ornitz, 2011), might be very important
(compare chapter 4.6).
84
Therefore, the most promising candidates of secreted FGFs (FGF-2, FGF-8b, FGF-15,
and FGF-17a) were selected to study additive effects in their capability to increase mDA
neuron number in vitro. The wild type and FGF-2 deficient mDA progenitors from E11.5
mice were dissected and transfected to over-express those FGFs. However,
overexpression of FGF-8b, FGF-15, FGF-17a or FGF-218kDa had no influence on wild
type or FGF-2 deficient VM cell cultures with regard to the number of mDA neurons. This
implies that the compensatory mechanisms are more complex or other factors are
responsible for the increased mDA neuron number in FGF-2 deficient mice.
4.4 FGF-2 modulates proliferation of nigral mDA progenitors in vivo
The neuronal progenitors of the pseudostratified layer of the ventricular zone posses
self-renewing capacity, which is characterized by symmetric division, whereby both
daughter cells maintain their self-renewing capacity, apico-basal polarity, mitotic spindle
orientation as well as the apical and basal processes. The asymmetric divisions are
associated with production of one self-renewing progenitor and one neuronal postmitotic
precursor, which in turn looses the apico-basal polarity, leaves the ventricular zone to
differentiate and form the marginal zone. The symmetric neurogenic divisions result in
production of two neuronal precursors, which leave the ventricular zone (Kosodo et al.,
2004, Doe, 2008, Lahti et al., 2011).
Previous in vitro studies convince that FGF-2 has a mitogenic function in neuronal
progenitors (Dono et al., 1998, Vaccarino et al., 1999, Bouvier and Mytilineou, 1995,
Studer et al., 1998). Within the mDA system, FGF-2 induces expansion and self-renewal
of mDA progenitors and inhibits their progression into differentiated mDA neurons
(Bouvier and Mytilineou, 1995). In fact, studies with complex FGFR mouse mutants with
midbrain-hindbrain specific disruption of three FGFRs (FGFR1, FGFR2, FGFR3)
showed, that FGF signaling is required for the balance between self-renewal, cell-cycle
exit and neurogenesis in the ventral midbrain (Saarimaki-Vire et al., 2007, Partanen,
2007, Lahti et al., 2011). The authors showed a shift in the cell cycle exit in FGFR-ko
85
mutants, which caused a gradual depletion of neuronal progenitors and resulted in
increased number of BrdU positive cells in the SVZ. The shifted balance of self-renewal
of progenitors and neurogenesis was proven by observed downregulation of Hes1
expression - antagonist of proneural signaling (compare also chapter 4.4) - and upregulation of proneural genes (Saarimaki-Vire et al., 2007, Lahti et al., 2011). Moreover
an in vitro experiment illustrates a switch of balanced symmetric proliferative,
asymmetric neurogenic and symmetric neurogenic division in the presence of FGF-2 to
mainly neurogenic divisions in the absence of FGF-2 (Lahti et al., 2011). In agreement
with this, E14.5 FGF-2 deficient mice displayed increased numbers of proliferating mDA
precursor cells (BrdU/Lmx1a-ir) in the SVZ of the rostral VM, without a substantial
depletion of self-renewing progenitors in the VZ. This suggests that endogenous FGF-2
is indispensable to prevent excessive transition of self-renewing symmetrically dividing
neuronal progenitors to mDA precursors in SVZ. This is also reflected in increased
production of mDA neurons in FGF-2 deficient mice. A similar anti-neurogenic potential
of FGF-2 was reported by Chen et al. (2007) for neuronal progenitor cells form rat
hippocampus, where FGF-2 increased the number of nestin-positive neuronal
precursors and significantly decreased their differentiation in β-tubulin III-ir or MAP2-ir
neurons in-vitro. This effect was counteracted by CTNF, GDNF, and by IGF-1 as well as
IGF 2 in a dose-dependent manner (Chen et al., 2007).
The enhanced proliferation of mDA precursors at E14.5 is coherent with increased mDA
neuron number generation in FGF-2 deficient SNpc and suggests rather a
compensatory mechanism for a loss of this sufficient neurotrophic factor in mature mDA
system than contradicts its mitogenic activity in vitro. Similar to FGF-2, FGF-8 has been
successfully applied in vitro as a mitogen to expand mDA progenitors. Additionally, the
prolonged supplementation with FGFs had a negative impact on mDA neuron number
(Jensen et al., 2008). It is likely that FGF-8 (and other FGFs, compare chapter 4.3)
alternates with FGF-2 in balance control of the self-renewing capacity of the neuronal
progenitors in the VZ. Due to failure of the switch between FGF-8 and -2, the repression
86
mDA differentiation might fail in FGF-2 deficient animals, since FGF-8 expression
declines normally, but the up-regulation of FGF-2 expression is missing.
4.5 Unchanged activation of cytosolic signal transduction pathways
The two main cytosolic transduction pathways subsequent to FGFR activation are the
ERK1/2 and Akt pathways, which are both FRS mediated. The maintenance of selfrenewal and proliferation of neuronal stem and progenitor cells in developing cortex is
regulated by FGF signaling (Kang et al., 2009), which is acting via FRSα maintaining
expression of Hes1 (Sato et al., 2010). Although the VM conditional FGFR1;FGFR2
knock-out mice show downregulation of Hes1, the ERK1/2 phosphorylation remained
unchanged (Lahti et al., 2011), suggesting other factors expressed in VM, like EGFs and
EGFRs (Abe et al., 2009), could maintain ERK1/2 activation. The present observation of
unchanged activation of ERK1/2 and Akt pathways in E14.5 FGF-2 deficient VM is
consistent with this.
Both transduction pathways, ERK1/2 as well as Akt, were discussed to mediate
developmental cross-talk between FGF and Wnt signaling pathways by inactivation of
GSK-3β (Frodin and Gammeltoft, 1999, Torres et al., 1999, Dailey et al., 2005, Katoh,
2006), which prevents degradation of ß-catenin – a nuclear effector of canonical Wnt
signaling (MacDonald et al., 2009). Further, the presence of FGF-2 in neurosphere
assay resulted in increased nuclear β-catenin pool, which in turn promoted reentry into
cell cycle and self-renewal of neural progenitor cells, whereas the absence of FGF-2
promoted neuronal differentiation (Israsena et al., 2004). However, present western blot
analysis of protein extracts from VM of E14.5 FGF-2 deficient and wild type embryos
revealed any coherent decrease of β-catenin pool neither in the cytosol nor in the
nucleus. Consistent with this finding transcript levels of β-catenin, Wnt-1, Wnt-5a and
Wnt-7b were unchanged as determined by qRT-PCR in E14.5 and P0 VM of wild-type
and FGF-2 deficient mice (Dr. A. Ratzka, personal observation). Thus, is seems unlikely
that the canonical Wnt pathway is affected by FGF-2 loss in the analyzed mouse strain.
87
4.6 FGF signaling is involved in terminal differentiation
Disruption of FGFR1 signaling in postmitotic mDA neurons by overexpression of a
dominant negative FGFR1 mutant under control of the TH-promoter resulted in a
reduced cell size and decreased density of mDA neurons in the SNpc and VTA of
newborn transgenic mice indicating that FGFR1 mediated FGF signaling is necessary
for terminal mDA differentiation and/or maintenance (Klejbor et al., 2006). Distribution of
FGFR1 changes during development of the SNpc. At P4 FGFR1-ir is localized
predominantly in cell nuclei where it co-localizes with its partner CBP, whereas at P10,
after SNpc projections into the striatum have been completed, FGFR1-ir localization
changes to the cytoplasm (Fang et al., 2005). This differential distribution of FGFR1
moreover underlines the role of nuclear FGFR1 signaling during the terminal
differentiation and maturation of SNpc. Findings of present study are consistent with this
developmental regulation of FGFR1, suggesting an increased activation of INFS in FGF2 deficient mice. Since, the enhanced mDA neuron production in FGF-2 deficient VM
was correlated with two fold increased nuclear FGFR1 levels in E14.5 VM extracts of
FGF-2 deficient compared to wild-type mice,. Effects of activated INFS, accumulating in
the cell nucleus, including cell cycle exit and cell differentiation, are distinct from the
mitogenic effects of extracellular FGFs (Stachowiak et al., 2003, Stachowiak et al.,
2007, Stachowiak et al., 2011). Recently, nuclear HMW FGF-2 isoforms were shown to
activate INFS, which in turn mediated transcriptional initiation of the TH-promoter in
bovine adrenal medullary cells via direct interaction with CBP and activation of CREB
signaling (Peng et al., 2002, Stachowiak et al., 2007). Whereas, our findings suggest
that extracellular 18 kD FGF-2 might antagonize INFS in wild type mice, as FGF-2
deficient VM explants treated with recombinant 18 kDa FGF-2 displayed decreased
nuclear FGFR1 levels. In fact, in contrast to the nuclear HMW FGF-2 the impact of
extracellular and predominantly cytosolic 18 kDa FGF-2 (Florkiewicz et al., 1991) on
nuclear accumulation is very diverse. For some cells, i.e. Swiss 3T3 fibroblasts
treatment with 18 kDa FGF-2 increases also nuclear FGFR1 levels (Maher, 1996), while
in other cells including human neural progenitors or mouse embryonic stem cells,
88
extracellular 18 kDa FGF-2 fails to promote nuclear FGFR1 accumulation (Peng et al.,
2002, Stachowiak et al., 2007). In this context it is interesting to note that 23 kDa HMW
FGF-2 is the predominant isoform in the SNpc of adult rats (Claus et al., 2004b,
Tooyama et al., 1992), while 18 kDa FGF-2 and 21 kDa is highly expressed throughout
the CNS during late embryogenesis becoming downregulated in postnatal brain
(Giordano et al., 1992). INFS can be activated by a variety of FGF-unrelated factors, like
BMP7, hormonal receptors, NGF, neurotransmitters, and retinoic acid (Stachowiak et al.,
2007, Stachowiak et al., 2011). A compensatory recruitment of these INFS-activating
factors could stimulate the nuclear accumulation of FGFR1 in the mDA neurons lacking
all isoforms of FGF-2. Additionally, as mentioned in the chapter 4.3, FGF-3 is expressed
in embryonic VM as well as in the striatum, shares the nuclear localization with FGF-2,
and might therefore also participate in compensatory recruitment of INFS.
So far, an increased cell cycle exit of neuronal progenitor cells of the VZ and mDA
differentiation due to over-activation of integrative nuclear FGFR1 signaling pathway
appears as an attractive model to account for the supernumerary mDA neurons of
postnatal FGF-2 deficient mice. This mechanism might also explain the specific increase
of mDA neurons in SNpc but not in VTA of juvenile FGF-2 deficient mice, as previous
studies showed high levels of FGFR1 in adult SNpc, but low levels in the VTA (Ohmachi
et al., 2003).
4.7 Crosstalk of mDA and FGF signaling
Both nuclear receptors, Nurr1 and FGFR1, play key roles in signaling integration in the
nucleus. As discussed in chapter 4.6, FGFR-1 was shown to participate in postmitotic
development of mDA neurons and to activate TH-gene transcription in medulloblastoma
cells in cooperation with CBP. Nurr1 is a central transcription factor coordinating
postmitotic mDA neurons (Jacobs et al., 2009a, Jacobs et al., 2009b, Smidt and
Burbach, 2009). Nurr1 was shown to bind to NBRE sites in the TH promoter and
activate TH gene transcription (Iwawaki et al., 2000, Kim et al., 2003), in addition to
other important genes required for mDA homeostasis (Jacobs et al., 2009a). Both
89
factors are also discussed to regulate maturation and maintenance of mDA neurons
(Klejbor et al., 2006, Calo et al., 2005, Kadkhodaei et al., 2009). While FGFR1 is
ubiquitously expressed within the developing CNS (Ozawa et al., 1996), Nurr1
expression is restricted to specific areas, while during development the mDA neurons
are the only dopaminergic subtype expressing Nurr1 within the CNS (Backman et al.,
1999). This is consistent with present immunohistochemical studies in the ventral
midbrain of E14.5 embryos. While FGFR1 is widely distributed throughout the midbrain,
Nurr1 is restricted to the postmitotic precursors and maturating neurons of the mDA
field. Moreover, both proteins co-localize in the nucleus of the neurons located in the
mantle zone of the mDA field suggesting a co-localization of both proteins in the same
nuclear
complexes.
Furthermore,
immunoprecipitation
experiments
in
SV-40
immortalized rat VM progenitors as well as in neuroblastoma cells overexpressing
Nurr1-FLAG and FGFR1 proved the co-localization of Nurr1 and FGFR1 in the same
nuclear complexes.
Previous Fluorescence Recovery After Photobleaching (FRAP) experiments were
carried out in neuroblastoma cells transfected with FGFR1-EGFP illustrating the
dynamic changes of nuclear FGFR1, which exists in fast, slow, and immobile fraction
(Dunham-Ems et al., 2009). The immobile fraction was proposed to represent the not
functional nuclear matrix bound pool of FGFR1. The slow fraction represents the
chromatin bound FGFR1 (Dunham-Ems et al., 2009), which was shown to localize in
speckle domains at transcriptionally active sides, as determined by presence of snRNP
and splicosome assembly factor SC-35 (Peng et al., 2001, Stachowiak et al., 2003,
Stachowiak et al., 2007). The fast fraction is supposed to be the unbound, inactive, fast
diffusing FGFR1 (Dunham-Ems et al., 2009). Several factors have been identified to
influence the mobility of nuclear FGFR1. The influences were show to be functional and
were correlated with specific changes in the relative fractions of the nuclear fast, slow
and immobile FGFR1 populations. For example, treatment with cAMP decreased the
fast and depleted the immobile pools increasing the slow transcriptionally active FGFR1
pool (Dunham-Ems et al., 2009). This was consistent with cAMP dependent gene
90
activating function of nuclear FGFR1 in neuronal differentiation (Stachowiak et al., 2003,
Fang et al., 2005, Stachowiak et al., 2007). Recently, Benjamin Förthmann, PhD student
in our lab, performed FRAP experiments on FGFR1 mobility after co-transfection of
Nurr1-FLAG and plasmid encoding only FLAG-tag as negative control. The cotransfection of Nurr1-FLAG resulted in significantly increased slow, transcriptionally
active fraction of FGFR1 specifically in the nucleus and not in the cytoplasm of human
neuroblastoma cells. Accordingly, he detected also a decrease in the fast, not functional
FGFR1 fraction if compared to the co-transfection with vector encoding only for the
FLAG-tag (Baron et al., in preparation). The slowing down specifically of nuclear FGFR1
after Nurr1 co-transfection suggests that both proteins are co-engaged in chromatin
binding and gene transcription.
In fact, the co-engagement of nuclear FGFR1 and Nurr1 in transcriptional activation was
investigated in collaboration with Prof. Michal Stachowiak, Sridhar Narhla, and Chris
Terranova (Pathology and Anatomical Sciences, University of Buffalo, NY). To prove a
functional relevance of Nurr1 and nuclear FGFR1 interaction in transcriptional activation,
functional luciferase assays were performed. Neuroblastoma cells were co-transfected
with Nurr1 and/or engineered nuclear form of FGFR1 [FGFR1(SP-/NLS)], lacking a
transmembrane domain and provided with nuclear localization sequence. The promoter
specific initiation of transcription of luciferase gene was evaluated for: 1.) Nur responsive
element (NurRE), which is activated by the Nurr1- dimers; 2.) NBRE activated by Nurr1monomers. The overexpression of Nurr1-FLAG, resulted in activation of NurRE and
NBRE-driven luciferase expression. The overexpression of FGFR1(SP-/NLS) alone did
not influence the NurRE or NBRE dependent activation, while co-transfection of
FGFR1(SP-/NLS) with Nurr1-FLAG potentiated the NurRE as well as NBRE driven
expression of luciferase gene (Baron et al., in preparation). However, it remains to be
proven, whether the cooperation of Nurr1 and FGFR1 also results in transcriptional
activation of TH-gene expression, which is regulated by a promoter sequence containing
multiple NBRE-sites activated by Nurr1 (Kim et al., 2003).
91
Thus, nuclear FGFR1 co-engage with Nurr1 in transcriptional activation, although the
concrete characteristics, whether this interaction is direct and/or mediated by other cofactors in the nuclear complex, remains to be resolved. For example both pathways,
INFS as well as mDA differentiation, were shown to be activated by cAMP (Tremblay et
al., 2010, Malmersjo et al., 2010, Stachowiak et al., 2003) or in case of retinoids
regulated by different ligands (Stachowiak et al., 2011, Castro et al., 2001, Volakakis et
al., 2009). As mentioned in chapter 1.2.7 Nurr1 heterodimerizes with RXR and
depending on the ligand docosahexanoic acid promotes neuroprotection (Volakakis et
al., 2009, Smidt and Burbach, 2009). The role of all-trans RA through RXR/RAR
pathway in developing mDA neurons remains unresolved (compare chapter 1.2.5,
(Smidt and Burbach, 2009). Smidt and Burbach hypothesized recently, that the
multifaceted interactions of Nurr1 with several interaction partners reflect the possible
role of Nurr1 in convergence of many cellular events (Smidt and Burbach, 2009). Briefly,
the interaction with ERK2 and p57Kip2 increased transcriptional activity of Nurr1 (Zhang
et al., 2007, Joseph et al., 2003, Sacchetti et al., 2006), while interaction with LimK and
Lef-1 repressed its activity (Sacchetti et al., 2006, Kitagawa et al., 2007). The activation
of Wnt signaling resulted in stabilization of β-catenin, which in turn builds
transcriptionally functional complexes with Nurr1 and Lef-1 (Kitagawa et al., 2007).
However, nuclear FGFR1 might be more than an essential additional player within the
complex interaction network coordinated by Nurr1. This interaction should be rather
valued as a novel integrative mechanism, mediated via cross-talk of two key players
within two complex developmental signaling cascades.
Altogether, the idea of increased INFS as underling mechanism for development of
increased mDA neurons number in FGF-2 deficient mice (compare chapter 4.6)
becomes even more compelling with regard to interaction of nuclear FGFR1 with Nurr1
during mDA development. Hypothetically, the increased FGFR1 should increase the
probability of interaction with Nurr1, which in turn might result in increased Nurr1
dependent gene expression, like TH, due to co-engagement of FGFR1 with Nurr1 in
92
transcriptional activity. Unfortunately in vivo IP of Nurr1 with FGFR1 was not
reproducible, although the first two experiments showed an interaction in FGF-2
deficient tissue but not wt tissue. The difficulties in the reproduction of endogenous CoIPs are due to very small nuclear protein amounts, which can be extracted from freshly
prepared E14.5 VMs. Additionally, FGFR1 protein seems to be unstable after isolation
and cryoconservation, which complicates the harvesting of the appropriate amount of
the tissue.
4.8 FGF-2 is involved in maturation and target innervation
Nigrostriatal wiring occurs between E18 und P4, followed by maturation until P28.
During this developmental process in the substantia nigra those mDA neurons which
failed to establish adequate projections to the striatum undergo natural cell death with
main peaks at P2 and P14 (Burke, 2003, Burke, 2004, Prakash and Wurst, 2006). Most
probably multiple trophic inputs are required for establishment of functionally adequate
nigrostriatal projections. In other neuronal systems FGF-2 participates in neuronal
network establishment in very diverse manner (compare (Umemori, 2009)). FGF-2 was
shown to regulate axon guidance of spinal motor neurons (Shirasaki et al., 2006), target
recognition of the Xenopus retinal ganglion cells (McFarlane et al., 1995, Webber et al.,
2003, Webber et al., 2005), as well as synaptic differentiation, inducing synaptic vesicle
aggregation in Xenopus spinal cord neurons (Dai and Peng, 1995) and promoting
neurite elongation and branching of hippocampal neurons in vitro (Li et al., 2002) and
regulate their spine morphology in vivo (Zechel et al., 2009). Consistent with this are the
findings of the present study, suggesting that FGF-2 loss compromises adequate
nigrostriatal pathway formation, including target recognition and maybe also path
finding. First evidence therefore is provided by decrease of apoptotic bodies missing
caspase-3 immunoreactivity at stage PO in FGF-2 deficient mice, indicating a probability
of reduced wiring control allowing inadequately wired fibers and the respective cells to
retain. Whether the significantly more caspase-3 negative apoptotic profiles within the
wild type compared to FGF-2 deficient SNpc represent the late apoptotic stages (Brecht
93
et al., 2001) or a different caspase-3 independent apoptotic cell death (Jeon et al., 1999)
remains unclear. Postnatal MFB lesioning experiments showed that besides the
conventional apoptotic signaling routes a caspase-3 independent cell death in the mDA
system is feasible (Jeon et al., 1999). The authors investigated developmental cell death
in the SNpc, natural developmental neuron death, and induced developmental death
following either striatal target injury with quinolinic acid or dopamine terminal lesion with
intrastriatal injection of 6-OHDA. Caspase-3 dependent apoptosis was shown to occur
during all three investigated cell death models. However, after 6-OHDA lesion only 16%
of apoptotic profiles were caspase-3-positive in contrast to 59% after striatal target
injury. Since the immunohistochemical techniques were unchanged, the authors
suggested, that this difference might be due to mDA specific toxicity of 6-OHDA.
Therefore they discussed a possibility of a different caspase-3 independent apoptotic
process in this context.
The reduced ontogenic cell death is correlated with increased fiber outgrowth in FGF-2
deficient VM and FB explant co-cultures. The heterogenous co-cultures missing FGF-2
in FB or VM, respectively, show a similar phenotype. Compared to co-cultures lacking
FGF-2 in both VM and FB they have significantly shorter fibers. Additionally, if compared
to pure wild type co-cultures they show significantly wider tracts. This indicates a
complex interplay between mesencephalic and telencephalic and/or diencephalic FGF
signaling during pathfinding. Previous VM explant culture studies revealed a biphasic
TH-ir fiber outgrowth, first occurs the glia-independent straight long distance outgrowth
followed by astrocyte-dependent network forming short distance fiber outgrowth
(Johansson and Stromberg, 2003). Another study indicates that GDNF specifically
regulates the astrocyte dependent TH-ir fiber outgrowth, by inducing the migration of
astrocytes (Bjerken et al., 2007). In fact, GDNF has been hypothesized to be a leading
striatum-derived neurotrophic factor for mDA neurons (Burke, 2006). One study also
reports a cooperative requirement of GDNF for FGF-2 mediated neuroprotection in
94
hippocampal neuron cultures, showing a regulatory function of FGF-2 on GDNF
expression (Lenhard et al., 2002).
However, it remains to be resolved how FGF-2 loss affects the astrocytes during the
nigrostriatal pathway formation, which isoforms participate hereby as well as what are
the sources of FGF-2 and the respective receptors. On the other hand a retrograde
(Ferguson and Johnson, 1991) as well as anterograde transport of FGF-2 within
nigrostriatal pathway (McGeer et al., 1992) imply a paracrine or autocrine mechanism
involving the secreted 18 kDa FGF-2 isoform in maturation and maintenance of the
nigrostriatal system. Hereby, the reports of induction of apoptosis via FGFR3 in the
peripheral nervous system (Jungnickel et al., 2004) as well as via FGFR2 in cortical
systems (Maric et al., 2007) support a hypothetical involvement of the secreted FGF-2.
However, additionally to extrinsically mediated, programmed cell death pathways, also a
direct effect of intrinsic mDA derived FGF-2 should be considered. Apart from the fact
that the LMW-FGF-2 is the predominant form during development, the 22kD HMW-FGF2 is also present in developing rat brain (Giordano et al., 1992), suggesting that HMWFGF-2 might also be involved in the regulation of adequate nigrostriatal pathway
formation. In fact, the HMW-FGF-2 has already been shown to induce apoptosis in vitro
(Ma et al., 2007). Accordingly, there are several putative sources for FGF-2 which may
ensure intact pathfinding and an adequate response to natural cell death initiation (Fig.
18): 1.) extrinsic VM derived FGF-2, which activates FGF-signaling via transmembrane
FGFR1; 2.) extrinsic FGF-2 produced by striatal astrocytes can also activate the FGFR
located on the membrane surface or become internalized by dopaminergic cells and be
transported to the soma; 3.) internalized extrinsic or intrinsic mDA neuron derived FGF-2
may modulate intracellular responses of other signaling pathways, in the soma, nucleus
or axonal growth cones.
95
1
3
2
Figure 18. Putative sources of FGF-2 during nigrostriatal pathway formation. 1.) extrinsic VM derived
FGF-2, which activates FGF-signaling via transmembrane FGFR1; 2.) extrinsic FGF-2 produced by striatal
astrocytes can also activate the FGFR located on the membrane surface or become internalized by
dopaminergic cells and be transported to the soma; 3.) internalized extrinsic or intrinsic VM derived FGF-2
may modulate intracellular responses of other signaling pathways, in the soma, nucleus or axonal growth
cones.
4.9 Multimodal role of FGF-2 in developing VM
Similar to the peripheral nervous system (Jungnickel et al., 2005) FGF-2 seems to have
multimodal functions in the CNS (Grothe and Wewetzer, 1996). The presence of FGF-2
is differentially regulated in embryonic (Weise et al., 1993), comparing to maturating and
adult brain stem (Grothe et al., 1991) indicating that the function of the protein
undergoes changes during development (Grothe and Wewetzer, 1996). This is
consistent with current study, which demonstrates that FGF-2 can differentially regulate
two independent developmental physiological processes in one functional system, like
mDA progenitor differentiation and nigrostriatal pathway formation. The multifaceted
functions of FGF-2 in mDA system might correlate with the expression pattern of the
different molecular isoforms of FGF-2 in the CNS. Since LMW-FGF-2 is the predominant
isoform in the embryonic human and rat brain showing decrease in the adult brain
(Giordano et al., 1992), this isoform might be the best candidate for regulation of
embryonic mDA development. The fact that 23 kDa HMW isoform becomes more
important in the maturing postnatal brain (Giordano et al., 1992) plus the finding that
exogenously applied HMW-FGF-2 leads to a better graft survival than LMW-FGF-2 in a
96
rat model of Parkinson disease (Timmer et al., 2004), indicates that the 23 kD HMWFGF-2 form is most probably required for maintenance and survival.
Other mDA relevant neurotrophic factors like GDNF, BDNF and TGF-ß, which are
known to support survival of VM mDA system (Krieglstein, 2004), also seem to have
multifunctional roles in developing and mature mDA system, similar to FGF-2. The
implication of GDNF in mDA differentiation (Roussa and Krieglstein, 2004) and
nigrostriatal innervation (Kholodilov et al., 2004), of BDNF in mDA neuron differentiation
(Baquet et al., 2005) and of TGF-β in ectopic induction of the mDA phenotype (Roussa
et al., 2006) during development, actually might underline their need in maintenance of
the adult nigrostriatal system. Interestingly, although TGF-ß is capable of induction of
mDA phenotype in vitro, the loss-of-function mouse mutants show likewise moderate the
phenotype (Roussa et al., 2006).
Most presumably, FGF-2 loss seems to be compensated during development by
presence of other functionally similar FGFs or other neurotrophic factors, like GDNF or
BDNF, implying a redundancy of this protein concerning development of SNpc. This is
reflected in a milder phenotype in the present loss-of-function model. Further, the
unchanged mRNA levels of FGFs do not exclude a possible redundancy effect
completely. The compensation of FGF-2 loss could also occur on protein level, like
shown here with FGFR1 accumulation (chapter 4.6), or simply due to presence of other
FGFs, like FGF-8 or FGF-3, during certain developmental time points (compare chapter
4.3 and 4.4). Anyway, it remains to be resolved, which mechanism is responsible for
increased accumulation of FGFR1 in the nucleus of FGF-2 deficient mice. Altogether,
the multimodal involvement of FGF-2 in SNpc development and maintenance imply a
presence of a regulatory loop, which ensures a functionally proper development of
SNpc. Possible evidences for this are: 1.) the opposite phenotype in the FGF-2 gain of
function model (FGF-2 overexpressing mice) suggests a coherent dose dependent
regulation (Timmer et al., 2007); 2.) since FGF-2 is involved in regulation of dopamine
turnover (Forget et al., 2006), some evolutionary evolved compensatory developmental
97
mechanisms might allow an establishment of a coherent cellular input with regard to
increased cell number and quality of the wiring, when the dopamine transmission is
diminished; 3.) also, the neuroprotective role of FGF-2 for mDA neurons might reflect
the requirement of an initial development of increased number of mDA neurons, to
compensate their increased vulnerability (compare chapter 4.1).
4.10 Concluding remarks
In
conclusion,
both
physiological
alterations,
the
increased
mDA
progenitor
differentiation and reduced ontogenic cell death in the developing SNpc of FGF-2
deficient mice, could explain the phenotype with increased mDA neuron number,
proving the regulatory role of FGF-2 in SNpc development. The regulatory participation
of FGF-2 during SNpc development underlines the importance of this neurotrophic factor
in maintenance and regulation of the mature nigrostriatal system. A recent gene analysis
in PD patients and controls showed association of single nucleotide polymorphism in the
FGF-2 gene with sporadic PD cases (Mizuta et al., 2008), emphasizing the previously
suspected involvement of FGF-2 in PD, after a reduced FGF-ir of mDA neurons has
been detected in postmortem tissue of PD patients (Tooyama et al., 1993, Tooyama et
al., 1994). Detailed understanding of the developmental processes in VM, including the
role of trophic factors like FGF-2, can enlighten the progredient pathophysiology of PD
and might lead to improved treatment options for PD patients, such as cell replacement
therapy. Exemplary, this study revealed a novel INFS / Nurr1 interactive mechanism for
gene activation during neuronal development. This regulatory mechanism may offer a
new therapeutic target for increasing production of mDA neurons in neurodevelopmental
(schizophrenia) and neurodegenerative (Parkinson’s) disorders. Future studies on mDA
neuron development in isoform specific FGF-2 mouse mutants may clarify the specific
involvement of different FGF-2 isoforms in distinct physiological events during SNpc
development.
98
5. References
ABE, Y., NAMBA, H., ZHENG, Y. & NAWA, H. 2009. In situ hybridization reveals
developmental regulation of ErbB1-4 mRNA expression in mouse midbrain:
implication of ErbB receptors for dopaminergic neurons. Neuroscience, 161, 95110.
ABELIOVICH, A. & HAMMOND, R. 2007. Midbrain dopamine neuron differentiation:
factors and fates. Dev Biol, 304, 447-54.
ABERCROMBIE, M. & JOHNSON, M. L. 1946. Quantitative histology of Wallerian
degeneration: I. Nuclear population in rabbit sciatic nerve. J Anat, 80, 37-50.
AKAI, J. & STOREY, K. 2003. Brain or brawn: how FGF signaling gives us both. Cell,
115, 510-2.
ALBERI, L., SGADO, P. & SIMON, H. H. 2004. Engrailed genes are cell-autonomously
required to prevent apoptosis in mesencephalic dopaminergic neurons.
Development, 131, 3229-36.
AMALRIC, F., BALDIN, V., BOSC-BIERNE, I., BUGLER, B., COUDERC, B., GUYADER,
M., PATRY, V., PRATS, H., ROMAN, A. M. & BOUCHE, G. 1991. Nuclear
translocation of basic fibroblast growth factor. Ann N Y Acad Sci, 638, 127-38.
ANDERSSON, E., TRYGGVASON, U., DENG, Q., FRILING, S., ALEKSEENKO, Z.,
ROBERT, B., PERLMANN, T. & ERICSON, J. 2006. Identification of intrinsic
determinants of midbrain dopamine neurons. Cell, 124, 393-405.
ANTOINE, M., REIMERS, K., DICKSON, C. & KIEFER, P. 1997. Fibroblast growth
factor 3, a protein with dual subcellular localization, is targeted to the nucleus and
nucleolus by the concerted action of two nuclear localization signals and a
nucleolar retention signal. J Biol Chem, 272, 29475-81.
ARENAS, E. 2010. Towards stem cell replacement therapies for Parkinson's disease.
Biochem Biophys Res Commun, 396, 152-6.
BACKMAN, C., PERLMANN, T., WALLEN, A., HOFFER, B. J. & MORALES, M. 1999. A
selective group of dopaminergic neurons express Nurr1 in the adult mouse brain.
Brain Res, 851, 125-32.
BAQUET, Z. C., BICKFORD, P. C. & JONES, K. R. 2005. Brain-derived neurotrophic
factor is required for the establishment of the proper number of dopaminergic
neurons in the substantia nigra pars compacta. J Neurosci, 25, 6251-9.
BARYSHNIKOVA, L. M., VON BOHLEN UND HALBACH, O., KAPLAN, S. & VON
BARTHELD, C. S. 2006. Two distinct events, section compression and loss of
particles ("lost caps"), contribute to z-axis distortion and bias in optical disector
counting. Microsc Res Tech, 69, 738-56.
BEAN, A. J., ELDE, R., CAO, Y. H., OELLIG, C., TAMMINGA, C., GOLDSTEIN, M.,
PETTERSSON, R. F. & HOKFELT, T. 1991. Expression of acidic and basic
fibroblast growth factors in the substantia nigra of rat, monkey, and human. Proc
Natl Acad Sci U S A, 88, 10237-41.
99
BEAN, A. J., OELLIG, C., PETTERSSON, R. F. & HOKFELT, T. 1992. Differential
expression of acidic and basic FGF in the rat substantia nigra during
development. Neuroreport, 3, 993-6.
BJERKEN, S., BOGER, H. A., NELSON, M., HOFFER, B. J., GRANHOLM, A. C. &
STROMBERG, I. 2007. Effects of glial cell line-derived neurotrophic factor
deletion on ventral mesencephalic organotypic tissue cultures. Brain Res, 1133,
10-9.
BJORKLUND, A., STENEVI, U., SCHMIDT, R. H., DUNNETT, S. B. & GAGE, F. H.
1983. Intracerebral grafting of neuronal cell suspensions. I. Introduction and
general methods of preparation. Acta Physiol Scand Suppl, 522, 1-7.
BJORKLUND, T., CEDERFJALL, E. A. & KIRIK, D. 2010. Gene therapy for dopamine
replacement. Prog Brain Res, 184, 221-35.
BJORKLUND, T. & KORDOWER, J. H. 2010. Gene therapy for Parkinson's disease.
Mov Disord, 25 Suppl 1, S161-73.
BLAKELY, B. D., BYE, C. R., FERNANDO, C. V., HORNE, M. K., MACHEDA, M. L.,
STACKER, S. A., ARENAS, E. & PARISH, C. L. 2011. Wnt5a regulates midbrain
dopaminergic axon growth and guidance. PLoS One, 6, e18373.
BORGAL, L., HONG, M., SADI, D. & MENDEZ, I. 2007. Differential effects of glial cell
line-derived neurotrophic factor on A9 and A10 dopamine neuron survival in vitro.
Neuroscience, 147, 712-9.
BORTA, A. & HOGLINGER, G. U. 2007. Dopamine and adult neurogenesis. J
Neurochem, 100, 587-95.
BOUVIER, M. M. & MYTILINEOU, C. 1995. Basic fibroblast growth factor increases
division and delays differentiation of dopamine precursors in vitro. J Neurosci, 15,
7141-9.
BRECHT, S., GELDERBLOM, M., SRINIVASAN, A., MIELKE, K., DITYATEVA, G. &
HERDEGEN, T. 2001. Caspase-3 activation and DNA fragmentation in primary
hippocampal neurons following glutamate excitotoxicity. Brain Res Mol Brain Res,
94, 25-34.
BUGLER, B., AMALRIC, F. & PRATS, H. 1991. Alternative initiation of translation
determines cytoplasmic or nuclear localization of basic fibroblast growth factor.
Mol Cell Biol, 11, 573-7.
BURKE, R. E. 2003. Postnatal developmental programmed cell death in dopamine
neurons. Ann N Y Acad Sci, 991, 69-79.
BURKE, R. E. 2004. Ontogenic cell death in the nigrostriatal system. Cell Tissue Res,
318, 63-72.
BURKE, R. E. 2006. GDNF as a candidate striatal target-derived neurotrophic factor for
the development of substantia nigra dopamine neurons. J Neural Transm Suppl,
41-5.
CALO, L., SPILLANTINI, M., NICOLETTI, F. & ALLEN, N. D. 2005. Nurr1 co-localizes
with EphB1 receptors in the developing ventral midbrain, and its expression is
enhanced by the EphB1 ligand, ephrinB2. J Neurochem, 92, 235-45.
100
CASTELO-BRANCO, G., WAGNER, J., RODRIGUEZ, F. J., KELE, J., SOUSA, K.,
RAWAL, N., PASOLLI, H. A., FUCHS, E., KITAJEWSKI, J. & ARENAS, E. 2003.
Differential regulation of midbrain dopaminergic neuron development by Wnt-1,
Wnt-3a, and Wnt-5a. Proc Natl Acad Sci U S A, 100, 12747-52.
CHADI, G., MOLLER, A., ROSEN, L., JANSON, A. M., AGNATI, L. A., GOLDSTEIN, M.,
OGREN, S. O., PETTERSSON, R. F. & FUXE, K. 1993. Protective actions of
human recombinant basic fibroblast growth factor on MPTP-lesioned nigrostriatal
dopamine neurons after intraventricular infusion. Exp Brain Res, 97, 145-58.
CHEN, H., TUNG, Y. C., LI, B., IQBAL, K. & GRUNDKE-IQBAL, I. 2007. Trophic factors
counteract elevated FGF-2-induced inhibition of adult neurogenesis. Neurobiol
Aging, 28, 1148-62.
CHUNG, C. Y., SEO, H., SONNTAG, K. C., BROOKS, A., LIN, L. & ISACSON, O. 2005.
Cell type-specific gene expression of midbrain dopaminergic neurons reveals
molecules involved in their vulnerability and protection. Hum Mol Genet, 14,
1709-25.
CHUNG, S., LEUNG, A., HAN, B. S., CHANG, M. Y., MOON, J. I., KIM, C. H., HONG,
S., PRUSZAK, J., ISACSON, O. & KIM, K. S. 2009. Wnt1-lmx1a forms a novel
autoregulatory loop and controls midbrain dopaminergic differentiation
synergistically with the SHH-FoxA2 pathway. Cell Stem Cell, 5, 646-58.
CINTRA, A., CAO, Y., OELLIG, C., TINNER, B., BORTOLOTTI, F., GOLDSTEIN, M.,
PETTERSSON, R. F. & FUXE, K. 1991. Basic FGF is present in dopaminergic
neurons of the ventral midbrain of the rat. Neuroreport, 2:597-600.
CLAUS, P., BRUNS, A. F. & GROTHE, C. 2004a. Fibroblast growth factor-2(23) binds
directly to the survival of motoneuron protein and is associated with small nuclear
RNAs. Biochem J, 384, 559-65.
CLAUS, P., WERNER, S., TIMMER, M. & GROTHE, C. 2004b. Expression of the
fibroblast growth factor-2 isoforms and the FGF receptor 1-4 transcripts in the rat
model system of Parkinson's disease. Neurosci Lett, 360, 117-20.
DAI, Z. & PENG, H. B. 1995. Presynaptic differentiation induced in cultured neurons by
local application of basic fibroblast growth factor. J Neurosci, 15, 5466-75.
DAILEY, L., AMBROSETTI, D., MANSUKHANI, A. & BASILICO, C. 2005. Mechanisms
underlying differential responses to FGF signaling. Cytokine Growth Factor Rev,
16, 233-47.
DOE, C. Q. 2008. Neural stem cells: balancing self-renewal with differentiation.
Development, 135, 1575-87.
DONO, R. 2003. Fibroblast growth factors as regulators of central nervous system
development and function. Am J Physiol Regul Integr Comp Physiol, 284, R86781.
DONO, R., TEXIDO, G., DUSSEL, R., EHMKE, H. & ZELLER, R. 1998. Impaired
cerebral cortex development and blood pressure regulation in FGF-2-deficient
mice. EMBO J, 17, 4213-25.
101
DORPH-PETERSEN, K. A., NYENGAARD, J. R. & GUNDERSEN, H. J. 2001. Tissue
shrinkage and unbiased stereological estimation of particle number and size. J
Microsc, 204, 232-46.
DUNHAM-EMS, S. M., LEE, Y. W., STACHOWIAK, E. K., PUDAVAR, H., CLAUS, P.,
PRASAD, P. N. & STACHOWIAK, M. K. 2009. Fibroblast growth factor receptor-1
(FGFR1) nuclear dynamics reveal a novel mechanism in transcription control. Mol
Biol Cell, 20, 2401-12.
DUNHAM-EMS, S. M., PUDAVAR, H. E., MYERS, J. M., MAHER, P. A., PRASAD, P. N.
& STACHOWIAK, M. K. 2006. Factors controlling fibroblast growth factor
receptor-1's cytoplasmic trafficking and its regulation as revealed by FRAP
analysis. Mol Biol Cell, 17, 2223-35.
EBERT, A. D., LAUSSMANN, M., WEGEHINGEL, S., KADERALI, L., ERFLE, H.,
REICHERT, J., LECHNER, J., BEER, H. D., PEPPERKOK, R. & NICKEL, W.
2010. Tec-kinase-mediated phosphorylation of fibroblast growth factor 2 is
essential for unconventional secretion. Traffic, 11, 813-26.
EISELLEOVA, L., MATULKA, K., KRIZ, V., KUNOVA, M., SCHMIDTOVA, Z., NERADIL,
J., TICHY, B., DVORAKOVA, D., POSPISILOVA, S., HAMPL, A. & DVORAK, P.
2009. A complex role for FGF-2 in self-renewal, survival, and adhesion of human
embryonic stem cells. Stem Cells, 27, 1847-57.
ESWARAKUMAR, V. P., LAX, I. & SCHLESSINGER, J. 2005. Cellular signaling by
fibroblast growth factor receptors. Cytokine Growth Factor Rev, 16, 139-49.
FANG, X., STACHOWIAK, E. K., DUNHAM-EMS, S. M., KLEJBOR, I. & STACHOWIAK,
M. K. 2005. Control of CREB-binding protein signaling by nuclear fibroblast
growth factor receptor-1: a novel mechanism of gene regulation. J Biol Chem,
280, 28451-62.
FARKAS, L. M., DUNKER, N., ROUSSA, E., UNSICKER, K. & KRIEGLSTEIN, K. 2003.
Transforming growth factor-beta(s) are essential for the development of midbrain
dopaminergic neurons in vitro and in vivo. J Neurosci, 23, 5178-86.
FAWCETT, J. W., BARKER, R. A. & DUNNETT, S. B. 1995. Dopaminergic neuronal
survival and the effects of bFGF in explant, three dimensional and monolayer
cultures of embryonic rat ventral mesencephalon. Experimental Brain Research,
106, 275-282.
FERGUSON, I. A. & JOHNSON, E. M. 1991. Fibroblast growth factor receptor-bearing
neurons in the CNS: Identification by receptor-mediated retrograde transport. J.
Comp. Neurol., 313, 693-706.
FERRARI, G., MINOZZI, M. C., TOFFANO, G., LEON, A. & SKAPER, S. D. 1989. Basic
fibroblast growth factor promotes the survival and development of mesencephalic
neurons in culture. Dev Biol, 133, 140-7.
FERRI, A. L., LIN, W., MAVROMATAKIS, Y. E., WANG, J. C., SASAKI, H., WHITSETT,
J. A. & ANG, S. L. 2007. Foxa1 and Foxa2 regulate multiple phases of midbrain
dopaminergic neuron development in a dosage-dependent manner.
Development, 134, 2761-9.
102
FISCHER, T., FAUS-KESSLER, T., WELZL, G., SIMEONE, A., WURST, W. &
PRAKASH, N. 2011. Fgf15-mediated control of neurogenic and proneural gene
expression regulates dorsal midbrain neurogenesis. Dev Biol, 350, 496-510.
FLORKIEWICZ, R. Z., BAIRD, A. & GONZALEZ, A. M. 1991. Multiple forms of bFGF:
differential nuclear and cell surface localization. Growth Factors, 4, 265-75.
FORD-PERRISS, M., ABUD, H. & MURPHY, M. 2001. Fibroblast growth factors in the
developing central nervous system. Clin Exp Pharmacol Physiol, 28, 493-503.
FORGET, C., STEWART, J. & TRUDEAU, L. E. 2006. Impact of basic FGF expression
in astrocytes on dopamine neuron synaptic function and development. Eur J
Neurosci, 23, 608-16.
FRODIN, M. & GAMMELTOFT, S. 1999. Role and regulation of 90 kDa ribosomal S6
kinase (RSK) in signal transduction. Mol Cell Endocrinol, 151, 65-77.
GALE, E. & LI, M. 2008. Midbrain dopaminergic neuron fate specification: Of mice and
embryonic stem cells. Mol Brain, 1, 8.
GALLEGUILLOS, D., VECCHIOLA, A., FUENTEALBA, J. A., OJEDA, V., ALVAREZ, K.,
GOMEZ, A. & ANDRES, M. E. 2004. PIASgamma represses the transcriptional
activation induced by the nuclear receptor Nurr1. J Biol Chem, 279, 2005-11.
GERMAN, D. C., MANAYE, K. F., SONSALLA, P. K. & BROOKS, B. A. 1992. Midbrain
dopaminergic cell loss in Parkinson's disease and MPTP-induced parkinsonism:
sparing of calbindin-D28k-containing cells. Ann N Y Acad Sci, 648, 42-62.
GIMENO, L. & MARTINEZ, S. 2007. Expression of chick Fgf19 and mouse Fgf15
orthologs is regulated in the developing brain by Fgf8 and Shh. Dev Dyn, 236,
2285-97.
GIORDANO, S., SHERMAN, L., LYMAN, W. & MORRISON, R. 1992. Multiple molecular
weight forms of basic fibroblast growth factor are developmentally regulated in
the central nervous system. Developmental Biology, 152, 293-303.
GONZALEZ, A. M., BERRY, M., MAHER, P. A., LOGAN, A. & BAIRD, A. 1995. A
comprehensive analysis of the distribution of FGF-2 and FGFR1 in the rat brain.
Brain Res, 701, 201-26.
GONZALEZ, A. M., BUSCAGLIA, M., ONG, M. & BAIRD, A. 1990. Distribution of basic
fibroblast growth factor in the 18-day rat fetus: localization in the basement
membranes of diverse tissues. J Cell Biol, 110, 753-65.
GROTHE, C., SCHULZE, A., SEMKOVA, I., MULLER-OSTERMEYER, F., REGE, A. &
WEWETZER, K. 2000. The high molecular weight fibroblast growth factor-2
isoforms (21,000 mol. wt and 23,000 mol. wt) mediate neurotrophic activity on rat
embryonic mesencephalic dopaminergic neurons in vitro. Neuroscience, 100, 7386.
GROTHE, C. & TIMMER, M. 2007. The physiological and pharmacological role of basic
fibroblast growth factor in the dopaminergic nigrostriatal system. Brain Res Rev,
54, 80-91.
GROTHE, C., TIMMER, M., SCHOLZ, T., WINKLER, C., NIKKHAH, G., CLAUS, P.,
ITOH, N. & ARENAS, E. 2004. Fibroblast growth factor-20 promotes the
103
differentiation of Nurr1-overexpressing neural stem cells into tyrosine
hydroxylase-positive neurons. Neurobiol Dis, 17, 163-70.
GROTHE, C. & WEWETZER, K. 1996. Fibroblast growth factor and its implications for
developing and regenerating neurons. Int J Dev Biol, 40, 403-10.
GROTHE, C., ZACHMANN, K. & UNSICKER, K. 1991. Basic FGF-like immunoreactivity
in the developing and adult rat brainstem. J Comp Neurol, 305, 328-36.
GUILLERY, R. W. 2002. On counting and counting errors. J Comp Neurol, 447, 1-7.
GUILLERY, R. W. & HERRUP, K. 1997. Quantification without pontification: choosing a
method for counting objects in sectioned tissues. J Comp Neurol, 386, 2-7.
GUNDERSEN, H. J., BAGGER, P., BENDTSEN, T. F., EVANS, S. M., KORBO, L.,
MARCUSSEN, N., MOLLER, A., NIELSEN, K., NYENGAARD, J. R.,
PAKKENBERG, B. & ET AL. 1988. The new stereological tools: disector,
fractionator, nucleator and point sampled intercepts and their use in pathological
research and diagnosis. APMIS, 96, 857-81.
GUO, Q., LI, K., SUNMONU, N. A. & LI, J. Y. 2010. Fgf8b-containing spliceforms, but
not Fgf8a, are essential for Fgf8 function during development of the midbrain and
cerebellum. Dev Biol, 338, 183-92.
HARDY, J. 2010. Genetic analysis of pathways to Parkinson disease. Neuron, 68, 2016.
HEDREEN, J. C. 1998. What was wrong with the Abercrombie and empirical cell
counting methods? A review. Anat Rec, 250, 373-80.
HERMANSON, E., BORGIUS, L., BERGSLAND, M., JOODMARDI, E. & PERLMANN,
T. 2006. Neuropilin1 is a direct downstream target of Nurr1 in the developing
brain stem. J Neurochem, 97, 1403-11.
HERNANDEZ-MONTIEL, H. L., TAMARIZ, E., SANDOVAL-MINERO, M. T. & VARELAECHAVARRIA, A. 2008. Semaphorins 3A, 3C, and 3F in mesencephalic
dopaminergic axon pathfinding. J Comp Neurol, 506, 387-97.
HU, Y., FANG, X., DUNHAM, S. M., PRADA, C., STACHOWIAK, E. K. &
STACHOWIAK, M. K. 2004. 90-kDa ribosomal S6 kinase is a direct target for the
nuclear fibroblast growth factor receptor 1 (FGFR1): role in FGFR1 signaling. J
Biol Chem, 279, 29325-35.
HYNES, M., PORTER, J. A., CHIANG, C., CHANG, D., TESSIER-LAVIGNE, M.,
BEACHY, P. A. & ROSENTHAL, A. 1995. Induction of midbrain dopaminergic
neurons by Sonic hedgehog. Neuron, 15, 35-44.
ISRASENA, N., HU, M., FU, W., KAN, L. & KESSLER, J. A. 2004. The presence of
FGF2 signaling determines whether beta-catenin exerts effects on proliferation or
neuronal differentiation of neural stem cells. Dev Biol, 268, 220-31.
ITOH, N. 2007. The Fgf families in humans, mice, and zebrafish: their evolutional
processes and roles in development, metabolism, and disease. Biol Pharm Bull,
30, 1819-25.
ITOH, N. & ORNITZ, D. M. 2008. Functional evolutionary history of the mouse Fgf gene
family. Dev Dyn, 237, 18-27.
104
ITOH, N. & ORNITZ, D. M. 2011. Fibroblast growth factors: from molecular evolution to
roles in development, metabolism and disease. J Biochem, 149, 121-30.
IWAWAKI, T., KOHNO, K. & KOBAYASHI, K. 2000. Identification of a potential nurr1
response element that activates the tyrosine hydroxylase gene promoter in
cultured cells. Biochem Biophys Res Commun, 274, 590-5.
JACKSON-LEWIS, V., VILA, M., DJALDETTI, R., GUEGAN, C., LIBERATORE, G., LIU,
J., O'MALLEY, K. L., BURKE, R. E. & PRZEDBORSKI, S. 2000. Developmental
cell death in dopaminergic neurons of the substantia nigra of mice. J Comp
Neurol, 424, 476-88.
JACOBS, F. M., SMITS, S. M., NOORLANDER, C. W., VON OERTHEL, L., VAN DER
LINDEN, A. J., BURBACH, J. P. & SMIDT, M. P. 2007. Retinoic acid counteracts
developmental defects in the substantia nigra caused by Pitx3 deficiency.
Development, 134, 2673-84.
JACOBS, F. M., VAN DER LINDEN, A. J., WANG, Y., VON OERTHEL, L., SUL, H. S.,
BURBACH, J. P. & SMIDT, M. P. 2009a. Identification of Dlk1, Ptpru and Klhl1 as
novel Nurr1 target genes in meso-diencephalic dopamine neurons. Development,
136, 2363-73.
JACOBS, F. M., VAN ERP, S., VAN DER LINDEN, A. J., VON OERTHEL, L.,
BURBACH, J. P. & SMIDT, M. P. 2009b. Pitx3 potentiates Nurr1 in dopamine
neuron terminal differentiation through release of SMRT-mediated repression.
Development, 136, 531-40.
JENSEN, P., PEDERSEN, E. G., ZIMMER, J., WIDMER, H. R. & MEYER, M. 2008.
Functional effect of FGF2- and FGF8-expanded ventral mesencephalic precursor
cells in a rat model of Parkinson's disease. Brain Res, 1218, 13-20.
JEON, B. S., KHOLODILOV, N. G., OO, T. F., KIM, S. Y., TOMASELLI, K. J.,
SRINIVASAN, A., STEFANIS, L. & BURKE, R. E. 1999. Activation of caspase-3
in developmental models of programmed cell death in neurons of the substantia
nigra. J Neurochem, 73, 322-33.
JOHANSSON, S. & STROMBERG, I. 2003. Fetal lateral ganglionic eminence attracts
one of two morphologically different types of tyrosine hydroxylase-positive nerve
fibers formed by cultured ventral mesencephalon. Cell Transplant, 12, 243-55.
JOSEPH, B., WALLEN-MACKENZIE, A., BENOIT, G., MURATA, T., JOODMARDI, E.,
OKRET, S. & PERLMANN, T. 2003. p57(Kip2) cooperates with Nurr1 in
developing dopamine cells. Proc Natl Acad Sci U S A, 100, 15619-24.
JUNGNICKEL, J., GRANSALKE, K., TIMMER, M. & GROTHE, C. 2004. Fibroblast
growth factor receptor 3 signaling regulates injury-related effects in the peripheral
nervous system. Mol Cell Neurosci, 25, 21-9.
JUNGNICKEL, J., KLUTZNY, A., GUHR, S., MEYER, K. & GROTHE, C. 2005.
Regulation of neuronal death and calcitonin gene-related peptide by fibroblast
growth factor-2 and FGFR3 after peripheral nerve injury: evidence from mouse
mutants. Neuroscience, 134, 1343-50.
KADKHODAEI, B., ITO, T., JOODMARDI, E., MATTSSON, B., ROUILLARD, C.,
CARTA, M., MURAMATSU, S., SUMI-ICHINOSE, C., NOMURA, T., METZGER,
105
D., CHAMBON, P., LINDQVIST, E., LARSSON, N. G., OLSON, L., BJORKLUND,
A., ICHINOSE, H. & PERLMANN, T. 2009. Nurr1 is required for maintenance of
maturing and adult midbrain dopamine neurons. J Neurosci, 29, 15923-32.
KANG, W., WONG, L. C., SHI, S. H. & HEBERT, J. M. 2009. The transition from radial
glial to intermediate progenitor cell is inhibited by FGF signaling during
corticogenesis. J Neurosci, 29, 14571-80.
KATOH, M. 2006. Cross-talk of WNT and FGF signaling pathways at GSK3beta to
regulate beta-catenin and SNAIL signaling cascades. Cancer Biol Ther, 5, 105964.
KELE, J., SIMPLICIO, N., FERRI, A. L., MIRA, H., GUILLEMOT, F., ARENAS, E. &
ANG, S. L. 2006. Neurogenin 2 is required for the development of ventral
midbrain dopaminergic neurons. Development, 133, 495-505.
KEUKER, J. I., VOLLMANN-HONSDORF, G. K. & FUCHS, E. 2001. How to use the
optical fractionator: an example based on the estimation of neurons in the
hippocampal CA1 and CA3 regions of tree shrews. Brain Res Brain Res Protoc,
7, 211-21.
KHOLODILOV, N., YARYGINA, O., OO, T. F., ZHANG, H., SULZER, D., DAUER, W. &
BURKE, R. E. 2004. Regulation of the development of mesencephalic
dopaminergic systems by the selective expression of glial cell line-derived
neurotrophic factor in their targets. J Neurosci, 24, 3136-46.
KIM, B. G., SHIN, D. H., JEON, G. S., SEO, J. H., KIM, Y. W., JEON, B. S. & CHO, S. S.
2000. Relative sparing of calretinin containing neurons in the substantia nigra of
6-OHDA treated rat Parkinsonian model. Brain Res, 855, 162-5.
KIM, K. S., KIM, C. H., HWANG, D. Y., SEO, H., CHUNG, S., HONG, S. J., LIM, J. K.,
ANDERSON, T. & ISACSON, O. 2003. Orphan nuclear receptor Nurr1 directly
transactivates the promoter activity of the tyrosine hydroxylase gene in a cellspecific manner. J Neurochem, 85, 622-34.
KIRIK, D., GEORGIEVSKA, B., ROSENBLAD, C. & BJORKLUND, A. 2001. Delayed
infusion of GDNF promotes recovery of motor function in the partial lesion model
of Parkinson's disease. Eur J Neurosci, 13, 1589-99.
KLAGSBRUN, M. & BAIRD, A. 1991. A dual receptor system is required for basic
fibroblast growth factor activity. Cell, 67, 229-31.
KLEINER-FISMAN, G., HERZOG, J., FISMAN, D. N., TAMMA, F., LYONS, K. E.,
PAHWA, R., LANG, A. E. & DEUSCHL, G. 2006. Subthalamic nucleus deep brain
stimulation: summary and meta-analysis of outcomes. Mov Disord, 21 Suppl 14,
S290-304.
KLEJBOR, I., MYERS, J. M., HAUSKNECHT, K., CORSO, T. D., GAMBINO, A. S.,
MORYS, J., MAHER, P. A., HARD, R., RICHARDS, J., STACHOWIAK, E. K. &
STACHOWIAK, M. K. 2006. Fibroblast growth factor receptor signaling affects
development and function of dopamine neurons - inhibition results in a
schizophrenia-like syndrome in transgenic mice. J Neurochem, 97, 1243-58.
KOLK, S. M., GUNPUT, R. A., TRAN, T. S., VAN DEN HEUVEL, D. M., PRASAD, A. A.,
HELLEMONS, A. J., ADOLFS, Y., GINTY, D. D., KOLODKIN, A. L., BURBACH,
106
J. P., SMIDT, M. P. & PASTERKAMP, R. J. 2009. Semaphorin 3F is a
bifunctional guidance cue for dopaminergic axons and controls their fasciculation,
channeling, rostral growth, and intracortical targeting. J Neurosci, 29, 12542-57.
KOSODO, Y., ROPER, K., HAUBENSAK, W., MARZESCO, A. M., CORBEIL, D. &
HUTTNER, W. B. 2004. Asymmetric distribution of the apical plasma membrane
during neurogenic divisions of mammalian neuroepithelial cells. EMBO J, 23,
2314-24.
KREITZER, A. C. & MALENKA, R. C. 2008. Striatal plasticity and basal ganglia circuit
function. Neuron, 60, 543-54.
KRIEGLSTEIN, K. 2004. Factors promoting survival of mesencephalic dopaminergic
neurons. Cell Tissue Res, 318, 73-80.
LAEMMLI, U. K. 1970. Cleavage of structural proteins during the assembly of the head
of bacteriophage T4. Nature, 227, 680-5.
LAHTI, L., SAARIMAKI-VIRE, J., RITA, H. & PARTANEN, J. 2011. FGF signaling
gradient maintains symmetrical proliferative divisions of midbrain neuronal
progenitors. Dev Biol, 349, 270-82.
LAW, S. W., CONNEELY, O. M., DEMAYO, F. J. & O'MALLEY, B. W. 1992.
Identification of a new brain-specific transcription factor, NURR1. Mol Endocrinol,
6, 2129-2135.
LEE, J. H. & KIM, K. T. 2004. Induction of cyclin-dependent kinase 5 and its activator
p35 through the extracellular-signal-regulated kinase and protein kinase A
pathways during retinoic-acid mediated neuronal differentiation in human
neuroblastoma SK-N-BE(2)C cells. J Neurochem, 91, 634-47.
LENHARD, T., SCHOBER, A., SUTER-CRAZZOLARA, C. & UNSICKER, K. 2002.
Fibroblast growth factor-2 requires glial-cell-line-derived neurotrophic factor for
exerting its neuroprotective actions on glutamate-lesioned hippocampal neurons.
Mol Cell Neurosci, 20, 181-97.
LI, A. J., SUZUKI, S., SUZUKI, M., MIZUKOSHI, E. & IMAMURA, T. 2002. Fibroblast
growth factor-2 increases functional excitatory synapses on hippocampal
neurons. Eur J Neurosci, 16, 1313-24.
LIN, W., METZAKOPIAN, E., MAVROMATAKIS, Y. E., GAO, N., BALASKAS, N.,
SASAKI, H., BRISCOE, J., WHITSETT, J. A., GOULDING, M., KAESTNER, K. H.
& ANG, S. L. 2009. Foxa1 and Foxa2 function both upstream of and
cooperatively with Lmx1a and Lmx1b in a feedforward loop promoting
mesodiencephalic dopaminergic neuron development. Dev Biol, 333, 386-96.
LINDVALL, O. & KOKAIA, Z. 2009. Prospects of stem cell therapy for replacing
dopamine neurons in Parkinson's disease. Trends Pharmacol Sci, 30, 260-7.
LIU, A. & JOYNER, A. L. 2001. Early anterior/posterior patterning of the midbrain and
cerebellum. Annu Rev Neurosci, 24, 869-96.
LIVAK, K. J. & SCHMITTGEN, T. D. 2001. Analysis of relative gene expression data
using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods,
25, 402-8.
107
MA, X., DANG, X., CLAUS, P., HIRST, C., FANDRICH, R. R., JIN, Y., GROTHE, C.,
KIRSHENBAUM, L. A., CATTINI, P. A. & KARDAMI, E. 2007. Chromatin
compaction and cell death by high molecular weight FGF-2 depend on its nuclear
localization, intracrine ERK activation, and engagement of mitochondria. J Cell
Physiol, 213, 690-8.
MACDONALD, B. T., TAMAI, K. & HE, X. 2009. Wnt/beta-catenin signaling:
components, mechanisms, and diseases. Dev Cell, 17, 9-26.
MAHER, P. A. 1996. Nuclear Translocation of fibroblast growth factor (FGF) receptors in
response to FGF-2. J Cell Biol, 134, 529-36.
MALMERSJO, S., LISTE, I., DYACHOK, O., TENGHOLM, A., ARENAS, E. & UHLEN,
P. 2010. Ca2+ and cAMP signaling in human embryonic stem cell-derived
dopamine neurons. Stem Cells Dev, 19, 1355-64.
MARIC, D., FIORIO PLA, A., CHANG, Y. H. & BARKER, J. L. 2007. Self-Renewing and
Differentiating Properties of Cortical Neural Stem Cells Are Selectively Regulated
by Basic Fibroblast Growth Factor (FGF) Signaling via Specific FGF Receptors. J.
Neurosci., 27, 1836-1852.
MARTINEZ-MORALES, J. R., DEL BENE, F., NICA, G., HAMMERSCHMIDT, M.,
BOVOLENTA, P. & WITTBRODT, J. 2005. Differentiation of the vertebrate retina
is coordinated by an FGF signaling center. Dev Cell, 8, 565-74.
MASON, I. 2007. Initiation to end point: the multiple roles of fibroblast growth factors in
neural development. Nat Rev Neurosci, 8, 583-96.
MAYER, E., FAWCETT, J. W. & DUNNETT, S. B. 1993. Basic fibroblast growth factor
promotes the survival of embryonic ventral mesencephalic dopaminergic
neurons--II. Effects on nigral transplants in vivo. Neuroscience, 56, 389-98.
MCFARLANE, S., MCNEILL, L. & HOLT, C. E. 1995. FGF signaling and target
recognition in the developing Xenopus visual system. Neuron, 15, 1017-28.
MCGEER, E. G., SINGH, E. A. & MCGEER, P. L. 1992. Apparent anterograde transport
of basic fibroblast growth factor in the rat nigrostriatal dopamine system.
Neuroscience Letters, 148, 31-33.
MIKIC, B., ENTWISTLE, R., ROSSMEIER, K. & BIERWERT, L. 2008. Effect of GDF-7
deficiency on tail tendon phenotype in mice. J Orthop Res, 26, 834-9.
MILLER, D. L., ORTEGA, S., BASHAYAN, O., BASCH, R. & BASILICO, C. 2000.
Compensation by fibroblast growth factor 1 (FGF1) does not account for the mild
phenotypic defects observed in FGF2 null mice. Mol Cell Biol, 20, 2260-8.
MIZUTA, I., TSUNODA, T., SATAKE, W., NAKABAYASHI, Y., WATANABE, M.,
TAKEDA, A., HASEGAWA, K., NAKASHIMA, K., YAMAMOTO, M., HATTORI, N.,
MURATA, M. & TODA, T. 2008. Calbindin 1, fibroblast growth factor 20, and synuclein in sporadic Parkinson’s disease. Human Genetics, 124, 89-94.
MORLEY, J. F. & HURTIG, H. I. 2010. Current understanding and management of
Parkinson disease: five new things. Neurology, 75, S9-15.
NAKAMURA, S., ITO, Y., SHIRASAKI, R. & MURAKAMI, F. 2000. Local directional cues
control growth polarity of dopaminergic axons along the rostrocaudal axis. J
Neurosci, 20, 4112-9.
108
NICKEL, W. 2011. The unconventional secretory machinery of fibroblast growth factor 2.
Traffic, 12, 799-805.
NICKEL, W. & RABOUILLE, C. 2009. Mechanisms of regulated unconventional protein
secretion. Nat Rev Mol Cell Biol, 10, 148-55.
NICKEL, W. & SEEDORF, M. 2008. Unconventional mechanisms of protein transport to
the cell surface of eukaryotic cells. Annu Rev Cell Dev Biol, 24, 287-308.
NIKKHAH, G., CUNNINGHAM, M. G., JODICKE, A., KNAPPE, U. & BJORKLUND, A.
1994. Improved graft survival and striatal reinnervation by microtransplantation of
fetal nigral cell suspensions in the rat Parkinson model. Brain Res, 633, 133-43.
NISHIMOTO, S. & NISHIDA, E. 2007. Fibroblast growth factor 13 is essential for neural
differentiation in Xenopus early embryonic development. J Biol Chem, 282,
24255-61.
NISHIO, T., FURUKAWA, S., AKIGUCHI, I. & SUNOHARA, N. 1998. Medial nigral
dopamine neurons have rich neurotrophin support in humans. Neuroreport, 9,
2847-51.
NIWA, H., YAMAMURA, K. & MIYAZAKI, J. 1991. Efficient selection for high-expression
transfectants with a novel eukaryotic vector. Gene, 108, 193-9.
NOBRE, A., KALVE, I., CESNULEVICIUS, K., RAGANCOKOVA, D., RATZKA, A.,
HALFER, N., WESEMANN, M., KRAMPFL, K., CLAUS, P. & GROTHE, C. 2010.
Characterization and differentiation potential of rat ventral mesencephalic
neuronal progenitor cells immortalized with SV40 large T antigen. Cell Tissue
Res, 340, 29-43.
O'KEEFFE, G. C., BARKER, R. A. & CALDWELL, M. A. 2009. Dopaminergic modulation
of neurogenesis in the subventricular zone of the adult brain. Cell Cycle, 8, 288894.
OHMACHI, S., MIKAMI, T., KONISHI, M., MIYAKE, A. & ITOH, N. 2003. Preferential
neurotrophic activity of fibroblast growth factor-20 for dopaminergic neurons
through fibroblast growth factor receptor-1c. Journal of Neuroscience Research,
72, 436-43.
OKANO, H., SAKAGUCHI, M., OHKI, K., SUZUKI, N. & SAWAMOTO, K. 2007.
Regeneration of the central nervous system using endogenous repair
mechanisms. J Neurochem, 102, 1459-65.
OPPENHEIM, R. W. 1991. Cell death during development of the nervous system. Annu
Rev Neurosci, 14, 453-501.
ORNITZ, D. M., XU, J., COLVIN, J. S., MCEWEN, D. G., MACARTHUR, C. A.,
COULIER, F., GAO, G. & GOLDFARB, M. 1996. Receptor specificity of the
fibroblast growth factor family. J Biol Chem, 271, 15292-7.
OZAWA, K., URUNO, T., MIYAKAWA, K., SEO, M. & IMAMURA, T. 1996. Expression of
the fibroblast growth factor family and their receptor family genes during mouse
brain development. Molecular Brain Research, 41, 279-288.
PALMER, T. D., RAY, J. & GAGE, F. H. 1995. FGF-2-responsive neuronal progenitors
reside in proliferative and quiescent regions of the adult rodent brain. Mol Cell
Neurosci, 6, 474-86.
109
PARK, T. H. & MYTILINEOU, C. 1992. Protection from 1-methyl-4-phenylpyridinium
(MPP+) toxicity and stimulation of regrowth of MPP(+)-damaged dopaminergic
fibers by treatment of mesencephalic cultures with EGF and basic FGF. Brain
Res, 599, 83-97.
PARTANEN, J. 2007. FGF signalling pathways in development of the midbrain and
anterior hindbrain. J Neurochem, 101, 1185-93.
PASCUAL, A., HIDALGO-FIGUEROA, M., PIRUAT, J. I., PINTADO, C. O., GOMEZDIAZ, R. & LOPEZ-BARNEO, J. 2008. Absolute requirement of GDNF for adult
catecholaminergic neuron survival. Nat Neurosci, 11, 755-61.
PENG, H., MOFFETT, J., MYERS, J., FANG, X., STACHOWIAK, E. K., MAHER, P.,
KRATZ, E., HINES, J., FLUHARTY, S. J., MIZUKOSHI, E., BLOOM, D. C. &
STACHOWIAK, M. K. 2001. Novel nuclear signaling pathway mediates activation
of fibroblast growth factor-2 gene by type 1 and type 2 angiotensin II receptors.
Mol Biol Cell, 12, 449-62.
PENG, H., MYERS, J., FANG, X., STACHOWIAK, E. K., MAHER, P. A., MARTINS, G.
G., POPESCU, G., BEREZNEY, R. & STACHOWIAK, M. K. 2002. Integrative
nuclear FGFR1 signaling (INFS) pathway mediates activation of the tyrosine
hydroxylase gene by angiotensin II, depolarization and protein kinase C. J
Neurochem, 81, 506-24.
PERLMANN, T. & WALLEN-MACKENZIE, A. 2004. Nurr1, an orphan nuclear receptor
with essential functions in developing dopamine cells. Cell Tissue Res, 318, 4552.
POEWE, W., ANTONINI, A., ZIJLMANS, J. C., BURKHARD, P. R. & VINGERHOETS,
F. 2010. Levodopa in the treatment of Parkinson's disease: an old drug still going
strong. Clin Interv Aging, 5, 229-38.
PRAKASH, N., BRODSKI, C., NASERKE, T., PUELLES, E., GOGOI, R., HALL, A.,
PANHUYSEN, M., ECHEVARRIA, D., SUSSEL, L., WEISENHORN, D. M.,
MARTINEZ, S., ARENAS, E., SIMEONE, A. & WURST, W. 2006. A Wnt1regulated genetic network controls the identity and fate of midbrain-dopaminergic
progenitors in vivo. Development, 133, 89-98.
PRAKASH, N. & WURST, W. 2006. Development of dopaminergic neurons in the
mammalian brain. Cell Mol Life Sci, 63, 187-206.
PRUSZAK, J., JUST, L., ISACSON, O. & NIKKHAH, G. 2009. Isolation and culture of
ventral mesencephalic precursor cells and dopaminergic neurons from rodent
brains. Curr Protoc Stem Cell Biol, Chapter 2, Unit 2D 5.
QUARTO, N., FINGER, F. P. & RIFKIN, D. B. 1991. The NH2-terminal extension of high
molecular weight bFGF is a nuclear targeting signal. J Cell Physiol, 147, 311-8.
RABALLO, R., RHEE, J., LYN-COOK, R., LECKMAN, J. F., SCHWARTZ, M. L. &
VACCARINO, F. M. 2000. Basic fibroblast growth factor (Fgf2) is necessary for
cell proliferation and neurogenesis in the developing cerebral cortex. J Neurosci,
20, 5012-23.
RATZKA A.*, B. O., STACHOWIAK M.K., GROTHE C 2011 submitted. FGF-2 regulates
dopaminergic neuron development in vivo.
110
REILLY, J. F. & MAHER, P. A. 2001. Importin beta-mediated nuclear import of fibroblast
growth factor receptor: role in cell proliferation. J Cell Biol, 152, 1307-12.
REILLY, J. F., MIZUKOSHI, E. & MAHER, P. A. 2004. Ligand dependent and
independent internalization and nuclear translocation of fibroblast growth factor
(FGF) receptor 1. DNA Cell Biol, 23, 538-48.
REUSS, B., DONO, R. & UNSICKER, K. 2003. Functions of fibroblast growth factor
(FGF)-2 and FGF-5 in astroglial differentiation and blood-brain barrier
permeability: evidence from mouse mutants. J Neurosci, 23, 6404-12.
REUSS, B. & VON BOHLEN UND HALBACH, O. 2003. Fibroblast growth factors and
their receptors in the central nervous system. Cell Tissue Res, 313, 139-57.
ROSS, R. A., JUDD, A. B., PICKEL, V. M., JOH, T. H. & REIS, D. J. 1976. Straindependent variations in number of midbrain dopaminergic neurones. Nature, 264,
654-6.
ROUSSA, E., FARKAS, L. M. & KRIEGLSTEIN, K. 2004. TGF-beta promotes survival
on mesencephalic dopaminergic neurons in cooperation with Shh and FGF-8.
Neurobiol Dis, 16, 300-10.
ROUSSA, E. & KRIEGLSTEIN, K. 2004. GDNF promotes neuronal differentiation and
dopaminergic development of mouse mesencephalic neurospheres. Neurosci
Lett, 361, 52-5.
ROUSSA, E., WIEHLE, M., DUNKER, N., BECKER-KATINS, S., OEHLKE, O. &
KRIEGLSTEIN, K. 2006. Transforming growth factor beta is required for
differentiation of mouse mesencephalic progenitors into dopaminergic neurons in
vitro and in vivo: ectopic induction in dorsal mesencephalon. Stem Cells, 24,
2120-9.
RUBENSTEIN, J. L., MARTINEZ, S., SHIMAMURA, K. & PUELLES, L. 1994. The
embryonic vertebrate forebrain: the prosomeric model. Science, 266, 578-80.
SAARIMAKI-VIRE, J., PELTOPURO, P., LAHTI, L., NASERKE, T., BLAK, A. A., VOGT
WEISENHORN, D. M., YU, K., ORNITZ, D. M., WURST, W. & PARTANEN, J.
2007. Fibroblast growth factor receptors cooperate to regulate neural progenitor
properties in the developing midbrain and hindbrain. J Neurosci, 27, 8581-92.
SACCHETTI, P., CARPENTIER, R., SEGARD, P., OLIVE-CREN, C. & LEFEBVRE, P.
2006. Multiple signaling pathways regulate the transcriptional activity of the
orphan nuclear receptor NURR1. Nucleic Acids Res, 34, 5515-27.
SASAKI, H., HUI, C., NAKAFUKU, M. & KONDOH, H. 1997. A binding site for Gli
proteins is essential for HNF-3beta floor plate enhancer activity in transgenics
and can respond to Shh in vitro. Development, 124, 1313-22.
SATO, T., SHIMAZAKI, T., NAKA, H., FUKAMI, S., SATOH, Y., OKANO, H., LAX, I.,
SCHLESSINGER, J. & GOTOH, N. 2010. FRS2alpha regulates Erk levels to
control a self-renewal target Hes1 and proliferation of FGF-responsive neural
stem/progenitor cells. Stem Cells, 28, 1661-73.
SAUCEDO-CARDENAS, O. & CONNEELY, O. M. 1996. Comparative distribution of
NURR1 and NUR77 nuclear receptors in the mouse central nervous system. J
Mol Neurosci, 7, 51-63.
111
SAUCEDO-CARDENAS, O., QUINTANA-HAU, J. D., LE, W.-D., SMIDT, M. P., COX, J.
J., MAYO, F., BURBACH, J. P. H. & CONNEELY, O. M. 1998. Nurr1 is essential
for the induction of the dopaminergic phenotype and the survival of ventral
mesencephalic late dopaminergic precursor neurons. Proceedings of the National
Academy of Sciences of the United States of America, 95, 4013-4018.
SAVICA, R., ROCCA, W. A. & AHLSKOG, J. E. 2010. When does Parkinson disease
start? Arch Neurol, 67, 798-801.
SCHAPIRA, A. H. 2008. Progress in neuroprotection in Parkinson's disease. Eur J
Neurol, 15 Suppl 1, 5-13.
SCHAPIRA, A. H. 2009. Neuroprotection in Parkinson's disease. Parkinsonism Relat
Disord, 15 Suppl 4, S41-3.
SHANMUGALINGAM, S., HOUART, C., PICKER, A., REIFERS, F., MACDONALD, R.,
BARTH, A., GRIFFIN, K., BRAND, M. & WILSON, S. W. 2000. Ace/Fgf8 is
required for forebrain commissure formation and patterning of the telencephalon.
Development, 127, 2549-61.
SHIRASAKI, R., LEWCOCK, J. W., LETTIERI, K. & PFAFF, S. L. 2006. FGF as a targetderived chemoattractant for developing motor axons genetically programmed by
the LIM code. Neuron, 50, 841-53.
SHULMAN, J. M., DE JAGER, P. L. & FEANY, M. B. 2011. Parkinson's disease:
genetics and pathogenesis. Annu Rev Pathol, 6, 193-222.
SIEGEL, G. J. & CHAUHAN, N. B. 2000. Neurotrophic factors in Alzheimer's and
Parkinson's disease brain. Brain Res Brain Res Rev, 33, 199-227.
SMIDT, M. P. & BURBACH, J. P. 2007. How to make a mesodiencephalic dopaminergic
neuron. Nat Rev Neurosci, 8, 21-32.
SMIDT, M. P. & BURBACH, J. P. 2009. Terminal differentiation ofmesodiencephalic
dopaminergic neurons: the role of Nurr1 and Pitx3. Adv Exp Med Biol, 651, 4757.
STACHOWIAK, E. K., FANG, X., MYERS, J., DUNHAM, S. & STACHOWIAK, M. K.
2003. cAMP-induced differentiation of human neuronal progenitor cells is
mediated by nuclear fibroblast growth factor receptor-1 (FGFR1). Journal of
Neurochemistry, 84, 1296-312.
STACHOWIAK, M. K., MAHER, P. A., JOY, A., MORDECHAI, E. & STACHOWIAK, E.
K. 1996. Nuclear accumulation of fibroblast growth factor receptors is regulated
by multiple signals in adrenal medullary cells. Mol Biol Cell, 7, 1299-317.
STACHOWIAK, M. K., MAHER, P. A. & STACHOWIAK, E. K. 2007. Integrative nuclear
signaling in cell development--a role for FGF receptor-1. DNA Cell Biol, 26, 81126.
STACHOWIAK, M. K., STACHOWIAK, E. K., ALETTA, J. M. & TZANAKAKIS, E. S.
2011. A common integrative nuclear signaling module for stem cell development.
In: STACHOWIAK, M. K. & TZANAKAKIS, M. (eds.) Stem Cells: From
Mechanisms to Technologies. World Scientific Pub Co Inc. .
STOPPINI, L., BUCHS, P. A. & MULLER, D. 1991. A simple method for organotypic
cultures of nervous tissue. J Neurosci Methods, 37, 173-82.
112
STUDER, L., TABAR, V. & MCKAY, R. D. 1998. Transplantation of expanded
mesencephalic precursors leads to recovery in parkinsonian rats. Nat Neurosci,
1, 290-5.
TAKAYAMA, H., RAY, J., RAYMON, H. K., BAIRD, A., HOGG, J., FISHER, L. J. &
GAGE, F. H. 1995. Basic fibroblast growth factor increases dopaminergic graft
survival and function in a rat model of Parkinson's disease. Nat Med, 1, 53-8.
THOMAS, B. & BEAL, M. F. 2007. Parkinson's disease. Hum Mol Genet, 16 Spec No. 2,
R183-94.
THOMPSON, L., BARRAUD, P., ANDERSSON, E., KIRIK, D. & BJORKLUND, A. 2005.
Identification of dopaminergic neurons of nigral and ventral tegmental area
subtypes in grafts of fetal ventral mesencephalon based on cell morphology,
protein expression, and efferent projections. J Neurosci, 25, 6467-77.
TIMMER, CESNULEVICIUS, K., WINKLER, C., KOLB, J., LIPOKATIC-TAKACS, E.,
JUNGNICKEL, J. & GROTHE, C. 2007. Fibroblast growth factor (FGF)-2 and
FGF receptor 3 are required for the development of the substantia nigra, and
FGF-2 plays a crucial role for the rescue of dopaminergic neurons after 6hydroxydopamine lesion. J Neurosci, 27, 459-71.
TIMMER, M., GROSSKREUTZ, J., SCHLESINGER, F., KRAMPFL, K., WESEMANN,
M., JUST, L., BUFLER, J. & GROTHE, C. 2006. Dopaminergic properties and
function after grafting of attached neural precursor cultures. Neurobiology of
disease, 21, 587-606.
TIMMER, M., MULLER-OSTERMEYER, F., KLOTH, V., WINKLER, C., GROTHE, C. &
NIKKHAH, G. 2004. Enhanced survival, reinnervation, and functional recovery of
intrastriatal dopamine grafts co-transplanted with Schwann cells overexpressing
high molecular weight FGF-2 isoforms. Exp Neurol, 187, 118-36.
TOOYAMA, I., KAWAMATA, T., WALKER, D., YAMADA, T., HANAI, K., KIMURA, H.,
IWANE, M., IGARASHI, K., MCGEER, E. G. & MCGEER, P. L. 1993. Loss of
basic fibroblast growth factor in substantia nigra neurons in Parkinson's disease.
Neurology, 43, 372-6.
TOOYAMA, I., MCGEER, E. G., KAWAMATA, T., KIMURA, H. & MCGEER, P. L. 1994.
Retention of basic fibroblast growth factor immunoreactivity in dopaminergic
neurons of the substantia nigra during normal aging in humans contrasts with
loss in Parkinson's disease. Brain Res, 656, 165-8.
TOOYAMA, I., WALKER, D., YAMADA, T., HANAI, K., KIMURA, H., MCGEER, E. G. &
MCGEER, P. L. 1992. High molecular weight basic fibroblast growth factor-like
protein is localized to a subpopulation of mesencephalic dopaminergic neurons in
the rat brain. Brain Research, 593, 274-280.
TORRES, M. A., ELDAR-FINKELMAN, H., KREBS, E. G. & MOON, R. T. 1999.
Regulation of ribosomal S6 protein kinase-p90(rsk), glycogen synthase kinase 3,
and beta-catenin in early Xenopus development. Mol Cell Biol, 19, 1427-37.
TREMBLAY, R. G., SIKORSKA, M., SANDHU, J. K., LANTHIER, P., RIBECCOLUTKIEWICZ, M. & BANI-YAGHOUB, M. 2010. Differentiation of mouse Neuro
2A cells into dopamine neurons. J Neurosci Methods, 186, 60-7.
113
UMEMORI, H. 2009. Weaving the neuronal net with target-derived fibroblast growth
factors. Dev Growth Differ, 51, 263-70.
VACCARINO, F. M., SCHWARTZ, M. L., RABALLO, R., NILSEN, J., RHEE, J., ZHOU,
M., DOETSCHMAN, T., COFFIN, J. D., WYLAND, J. J. & HUNG, Y. T. 1999.
Changes in cerebral cortex size are governed by fibroblast growth factor during
embryogenesis. Nat Neurosci, 2, 246-53.
VADASZ, C., SMILEY, J. F., FIGARSKY, K., SAITO, M., TOTH, R., GYETVAI, B. M.,
OROS, M., KOVACS, K. K., MOHAN, P. & WANG, R. 2007. Mesencephalic
dopamine neuron number and tyrosine hydroxylase content: Genetic control and
candidate genes. Neuroscience, 149, 561-72.
VICARIO-ABEJON, C., JOHE, K. K., HAZEL, T. G., COLLAZO, D. & MCKAY, R. D.
1995. Functions of basic fibroblast growth factor and neurotrophins in the
differentiation of hippocampal neurons. Neuron, 15, 105-14.
VIEIRA, C., POMBERO, A., GARCIA-LOPEZ, R., GIMENO, L., ECHEVARRIA, D. &
MARTINEZ, S. 2010. Molecular mechanisms controlling brain development: an
overview of neuroepithelial secondary organizers. Int J Dev Biol, 54, 7-20.
VOLAKAKIS, N., JOODMARDI, E. & PERLMANN, T. 2009. NR4A orphan nuclear
receptors influence retinoic acid and docosahexaenoic acid signaling via upregulation of fatty acid binding protein 5. Biochem Biophys Res Commun, 390,
1186-91.
VON BOHLEN UND HALBACH, O., MINICHIELLO, L. & UNSICKER, K. 2005.
Haploinsufficiency for trkB and trkC receptors induces cell loss and accumulation
of alpha-synuclein in the substantia nigra. FASEB J, 19, 1740-2.
VOORN, P., KALSBEEK, A., JORRITSMA-BYHAM, B. & GROENEWEGEN, H. J. 1988.
The pre- and postnatal development of the dopaminergic cell groups in the
ventral mesencephalon and the dopaminergic innervation of the striatum of the
rat. Neuroscience, 25, 857-87.
WALLEN, A. & PERLMANN, T. 2003. Transcriptional control of dopamine neuron
development. Ann N Y Acad Sci, 991, 48-60.
WALLEN, A., ZETTERSTROM, R. H., SOLOMIN, L., ARVIDSSON, M., OLSON, L. &
PERLMANN, T. 1999. Fate of mesencephalic AHD2-expressing dopamine
progenitor cells in NURR1 mutant mice. Exp Cell Res, 253, 737-46.
WANG, Z., BENOIT, G., LIU, J., PRASAD, S., AARNISALO, P., LIU, X., XU, H.,
WALKER, N. P. & PERLMANN, T. 2003. Structure and function of Nurr1 identifies
a class of ligand-independent nuclear receptors. Nature, 423, 555-60.
WEISE, B., JANET, T. & GROTHE, C. 1993. Localization of bFGF and FGF-receptor in
the developing nervous system of the embryonic and newborn rat. J Neurosci
Res, 34, 442-53.
WEST, M. J. 1993. New stereological methods for counting neurons. Neurobiol Aging,
14, 275-85.
WEST, M. J., SLOMIANKA, L. & GUNDERSEN, H. J. 1991. Unbiased stereological
estimation of the total number of neurons in thesubdivisions of the rat
hippocampus using the optical fractionator. Anat Rec, 231, 482-97.
114
WU, Y., LE, W. & JANKOVIC, J. 2011. Preclinical biomarkers of Parkinson disease.
Arch Neurol, 68, 22-30.
XIAO, Q., CASTILLO, S. O. & NIKODEM, V. M. 1996. Distribution of messenger RNAs
for the orphan nuclear receptors Nurr1 and Nur77 (NGFI-B) in adult rat brain
using in situ hybridization. Neuroscience, 75, 221-30.
YAGAMI, T., TAKASE, K., YAMAMOTO, Y., UEDA, K., TAKASU, N., OKAMURA, N.,
SAKAEDA, T. & FUJIMOTO, M. 2010. Fibroblast growth factor 2 induces
apoptosis in the early primary culture of rat cortical neurons. Exp Cell Res, 316,
2278-90.
YAMAUCHI, K., MIZUSHIMA, S., TAMADA, A., YAMAMOTO, N., TAKASHIMA, S. &
MURAKAMI, F. 2009. FGF8 signaling regulates growth of midbrain dopaminergic
axons by inducing semaphorin 3F. J Neurosci, 29, 4044-55.
YAYON, A., KLAGSBRUN, M., ESKO, J. D., LEDER, P. & ORNITZ, D. M. 1991. Cell
surface, heparin-like molecules are required for binding of basic fibroblast growth
factor to its high affinity receptor. Cell, 64, 841-8.
YE, W., SHIMAMURA, K., RUBENSTEIN, J. L., HYNES, M. A. & ROSENTHAL, A.
1998. FGF and Shh signals control dopaminergic and serotonergic cell fate in the
anterior neural plate. Cell, 93, 755-66.
ZECHEL, S., JAROSIK, J., KIPRIANOVA, I., SCHOBER, A., UNSICKER, K. & VON
BOHLEN UND HALBACH, O. 2006. FGF-2 deficiency does not alter vulnerability
of the dopaminergic nigrostriatal system towards MPTP intoxication in mice. Eur J
Neurosci, 23, 1671-5.
ZECHEL, S., UNSICKER, K. & VON BOHLEN UND HALBACH, O. 2009. Fibroblast
growth factor-2 deficiency affects hippocampal spine morphology, but not
hippocampal catecholaminergic or cholinergic innervation. Dev Dyn, 238, 343-50.
ZETTERSTROM, R. H., SOLOMIN, L., JANSSON, L., HOFFER, B. J., OLSON, L. &
PERLMANN, T. 1997. Dopamine neuron agenesis in Nurr1-deficient mice.
Science, 276, 248-50.
ZETTERSTROM, R. H., SOLOMIN, L., MITSIADIS, T., OLSON, L. & PERLMANN, T.
1996. Retinoid X receptor heterodimerization and developmental expression
distinguish the orphan nuclear receptors NGFI-B, Nurr1, and Nor1. Mol
Endocrinol, 10, 1656-66.
ZHANG, T., JIA, N., FEI, E., WANG, P., LIAO, Z., DING, L., YAN, M., NUKINA, N.,
ZHOU, J. & WANG, G. 2007. Nurr1 is phosphorylated by ERK2 in vitro and its
phosphorylation upregulates tyrosine hydroxylase expression in SH-SY5Y cells.
Neurosci Lett, 423, 118-22.
ZHANG, X., IBRAHIMI, O. A., OLSEN, S. K., UMEMORI, H., MOHAMMADI, M. &
ORNITZ, D. M. 2006. Receptor specificity of the fibroblast growth factor family.
The complete mammalian FGF family. J Biol Chem, 281, 15694-700.
115
6. Acknowledgement/ Danksagung
Ich danke insbesondere Frau Professor Grothe für die Bereitstellung des sehr
interessanten Themas, sowie für die stetige Unterstützung und Anregungen.
Ich danke auch meinen Co-Supervisoren Herrn Professor Dengler und Herrn Professor
Lanfermann für die Betreuung meiner Doktorarbeit, sowie Frau Professor Brandt-Saberi
für die Begutachtung.
Weiterhin danke ich Herrn Dr. Ratzka für die Einarbeitung und Betreuung im Labor.
Mein Besonderer Dank gilt Kerstin Kuhlemann, Natascha Heidrich, Silke Fischer und
Hildegard Streich für ihre exzellente technische Unterstützung und für die sehr
angenehme Zusammenarbeit.
Ebenso bedanke ich mich bei Hella Brinkmann und Maike Wesemann für die immer
passenden Lösungsvorschläge, für ihre Hilfsbereitschaft und für das offene Ohr zu jeder
Zeit. Ich möchte mich insgesamt bei allen Mitarbeitern im Institut für die kollegiale
Atmosphäre bedanken.
Ich danke auch Herrn Professor Brinker für die Bereitstellung des Stereologischen SetUps, Herrn Professor Nikkhah, Herrn Dr. Döbrössy und Fabian Büchele aus Freiburg für
die Einführung in die Stereologie. Nicht zuletzt danke ich auch Herrn Professor
Stachowiak und seinen Kollegen, sowie Herrn Professor Claus und Benjamin
Förthmann für eine wunderbare Zusammenarbeit und Kooperation.
Ein großer Dank gilt Andre Nobre, Ieva Kalve und Miriam Schiff für Ihre Hilfe und
Unterstützung insbesondere am Anfang meiner Doktoranden-Zeit im Wissenschaftlichen
sowie Privaten. Ich bedanke mich außerdem bei allen Doktoranden, PhD-, Diplom-,
Master- und Bachelor-Studenten für einen anregenden Ideenaustausch während
unserer unterhaltsamen Kaffeepausen!
Ich danke den Koordinatoren des PhD-Programmes Dagmar Esser, Nadja Borsum und
Beatrice Grummer für die administrative Unterstützung und ihre stetige Hilfsbereitschaft.
Ganz besonders danke ich meiner Familie und auch meinen Freunden für die Liebe,
Geduld, Unterstützung aus der Ferne und die Zeit, die sie mir geschenkt haben, die ich
für das Fertigstellen dieser Arbeit gebraucht habe.
116
7. Curriculum vitae
Personal Information
Name:
Address:
Date/ place of birth:
Citizenship:
Olga Baron
Nordring 1
30163 Hannover
Germany
Tel.: 0176-22844403
[email protected]
21.06.1981 in Shdanowka (Russia)
German
Education
[ 1988, September]
1. class
Enrollment on elementary school in Nizshnjaja Pjosha (USSR) near NaryanMar
[ 1989, February]
2. - 7. class
Continuation on elementary school/ transfer to secondary school in NaryanMar (Russia)
[ 1994, August ]
7. - 8. class
Immigration to Federal Republic of Germany/ Integration at the LeibnizGymnasium in Dormagen-Hackenbroich (Germany)
[ 1995, March –
2001, June]
8. - 10. class, Abitur
Graduation with higher education entrance qualification at Gymnasium
Leopoldinum in Detmold (Germany)
Academic Studies
[2001, October 2008, February]
Study of biological sciences at the University of Rostock (Germany)
[ 2007, December]
Graduation with Diploma degree in biology, diploma thesis at the Institute for Cell
Biology and Biosystems Technology (Chair in Animal Physiology Prof. Dieter G.
Weiss), University of Rostock
[Since 2008, October]
Enrolled in the international Ph.D.-Program at the Center for Systemic
Neurosciences, Hannover
[Since 2008, October]
Doctorate in Neurosciences at the Institute of Neuroanatomy (Chair: Prof. Claudia
Grothe), Hannover Medical School
117
Additional activities during studies
[2004, August]
Student exchange program between University of Rostock and Moscow
Lomonossow University – visit of “Nikolaj Pertzov” biostation in the White Sea
[ 2005, 2007]
Student assistant – „nourishment analysis of cormorant“, Institute of Zoology at
the University of Rostock (Germany)
[ 2006, August December]
Internship at BIONAS GmbH, the laboratory for real-time cell monitoring
systems (Rostock, Germany)
[ 2006, October December]
Student assistant – „Tycho - Cell-Material-Dialogue“, a collaboration project of the
Institute for Physics (University of Rostock) and Institute for Cell Biology and
Biosystems Technology (chair in Animal Physiology, University of Rostock)
[2009-2010]
Assistance in civil educational program “ Patient University” at Hannover Medical
School
[2010, June]
Organization and realization of “Neuroscience Meets School 2010”, a workshop of
the Center for Systemic Neurosciences for schools
Language skills
German, English, Russian, Latin
118
Affidavit
I, Olga Baron, herewith declare that I autonomously carried out the doctoral thesis
entitled “On role of basic fibroblast growth factor (FGF-2) during development of
mesencephalic dopaminergic neurons of substantia nigra in mice”.
Following third party assistance has been used for help with:
Genotyping: Natascha Heidrich and Silke Fischer
Immunohistochemistry: Kerstin Kuhlemann, Natascha Heidrich and Silke Fischer
Western blot: Kerstin Kuhlemann
qRT-PCR: Dr. Andreas Ratzka and Kerstin Kuhlemann
I did not receive any assistance in return for payment by consulting agencies or any
other person. No one received any kind of payment for direct or indirect assistance in
correlation to the content of the submitted thesis.
I conducted the project at the following institution: Institute of Neuroanatomy at
Hannover Medical School
The thesis has not been submitted elsewhere for an exam, as thesis or for evaluation in
a similar context.
I hereby affirm the above statements to be complete and true to the best of my
knowledge.
_______________________________