Epidermal Hyperplasia and Appendage Abnormalities in Mice

1
Epidermal Hyperplasia and Appendage Abnormalities in Mice
2
Lacking CD109
3
Shinji Mii,1 Yoshiki Murakumo,1,* Naoya Asai,2 Mayumi Jijiwa,1 Sumitaka Hagiwara3,
4
Takuya Kato,1 Masato Asai,1 Atsushi Enomoto,1 Kaori Ushida,1 Sayaka Sobue,4
5
Masatoshi Ichihara,4 and Masahide Takahashi1,2,*
6
1
Department of Pathology, Nagoya University Graduate School of Medicine, Nagoya, Japan.
7
2
Division of Molecular Pathology, Center for Neurological Disease and Cancer, Nagoya
8
University Graduate School of Medicine, Nagoya, Japan.
9
3
Department of Oral and Maxillofacial Surgery, Nagoya University Graduate School of Medicine,
10
Nagoya, Japan.
11
4
12
Kasugai, Japan.
13
*
14
Graduate School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya 466-8550, Japan. Tel: +81
15
52 744 2093; Fax: +81 52 744 2098; E-mail: [email protected];
16
[email protected]
17
Running title: Skin abnormalities in mice lacking CD109
18
This work was supported by Grants-in-Aid for Global Center of Excellence (GCOE) research,
19
Scientific Research (A) commissioned by the Ministry of Education, Culture, Sports, Science and
Department of Biomedical Sciences, College of Life and Health Sciences, Chubu University,
Correspondence: Y Murakumo and M Takahashi, Department of Pathology, Nagoya University
1
1
Technology (MEXT) of Japan (to MT), by Scientific Research (C) commissioned by MEXT of
2
Japan (to YM), and by the Toyoaki Scholarship Foundation (to YM).
3
Number of text pages: 36
4
Number of tables: 1
5
Number of figures: 5 + 5 supplemental figures
6
Conflict of Interest: The authors declare no conflict of interest.
2
1
2
Abstract
CD109 is a glycosylphosphatidylinositol-anchored glycoprotein that is highly expressed in
3
several types of human cancer tissues, particularly squamous cell carcinomas. In normal human
4
tissues, human CD109 expression is limited to certain cell types, including myoepithelial cells of
5
the mammary, lacrimal, salivary and bronchial glands and basal cells of the prostate and
6
bronchial epithelia. While CD109 is reported to negatively regulate TGF-β signaling in
7
keratinocytes in vitro, its physiological role in vivo remains largely unknown. To investigate the
8
function of CD109 in vivo, we generated CD109–deficient (CD109–/–) mice. Although CD109–/–
9
mice were born normally, transient impairment of hair growth was observed. Histologically,
10
kinked hair shafts, ectatic hair follicles with an accumulation of sebum, and persistent
11
hyperplasia of the epidermis and sebaceous glands were observed in CD109–/– mice.
12
Immunohistochemical analysis revealed thickening of the basal / suprabasal layer in the
13
epidermis of CD109–/– mice, which is where endogenous CD109 is expressed in wild-type mice.
14
Although CD109 was reported to negatively regulate TGF-β signaling, no significant difference in
15
levels of Smad2 phosphorylation was observed in the epidermis between wild-type and CD109–/–
16
mice. Instead, Stat3 phosphorylation levels were significantly elevated in the epidermis of
17
CD109–/– mice compared with wild-type mice. These results suggest that CD109 regulates
18
differentiation of keratinocytes via a signaling pathway involving Stat3.
3
1
2
Introduction
CD109, a glycosylphosphatidylinositol (GPI)-anchored cell surface glycoprotein, is a
3
member of the α2-macroglobulin/C3, C4, C5 family of thioester-containing proteins.1–4 CD109 is a
4
cell surface antigen expressed on a subset of fetal and adult CD34+ bone marrow mononuclear
5
cells, phytohemagglutinin-activated T lymphoblasts, thrombin-activated platelets, leukemic
6
megakaryoblasts, endothelial cells and mesenchymal stem cell subsets.5–7 In addition, we have
7
reported that human CD109 is expressed in a limited number of cell types in normal tissues such
8
as myoepithelial cells of the mammary, lacrimal, salivary and bronchial glands and basal cells of
9
the prostate and bronchial epithelia.8–12 High levels of CD109 expression were also detected in
10
various tumor cell lines and in tumor tissues including squamous cell carcinomas (SCCs) of the
11
lung, esophagus, uterus and oral cavity, malignant melanoma of the skin, and urothelial
12
carcinoma of the urinary bladder.8–16 CD109 expression was significantly higher in
13
well-differentiated SCCs of the oral cavity and in low-grade urothelial carcinomas of the urinary
14
bladder than in moderately- or poorly-differentiated SCCs and in high-grade urothelial
15
carcinomas, respectively.14,16 These findings suggest that CD109 expression is strictly controlled
16
in normal tissues and is associated with tumor development.
17
Signals through the transforming growth factor (TGF)-β receptor system induce a wide
18
range of biological responses including cell proliferation, differentiation, migration and apoptosis,
19
tissue remodeling, immune response and angiogenesis.17–19 Ligand-mediated assembly of
4
1
TGF-β receptors I and II (TGFBRI and TGFBRII, respectively) initiates an intracellular
2
phosphorylation cascade; activated TGFBRII transphosphorylates TGFBRI, which subsequently
3
phosphorylates receptor-regulated Smads (R-Smads such as Smad2/3), which allows the
4
R-Smads to bind a common mediator, Smad4. R-Smad/Smad4 complexes accumulate in the
5
nucleus where they act as transcription factors for target genes.20–22 TGF-β-mediated receptor
6
activation also induces inhibitory Smads (I-Smads such as Smad7), which compete with
7
R-Smads in binding to activated TGFBRI, thus negatively regulating signals.23
8
Reportedly, CD109 functions as a negative regulator of TGF-β signaling in human
9
keratinocytes; CD109, as a component of the TGF-β receptor system, inhibits activation of
10
R-Smad, probably by direct modulation of receptor activity.24,25 Our recent study using cultured
11
cells showed CD109 to be cleaved by furin, generating 180- and 25-kDa fragments; the 180 kDa
12
fragment is partially secreted into the medium.26 The negative effect of CD109 on TGF-β
13
signaling requires this furin-mediated cleavage of CD109, as the resulting 180 kDa fragment is
14
responsible for this effect.26 CD109 is also reported to associate with caveolin-1, a major
15
component of caveolae, and to promote localization of TGFBRs into caveolar compartments to
16
facilitate their degradation.27 While these findings show the importance of CD109 in the TGF-β
17
signaling pathway in vitro, its physiological functions in vivo have not yet been elucidated.
18
Recombinant CD109 is reported to regulate signal transducers and activators of
19
transcription (STAT)3 activation in human keratinocytes.28 Various cytokines and growth factors,
5
1
such as IL-6, IL-20, IL-22 and EGF, activate Stat3 in keratinocytes,29,30 whereas TGF-β
2
suppresses Stat3 activation through IL-6 in epithelial cells.31,32 Stat3 is critical to such biological
3
activities as cell proliferation, differentiation, migration, survival, and oncogenesis.29,30
4
In this study, we generated CD109-deficient mice (CD109–/– mice) to investigate the
5
physiological roles of CD109 in vivo. The CD109–/– mice showed transient impairment of hair
6
growth, accompanied by kinked hair shafts, ectatic hair follicles with an accumulation of sebum,
7
and persistent hyperplasia of the epidermis and sebaceous glands. These findings suggest that
8
CD109 plays a role in the normal development of skin.
9
Materials and Methods
10
11
Targeting Construct
Construction of the targeting vector started by modifying a pBlueScript II KS vector (Agilent
12
Technologies, Santa Clara, CA) containing a LacZ reporter with a mouse nuclear localization
13
signal (NLS) upstream of a phosphoglycerine kinase (PGK) promoter driving a neomycin (neo)
14
selection marker. The CD109 5′ homology arm (1666 bp) was generated by PCR with PfuUltra™
15
High-Fidelity DNA polymerase (Agilent Technologies) using genomic DNA of the 129S6 mouse
16
strain as a template and inserted into the EcoRV site of the vector. The CD109 3′ homology arm
17
(4893 bp) of intron 2 was inserted into the NotI site. This targeting vector (Figure 1A: middle) for
18
the CD109 locus (Figure 1A: top) was designed to insert a lacZ-PGK-neo cassette 17 amino
19
acids downstream from the start methionine, resulting in disruption of the remainder of the first
6
1
coding exon and the whole of the second exon. Structure of the targeted CD109 allele is shown
2
(Figure 1A: bottom). The final targeting vector was confirmed by DNA sequencing and restriction
3
mapping.
4
Generation of CD109 Knockout / lacZ Knock-in Mice
5
The targeting vector was linearized by digestion with KpnI, and introduced by
6
electroporation into embryonic stem (ES) cells derived from 129S6 mice. After G418/diphtheria
7
toxin A positive-negative selection, 10 ES clones with successful homologous recombination
8
were isolated by Southern blot screening of SpeI-digested genomic DNAs with a 5′ probe
9
(Figure 1A). One clone was injected into C57BL/6J blastocysts; chimeric mice were generated
10
by PhoenixBio (Higashihiroshima, Japan). Genetic background of the mice used in this study
11
was C57BL6J/129S6. All mice were housed in polycarbonate cages containing hardwood chip
12
bedding at 25°C on a 12-h light/dark cycle. All animal protocols were approved by the Animal
13
Care and Use Committee of Nagoya University Graduate School of Medicine (Approval ID
14
number: 23121).
15
Genotyping of Mice after Germ-line Transmission
16
Genomic DNAs from offspring were extracted from their tails. Genotyping of mice was
17
performed through PCR, based on four primers:
18
primer P1 (forward): 5′-GTCCCGCTTTCTGGTGACAG-3′;
19
primer P2 (reverse): 5′-GTGTGACTGTTAGACAGTGCAG-3′;
7
1
primer P3 (forward): 5′-CCATCGCCATCTGCTGCACG-3′; and
2
primer P4 (reverse): 5′-ACGATCCTGAGACTTCCACAC-3′ (Figure 1A). The PCR with rTaq
3
polymerase (Takara Bio Inc., Ohtsu, Japan) was performed as follows: 96°C for 2 min; 32 cycles
4
of 94°C for 30 s, 65°C for 30 s, and 72°C for 30 s. The expected PCR product size from wild-type
5
and targeted alleles were 205 and 603 bp, respectively.
6
Tissue Preparation
7
After body weight measurement, mice were sacrificed under general anesthesia. Complete
8
autopsies were performed and resected organs were cut into 5-mm3 specimens and quickly
9
frozen for protein extraction.
10
Antibodies
11
Primary antibodies used in this study include anti-CD109 and anti-Smad7 monoclonal or
12
polyclonal antibodies (Santa Cruz Biotechnology, Santa Cruz, CA), anti-β-actin and anti-p63
13
monoclonal antibodies (Sigma-Aldrich, St. Louis, MO), anti-CD3 polyclonal antibody (Dako,
14
Glostrup, Denmark), anti-CD45R and anti-Gr-1 monoclonal antibodies (eBioscience, San Diego,
15
CA), anti-CK10, anti-CK14 and anti-filaggrin polyclonal antibodies (Covance Inc, Princeton, NJ),
16
anti-BrdU monoclonal antibody (BD Biosciences, San Jose, CA), and anti-cleaved caspase-3,
17
anti-phospho-Smad2, anti-Smad2, anti-phospho-Stat3 and anti-Stat3 monoclonal or polyclonal
18
antibodies (Cell Signaling Technology, Danvers, MA). Alexa Fluor 488-conjugated anti-rabbit
19
IgG secondary antibody was purchased from Invitrogen (Carlsbad, CA) and horseradish
8
1
peroxidase (HRP)-conjugated anti-rabbit IgG secondary antibody was purchased from Dako.
2
(Table 1.)
3
Western Blot Analysis
4
Frozen mouse tissues were homogenized in sodium dodecyl sulfate (SDS) sample buffer
5
(62.5 mM Tris-HCl, pH 6.8, 2% SDS, 25% glycerol, 20 µg/ml bromophenol blue) and sonicated
6
until no longer viscous. After measuring protein concentration using the DC protein Assay Kit
7
(Bio-Rad Laboratories, Hercules, CA), the lysates were boiled at 100°C for 2 min in the presence
8
of 2% β-mercaptoethanol. The lysates, containing 40 µg of proteins, were subjected to
9
SDS-polyacrylamide gel electrophoresis and transferred to polyvinylidene fluoride membranes
10
(Millipore Corporation, Bedford, MA). Membranes were blocked for 1 h at room temperature (RT)
11
in Blocking One (Nacalai Tesque, Kyoto, Japan) with gentle agitation and incubated with the
12
primary antibody for 1 h at RT. After washing the membranes three times with TBST buffer (20
13
mM Tris-HCl, pH7.6, 137 mM NaCl, 0.1% Tween 20), they were incubated with secondary
14
antibody conjugated to HRP for 1 h at RT. After washing the membranes, the reaction was
15
visualized by the ECL Detection Kit (GE Healthcare, Buckinghamshire, UK) according to
16
manufacturer’s instructions.
17
Histological Analysis
18
19
The major organs were resected as described above. All tissues were fixed in 10%
neutral-buffered formalin, dehydrated and embedded in paraffin. Sections 4-µm thick were
9
1
prepared for hematoxylin and eosin (H-E) staining (which was performed by conventional
2
methods) and immunohistochemistry. Epidermal thickness of the dorsal skin was measured from
3
the basal lamina to the lower border of the stratum corneum using WinROOF software (Mitani
4
Corporation, Fukui, Japan). Lipid accumulation was visualized by Oil-red-O (Sigma-Aldrich)
5
staining on frozen sections of formalin-fixed skin tissues.
6
In situ Hybridization
7
After mice were anesthetized and perfused intravascularly with 4% (w/v) paraformaldehyde
8
solution, organs were resected and frozen in 2-methylbutane cooled in liquid nitrogen. Frozen
9
sections 10 µm-thick were prepared using a cryostat (Leica Microsystems, Wetzlar, Germany).
10
CD109-specific PCR products of ~400 bp with SP6 and T7 promoter fragments were
11
generated with primers 5′-ccaagctATTTAGGTGACACTATAGAgaagtgaaccttctcagtggc-3′ and
12
5′-tgaattgTAATACGACTCACTATAGGGgcacaaagtacagaaggacgg-3′ (SP6 and T7 promoter
13
sequences are in uppercase). The PCR products were gel-purified and transcribed in vitro using
14
SP6 or T7 RNA polymerase (Roche Applied Science, Penzberg, Germany) incorporating
15
33
16
control; T7 RNA polymerase generated the anti-sense probe. The probes were subsequently
17
DNase-treated and purified using the RNeasy Mini Kit (QIAGEN, Hilden, Germany). Slides were
18
hybridized overnight at 60°C using 200 µl per slide of hybridization solution consisting of 1 × 106
19
counts per minute (cpm) radiolabeled probe and hybridization buffer (50% formamide, 10%
P-UTP (PerkinElmer, Waltham, MA). SP6 RNA polymerase generated the sense probe as a
10
1
dextran sulfate, 0.5 M NaCl, 1 × Denhardt's, 10 mM Tris, pH 8.0, 1 mM EDTA, 500 µg/ml yeast
2
tRNA, and 10 mM DTT). Following hybridization, slides were immersed in 2 × standard saline
3
citrate for 15 min at RT and RNase buffer (RNase A 20 µg/ml, 0.5 M NaCl, 10 mM Tris, pH 8.0,
4
and 1 mM EDTA) for 30 min at 37°C. After intensive washing and dehydration, slides were
5
dipped in twice-diluted Kodak Autoradiography Emulsion, Type NTB (Eastman Kodak,
6
Rochester, NY) and dried at RT for 30 min in a dark room. Slides were then stored at 4°C for
7
approximately 2 weeks in a dark place and developed using Kodak D19 Developer and Fixer
8
(Eastman Kodak).
9
Immunohistochemistry
10
Paraffin sections were prepared as described above. Slides were deparaffinized in xylene
11
and rehydrated in a graded series of ethanol. For antigen retrieval, they were immersed into
12
Target Retrieval Solution, pH 9.0 (Dako) and heated for 15 min at 121°C by autoclaving or
13
incubated for 30 min at 100°C in a water bath. In the case of staining with anti-Gr-1 antibody,
14
slides were pretreated with Proteinase K (100 µg/ml; Wako Pure Chemical Industries, Osaka,
15
Japan) for 10 min at RT for antigen retrieval. Non-specific binding was blocked with 10% normal
16
goat serum for 10 min at RT. Sections were incubated with primary antibodies for 1 h at RT.
17
Endogenous peroxidase was inhibited with 3% hydrogen peroxide in PBS for 15 min. The slides
18
were incubated with the secondary antibody conjugated to HRP-labeled polymer (EnVision+;
19
Dako) for 15 min at RT except for the slides incubated with anti-CD45R or anti-Gr-1 antibodies,
11
1
which were incubated with N-Histofine Simple Stain Mouse MAX PO (Rat) (Nichirei Bioscience,
2
Tokyo, Japan) for 30 min at RT. Reaction products were visualized with diaminobenzidine
3
(Dako); nuclear counterstaining was performed with hematoxylin.
4
BrdU Incorporation Analysis
5
Wild-type (CD109+/+) and homozygous (CD109–/–) mice were injected intraperitoneally with
6
BrdU (5 mg per 100 g body weight; Sigma-Aldrich) at postnatal day 14 (P14) and P28. After 2 h,
7
the mice were sacrificed under general anesthesia, and dorsal skin was resected. The skin
8
tissues were fixed in 10% neutral-buffered formalin, dehydrated and embedded in paraffin.
9
Sections 4-µm thick were prepared on slides. Slides were deparaffinized, rehydrated and
10
immersed in 2N HCl for 30 min at RT for antigen retrieval. Immunohistochemical analysis using
11
anti-BrdU antibody was performed as described above.
12
Immunofluorescence Staining
13
Deparaffinization, hydration, antigen retrieval and blocking steps were performed as
14
described above. The sections were incubated with primary antibody for 2 h at RT. The slides
15
were incubated with Alexa Fluor 488 (1:500)-labeled secondary antibody for 30 min at RT. The
16
images were visualized under a fluorescence microscope (Olympus, Tokyo, Japan).
17
Cell Culture
18
19
Both wild-type and CD109-deficient keratinocytes were isolated from neonatal (P0) dorsal
skin using the CELLnTEC Advanced Cell Systems (CELLnTEC, Bern, Switzerland). They were
12
1
maintained in fully defined, low-calcium (0.07 mM) medium, CnT-07 (CELLnTEC) in an incubator
2
at 37°C and 5% CO2. They were then subcultured by trypsinization and plated on 3.5 cm dishes
3
in CnT-07 media. At 70–80% confluency, cells were starved for 6 h in growth factor-free medium.
4
They were washed once, treated with 0.1nM TGF-β1 (PeproTech, Rocky Hill, NJ) for 1–4 h, and
5
then lysed as described above.
6
In vivo Wound Healing Assay
7
Male CD109+/+ and CD109–/– sibling mice (approximately 22–28g body weight and 8–12
8
weeks old) were anesthetized; their dorsal skin was shaved and swabbed with 70% ethanol prior
9
to the procedure. For wounding, 4-mm punch (Kai Industries, Seki, Japan) biopsies were
10
performed on the shaved area. Wounds were separated by a minimum of 6 mm of uninjured skin.
11
The diameters of the wound area were measured at 0, 1, 3, 5, and 7 days after wounding and
12
wound closure was evaluated as a percentage of the initial wound size.
13
Statistical Analysis
14
Student’s t-test was used to analyze differences in epidermal thickness, wound healing and
15
immunohistochemical positive ratios between CD109–/– and CD109+/+ mice. P < 0.05 was
16
considered significant.
13
1
Results
2
Generation of CD109 knockout / lacZ Knock-in Mice
3
To inactivate the CD109 gene, we engineered a targeting vector in which the 3′ part of
4
exon 1 and the whole of exon 2 were replaced with a lacZ-PGK-neo cassette (Figure 1A). The
5
lacZ sequence was placed in-frame with the start codon of CD109. Male chimeras were
6
generated from one of the targeted ES clones and germ-line transmission of the targeted locus
7
was ascertained by mating to C57BL/6J females. The genotypes of wild-type (CD109+/+),
8
heterozygous (CD109+/-), and homozygous (CD109–/–) mice were determined by Southern blot
9
analysis (Figure 1B) and PCR (Figure 1C). Both heterozygous and homozygous mice were born
10
normally and there was no significant difference in body weight or life span among wild-type,
11
heterozygous, and homozygous mice.
12
Mouse CD109 Is Expressed in the Skin and Testis
13
To assess tissue distribution of CD109, whole lysates from various organs of CD109+/+ and
14
CD109–/– mice were prepared and subjected to western blotting with anti-CD109 antibody. Two
15
bands with molecular masses of about 150 and 180 kDa were detected in the lysates from skin
16
and testis of CD109+/+ mice, but were undetectable in CD109–/– mice, indicating that the two
17
bands were CD109-specific (Figure 2A). The bands detected in cerebrum, cerebellum and liver
18
were non-specific because they were detected both in CD109+/+ and CD109–/– mice.
19
CD109 mRNA expression in the skin and testis was also examined by in situ hybridization.
14
1
CD109-specific signal was detected in the epidermis and the seminiferous tubules of CD109+/+
2
mice, but not in CD109–/– mice (Figure 2B).
3
We subsequently performed immunohistochemical analysis of endogenous CD109 protein
4
expression in various mouse tissues. CD109 protein was expressed in squamous epithelia of the
5
skin and tongue, and seminiferous tubules of the testis in CD109+/+ mice, but not CD109–/– mice
6
(Figure 2C). CD109 expression was detected in the basal and suprabasal layers of the epidermis,
7
including the infundibular portion of the hair follicle (Figure 2C). Tissue distribution of CD109 was
8
compatible with the results of X-gal staining for LacZ expression in CD109-deficient (lacZ
9
knock-in) mice (data not shown). Mouse CD109 expression was not detected in either
10
myoepithelial cells of the mammary and salivary glands or in bronchial basal cells in this study.
11
CD109–/– Mice Display Transient Impairment of Hair Growth
12
Macroscopically, impairment of hair growth in CD109–/– mice was first observed at postnatal
13
day 7 (P7) and persisted until P28 (Figure 3A). Hair of CD109–/– mice was much sparser and less
14
directed than CD109+/+ mice. Hair growth then recovered and no severe impairment was
15
observed after P35 (Figure 3A). However, hair direction remained irregular in CD109–/– mice
16
after P35. All CD109–/– mice displayed the same phenotype to varying degrees. CD109–/– mice,
17
which were generated from CD109+/- mice backcrossed 10 times to the C57BL/6J strain, also
18
showed the same phenotype, eliminating the influence of mouse genetic background of
15
1
C57BL6J/129S6. No apparent abnormalities were observed in other organs, including testis in
2
CD109–/– mice.
3
CD109 Deficiency Results in Epidermal Hyperplasia and Appendage Abnormalities
4
To examine histological aberrations causing transient impairment of hair growth in CD109–/–
5
mice, H-E stained skin specimens were prepared from P0 to P70 CD109+/+ and CD109–/– mice
6
(Figure 3B). There were no apparent differences in hair follicles between CD109+/+ and CD109–/–
7
mice at P0, P3 and P5 (Figure 3B and Supplemental Figure S1 at http://ajp.amjpathol.org/), or in
8
inferior segments of hair follicles at P7 (see Supplemental Figure S1 at http://ajp.amjpathol.org/).
9
At P14, some hairs of CD109–/– mice failed to penetrate the epidermis and their shafts were kinky
10
or zigzagged (Figure 3B; arrow), whereas all hairs in CD109+/+ mice penetrated the epidermis.
11
Many hair follicles of CD109–/– mice at P21 were ectatic; some did not have hair shafts (Figure
12
3B; arrowheads). On the other hand, number of hair follicles did not significantly differ between
13
CD109+/+ and CD109–/– mice from P0 to P70 except for P21. At P21, hair follicles of CD109–/–
14
mice were difficult to count because of distortion of the hair follicles. Oil-red-O staining of the skin
15
of CD109–/– mice revealed accumulation of sebum in the ectatic hair follicles (Figure 3C).
16
Inflammatory cell infiltration was observed in the dermis of CD109–/– mice, which was
17
accompanied by the appearance of skin appendage abnormalities (Figure 3B and Supplemental
18
Figure S2 at http://ajp.amjpathol.org/). The infiltrating cells included CD3+ cells (T lymphocytes),
16
1
Gr-1+ cells (neutrophils) and a small number of CD45R+ cells (B lymphocytes) (see
2
Supplemental Figure S2 at http://ajp.amjpathol.org/).
3
Hyperplasia of the epidermis and sebaceous glands became apparent at P7 in CD109–/–
4
mice (Figure 3B and Supplemental Figure S1 at http://ajp.amjpathol.org/) and sustained until
5
P70. Epidermal thickness of the dorsal skin was quantified by measuring length between basal
6
lamina and lower border of the stratum corneum; differences between CD109+/+ and CD109–/–
7
mice (3 mice per group) were statistically analyzed. Significant differences were observed
8
between the two groups at all time points after P7 except for P35 and P42 (Figure 3D).
9
Epidermal hyperplasia was also seen at infundibular portions of hair follicles in CD109–/– mice.
10
Epidermal hyperplasia was seen in both dorsal skin and sole skin, which does not have hairs
11
(Figure 4B). These findings indicated that CD109 deficiency causes epidermal hyperplasia
12
accompanied by ectatic hair follicles and impairment of normal hair growth.
13
Thickening of the Basal / Suprabasal Layer Is the Cause of Epidermal Hyperplasia
14
To further investigate the epidermal hyperplasia, the dorsal skin epidermis of CD109+/+ and
15
CD109–/– mice was immunohistochemically analyzed using proliferation, differentiation and
16
apoptotic markers (Figure 4A, 4C). The epidermis of CD109–/– mice showed apparent thickening
17
of the basal / suprabasal layer, which was positive for the basal cell markers, p63 and CK14,
18
compared with CD109+/+ mice.33,34 Thickness of the spinous and granular layers, which were
19
positive for CK10 and filaggrin, respectively, were similar between CD109+/+ and CD109–/– mice.
17
1
We also examined the sole skin of adult CD109+/+ and CD109–/– mice at P56. Epidermal
2
hyperplasia and basal / suprabasal layer thickening were observed in the glabrous epidermis of
3
CD109–/– mice (Figure 4B) suggesting that the presence of hair follicles is not related to
4
epidermal hyperplasia in CD109-deficient mice. Proliferation of the basal cells was assessed by
5
BrdU incorporation; apoptosis of keratinocytes was assessed by cleaved caspase-3 staining.
6
While no significant difference in cleaved caspase-3-positive ratio was detected between the two
7
mouse groups (Figure 4C), BrdU-positive ratio was significantly increased in the epidermis of the
8
CD109–/– mice compared with CD109+/+ mice (Figure 4C and Supplemental Figure S3 at
9
http://ajp.amjpathol.org/). These findings suggest that CD109 regulates proliferation and
10
differentiation of keratinocytes.
11
Effect of CD109 Deficiency on TGF-β Signal Is Undetectable in the Epidermis
12
A previous study showed CD109 to be a component of the TGF-β receptor complex and to
13
inhibit TGF-β/Smad signaling in vitro.25–27 To understand the mechanism underlying the
14
epidermal hyperplasia and appendage abnormalities in CD109-deficient mice, we examined the
15
influence of CD109 deficiency on TGF-β/Smad signaling using fluorescence
16
immunohistochemistry. Smad2 phosphorylation, which is reportedly increased by knockdown of
17
CD109 expression,26 was assessed in the epidermis by staining with anti-phospho-Smad2
18
(pSmad2) antibody. However, we found no significant difference in the number of
19
pSmad2-positive cells in the epidermis between CD109+/+ and CD109–/– mice (4 mice per group)
18
1
from P7 to P28 (Figure 5A and Supplemental Figure S4 at http://ajp.amjpathol.org/). We also
2
examined expression of the TGF-β-inducible protein, Smad7, in the epidermis which negatively
3
regulates the strength and duration of TGF-β signaling.35 No apparent difference in Smad7
4
staining was observed between CD109+/+ and CD109–/– mice (data not shown).
5
Stat3 Phosphorylation Was Enhanced in the Epidermis of CD109–/– Mice
6
Recently, recombinant CD109 protein was reported to both downregulate TGF-β signaling,
7
and upregulate STAT3 phosphorylation in human N/TERT-1 keratinocytes in vitro.28 In addition,
8
STAT3 reportedly affects psoriasis-like epidermal hyperplasia.36,37 To examine the effect of
9
CD109 deficiency on Stat3 activation in vivo, we assessed Stat3 phosphorylation in the
10
epidermis by staining with anti-phospho-Stat3 (pStat3) antibody. Interestingly, we found that
11
Stat3 phosphorylation levels were significantly elevated in the epidermis of CD109–/– mice
12
compared with that of CD109+/+ mice, on and after P7 (Figure 5B and Supplemental Figure S5 at
13
http://ajp.amjpathol.org/). We also isolated primary keratinocytes from neonatal (P0) dorsal skin
14
and examined Stat3 phosphorylation and TGF-β/Smad signaling induced by TGF-β1 in them.
15
Primary keratinocytes were cultured for at least 7 days; Stat3 phosphorylation was assessed in
16
the absence or presence of TGF-β1. The level of Stat3 phosphorylation in CD109–/–
17
keratinocytes was high without TGF-β1 stimulation compared with that in CD109+/+ keratinocytes,
18
although the levels of Stat3 phosphorylation were decreased by TGF-β1 stimulation in both
19
CD109+/+ and CD109–/– keratinocytes (see Supplemental Figure 5C at http://ajp.amjpathol.org/).
19
1
On the other hand, the levels of Smad2 phosphorylation induced by TGF-β1 were almost the
2
same at each time point between CD109+/+ and CD109–/– keratinocytes (see Supplemental
3
Figure 5C at http://ajp.amjpathol.org/).
4
In addition, we performed an in vivo wound healing assay to evaluate the effect of CD109 on
5
wound healing, which is known to be regulated by not only TGF-β signaling,38 but also Stat3
6
phosphorylation.29,30,36 The dorsal skin of CD109+/+ and CD109–/– mice were wounded by punch
7
biopsies and wound closure was measured as described in Materials and Methods. However,
8
there was no significant difference in wound closure between CD109+/+ and CD109–/– mice at
9
each time point (Figure 5D).
10
11
Discussion
CD109 is a GPI-anchored cell surface protein, whose expression is normally confined to a
12
very limited number of cell types. By northern blot analysis, CD109 expression was detected only
13
in testis in human and mouse tissues.8 By immunohistochemical analyses, CD109
14
immunoreactivity was detected in human myoepithelial cells of the mammary, salivary, lacrimal
15
and bronchial glands and basal cells of the prostate and bronchial epithelia.10,11 In the present
16
study, we analyzed mouse CD109 expression by immunohistochemistry and detected CD109 in
17
the epidermis of the skin, the squamous epithelia of the tongue, and the seminiferous tubules of
18
the testis. However, we could not detect its expression in myoepithelial cells of the mammary
20
1
and salivary glands or basal cells of the bronchial epithelia indicating that the expression pattern
2
of mouse CD109 is different than human CD109.
3
To investigate the function of CD109 in vivo, we generated CD109–/– mice, which displayed
4
skin abnormalities but not any apparent abnormalities in other tissues. Skin abnormalities in
5
CD109–/– mice included epithelial hyperplasia mainly due to basal / suprabasal layer thickening
6
with increased basal cell proliferation, impairment of hair growth, ectatic hair follicles and
7
sebaceous gland hyperplasia. Among these abnormalities, epithelial hyperplasia could be the
8
primary phenotype as a result of CD109 deficiency, because the epidermal hyperplasia was
9
observed not only in the dorsal skin but also in the sole skin, which does not have skin
10
appendages. Impairment of hair growth and ectatic hair follicles may be secondary changes
11
caused by narrowing of the infundibular portion of the hair follicles due to epithelial hyperplasia,
12
because no developmental or differentiation abnormalities were observed histologically from
13
P0–P5, and hyperplasia of epidermis and sebaceous glands emerged at P7, prior to kinked hair
14
shafts and ectatic hair follicles at P14. Inflammatory cell infiltration was also observed in the skin
15
of CD109–/– mice. The infiltrating cells include T lymphocytes, neutrophils, and a small number of
16
B lymphocytes. While the precise mechanism of the infiltration of these cells in the skin of
17
CD109–/– mice remains to be elusive, it may be due to psoriasis-like skin alterations36,37 or a
18
reaction to accumulation of sebum. Alternatively, the inflammation might contribute to epidermal
19
hyperplasia and hair follicle degeneration. Although CD109 is highly expressed in the
21
1
seminiferous tubules of the testis in both human and mouse, CD109–/– mice did not show any
2
apparent abnormality in the testis; hence its function in the testis remains unclear.
3
In immunohistochemical analyses using human cancer tissues, we previously revealed that
4
CD109 is highly expressed in SCCs of the lung and oral cavity.10,14 In our study using SCCs of
5
the oral cavity, CD109 was expressed in all well-differentiated SCCs, but only in 89% and 64% of
6
moderately- and poorly-differentiated SCCs, respectively, implying that CD109 expression is
7
correlated with the differentiation stage of SCCs. In the present study, we found that a CD109
8
deficiency results in thickening of the basal / suprabasal cell layer in the epidermis with increased
9
basal cell proliferation. These findings suggest that CD109 regulates proliferation and
10
differentiation of keratinocytes in vivo, even in malignant tumors. Further investigation is
11
necessary to clarify the roles of CD109 in development of human cancer.
12
CD109 is reported to be a negative regulator of TGF-β signaling in keratinocytes in vitro;
13
over-expression of CD109 inhibits TGF-β signaling, whereas knockdown of CD109 up-regulates
14
its signaling.25–28 Because activation of TGF-β signaling suppresses cell proliferation of
15
keratinocytes in vitro,17 we evaluated the status of Smad2, one of the R-Smads, and Smad7, one
16
of the I-Smads,20–23 in CD109–/– mice. Cutaneous wound healing is reportedly delayed in
17
transgenic mice that over-express Smad2 in the epidermis,39 and hyperplasia of the epidermis
18
and sebaceous glands is induced in transgenic mice that over-express Smad7 in the
19
epidermis.35,40 The skin phenotype of Smad7 transgenic mice was similar to CD109–/– mice
22
1
observed in this study. Epidermal hyperplasia in Smad7 transgenic mice may be due to
2
suppression of R-Smad (e.g. Smad2) phosphorylation and a reduction of TGFBRI and TGFBRII
3
protein levels, resulting in blockage of TGF-β-induced growth arrest in keratinocytes.40 However,
4
CD109 deficiency influenced neither levels of Smad2 phosphorylation and Smad7 expression,
5
nor the wound-healing ability in the epidermis of CD109–/– mice. These findings suggest that
6
either (a) long-lasting effects of CD109 deficiency on the TGF-β signaling pathway may be
7
masked by feedback signals or other compensatory mechanisms; or (b) CD109 does not
8
modulate TGF-β/Smad2 signaling in mouse epidermis, but is associated with a different
9
signaling pathway.
10
CD109 can reportedly regulate STAT3 activation in human keratinocytes.28 It was also
11
reported that upregulation of Stat3 is associated with psoriasis,36,37 which is a common disease
12
characterized histologically by epidermal hyperplasia, altered epidermal differentiation, and local
13
accumulation of acute and chronic inflammatory cells.30,36,41 We therefore evaluated the level of
14
Stat3 phosphorylation in CD109–/– mice. Interestingly, immunofluorescence analysis showed
15
Stat3 phosphorylation was elevated in the epidermis of CD109–/– mice compared with CD109+/+
16
mice. Enhanced Stat3 phosphorylation was also observed in primary keratinocytes isolated from
17
CD109–/– mice under cytokine- and growth factor-free conditions, although TGF-β1 stimulation
18
markedly down-regulated Stat3 phosphorylation. Stat3 upregulation is reportedly associated with
19
increased proliferation and impaired differentiation of keratinocytes and results in psoriasis-like
23
1
skin alterations, including hyperkeratosis and acanthosis with inflammatory infiltrates in
2
vivo.30,36,37,41 This phenotype is similar to the skin abnormalities in CD109–/– mice. Thus, our
3
findings suggest that CD109 regulates Stat3 activation, which could be associated with
4
epidermal hyperplasia in CD109–/– mice. Further analyses of cytokine or growth factor signaling
5
pathways will provide insight into the precise mechanism of skin abnormalities developed in
6
CD109–/– mice.
7
Acknowledgments
8
We thank Mr. Koichi Imaizumi, Mr. Kozo Uchiyama and Mrs. Akiko Itoh (Department of
9
Pathology), Mr. Nobuyoshi Hamada and Mr. Yoshiyuki Nakamura (Radioisotope Center Medical
10
Branch) and Mr. Yasutaka Ohya and Mrs. Kumiko Yano-Ohya (Division for Research of
11
Laboratory Animals, Center for Research of Laboratory Animals and Medical Research
12
Engineering) for technical assistance.
13
24
1
2
References
1.
Sutherland DR, Yeo E, Ryan A, Mills GB, Bailey D, Baker MA: Identification of a cell-surface
3
antigen associated with activated T lymphoblasts and activated platelets. Blood 1991,
4
77:84-93
5
2.
Haregewoin A, Solomon K, Hom RC, Soman G, Bergelson JM, Bhan AK, Finberg RW:
6
Cellular expression of a GPI-linked T cell activation protein. Cell Immunol 1994,
7
156:357-370
8
3.
Smith JW, Hayward CP, Horsewood P, Warkentin TE, Denomme GA, Kelton JG:
Characterization and localization of the Gova/b alloantigens to the
9
10
glycosylphosphatidylinositol-anchored protein CDw109 on human platelets. Blood 1995,
11
86:2807-2814
12
4.
Lin M, Sutherland DR, Horsfall W, Totty N, Yeo E, Nayar R, Wu XF, Schuh AC: Cell surface
13
antigen CD109 is a novel member of the α2-macroglobulin/C3, C4, C5 family of
14
thioester-containing proteins. Blood 2002, 99:1683-1691
15
5.
16
17
Kelton JG, Smith JW, Horsewood P, Humbert JR, Hayward CP, Warkentin TE: Gova/b
alloantigen system on human platelets. Blood 1990, 75:2172-2176
6.
Murray LJ, Bruno E, Uchida N, Hoffman R, Nayar R, Yeo EL, Schuh AC, Sutherland DR:
18
CD109 is expressed on a subpopulation of CD34+ cells enriched in hematopoietic stem and
19
progenitor cells. Exp Hematol 1999, 27:1282-1294
25
1
7.
Giesert C, Marxer A, Sutherland DR, Schuh AC, Kanz L, Burring H-J: Antibody W7C5
2
defines a CD109 epitope expressed on CD34+ and CD34- hematopoietic and mesenchymal
3
stem cell subsets. Ann NY Acad Sci 2003, 996:227-230
4
8.
Hashimoto M, Ichihara M, Watanabe T, Kawai K, Koshikawa K, Yuasa N, Takahashi T,
5
Yatabe Y, Murakumo Y, Zhang JM, Nimura Y, Takahashi M: Expression of CD109 in human
6
cancer. Oncogene 2004, 23:3716-3720
7
9.
Zhang JM, Hashimoto M, Kawai K, Murakumo Y, Sato T, Ichihara M, Nakamura S,
8
Takahashi M: CD109 expression in squamous cell carcinoma of the uterine cervix. Pathol Int
9
2005, 55:165-169
10
10. Sato T, Murakumo Y, Hagiwara S, Jijiwa M, Suzuki C, Yatabe Y, Takahashi M: High-level
11
expression of CD109 is frequently detected in lung squamous cell carcinomas. Pathol Int
12
2007, 57:719-724
13
11. Hasegawa M, Hagiwara S, Sato T, Jijiwa M, Murakumo Y, Maeda M, Moritani S, Ichihara S,
14
Takahashi M: CD109, a new marker for myoepithelial cells of mammary, salivary, and
15
lacrimal glands and prostate basal cells. Pathol Int 2007, 57:245-250
16
12. Hasegawa M, Moritani S, Murakumo Y, Sato T, Hagiwara S, Suzuki C, Mii S, Jijiwa M,
17
Enomoto A, Asai N, Ichihara S, Takahashi M: CD109 expression in basal-like breast
18
carcinoma. Pathol Int 2008, 58:288-294
26
1
13. Järvinen AK, Autio R, Kilpinen S, Saarela M, Leivo I, Grénman R, Mäkitie AA, Monni O:
2
High-resolution copy number and gene expression microarray analyses of head and neck
3
squamous cell carcinoma cell lines of tongue and larynx. Genes Chromosomes Cancer
4
2008, 47:500-509
5
14. Hagiwara S, Murakumo Y, Sato T, Shigetomi T, Mitsudo K, Tohnai I, Ueda M, Takahashi M:
6
Up-regulation of CD109 expression is associated with carcinogenesis of the squamous
7
epithelium of the oral cavity. Cancer Sci 2008, 99:1916-1923
8
9
10
11
15. Ohshima Y, Yajima I, Kumasaka MY, Yanagishita T, Watanabe D, Takahashi M, Inoue Y,
Ihn H, Matsumoto Y, Kato M: CD109 expression levels in malignant melanoma. J Dermatol
Sci 2010, 57:140-142
16. Hagikura M, Murakumo Y, Hasegawa M, Jijiwa M, Hagiwara S, Mii S, Hagikura S,
12
Matsukawa Y, Yoshino Y, Hattori R, Wakai K, Nakamura S, Gotoh M, Takahashi M:
13
Correlation of pathological grade and tumor stage of urothelial carcinomas with CD109
14
expression. Pathol Int 2010, 60:735-743
15
16
17. Rahimi RA, Leof EB: TGF-β signaling: a tale of two responses. J Cell Biochem 2007,
102:593-608
17
18. Massagué J: TGFβ in Cancer. Cell 2008, 134:215-230
18
19. Li MO, Flavell RA: TGF-β: a master of all T cell trades. Cell 2008, 134:392-404
27
1
2
3
4
5
6
7
8
9
20. Shi Y, Massagué J: Mechanisms of TGF-β signaling from cell membrane to the nucleus. Cell,
2003 113:685-700
21. Schmierer B, Hill CS: TGFβ-SMAD signal transduction: molecular specificity and functional
flexibility. Nat Rev Mol Cell Biol 2007, 8:970-982
22. Moustakas A, Heldin CH: The regulation of TGFβ signal transduction. Development 2009,
136:3699-3714
23. Itoh S, ten Dijke P: Negative regulation of TGF-β receptor/Smad signal transduction. Curr
Opin Cell Biol 2007, 19:176-184
24. Tam BY, Germain L, Philip A: TGF-β receptor expression on human keratinocytes: a 150
10
kDa GPI-anchored TGF-β1 binding protein forms a heteromeric complex with type I and type
11
II receptors. J Cell Biochem 1998, 70:573-586
12
25. Finnson KW, Tam BY, Liu K, Marcoux A, Lepage P, Roy S, Bizet AA, Philip A: Identification
13
of CD109 as part of the TGF-β receptor system in human keratinocytes. FASEB J 2006,
14
20:1525-1527
15
26. Hagiwara S, Murakumo Y, Mii S, Shigetomi T, Yamamoto N, Furue H, Ueda M, Takahashi M:
16
Processing of CD109 by furin and its role in the regulation of TGF-β signaling. Oncogene
17
2010, 29:2181-2191
28
1
27. Bizet AA, Liu K, Tran-Khanh N, Saksena A, Vorstenbosch J, Finnson KW, Buschmann MD,
2
Philip A: The TGF-β co-receptor, CD109, promotes internalization and degradation of TGF-β
3
receptors. Biochim Biophys Acta 2011, 1813:742-753
4
28. Litvinov IV, Bizet AA, Binamer Y, Jones DA, Sasseville D, Philip A: CD109 release from the
5
cell surface in human keratinocytes regulates TGF-β receptor expression, TGF-β signalling
6
and STAT3 activation: relevance to psoriasis. Exp Dermatol 2011, 20:627-632
7
8
9
10
11
12
13
29. Yu H, Kortylewski M, Pardoll D: Crosstalk between cancer and immune cells: role of STAT3
in the tumour microenvironment. Nat Rev Immunol 2007, 7:41-51
30. Sano S, Chan KS, DiGiovanni J: Impact of Stat3 activation upon skin biology: a dichotomy of
its role between homeostasis and diseases. J Dermatol Sci 2008 50:1-14
31. Walia B, Wang L, Merlin D, Sitaraman SV: TGF-β down-regulates IL-6 signaling in intestinal
epithelial cells: critical role of SMAD-2. FASEB J 2003 17:2130-2132
32. Staršíchova A, Lincová E, Pernicová Z, Kozubík A, Souček K: TGF-β1 suppresses
14
IL-6-induced STAT3 activation through regulation of Jak2 expression in prostate epithelial
15
cells. Cell Signal 2010 22:1734-1744
16
33. Parsa R, Yang A, McKeon F, Green H: Association of p63 with proliferative potential in
17
normal and neoplastic human keratinocytes. J Invest Dermatol 1999, 113:1099-1105
18
34. Fuchs E: Epidermal differentiation: the bare essentials. J Cell Biol 1990, 111:2807-2814
29
1
35. He W, Li AG, Wang D, Han S, Zheng B, Goumans MJ, ten Dijke P, Wang XJ:
2
Overexpression of Smad7 results in severe pathological alterations in multiple epithelial
3
tissues. EMBO J 2002, 21:2580-2590
4
36. Sano S, Chan KS, Carbajal S, Clifford J, Peavey M, Kiguchi K, Itami S, Nickoloff BJ,
5
DiGiovanni J: Stat3 links activated keratinocytes and immunocytes required for development
6
of psoriasis in a novel transgenic mouse model. Nat Med 2005 11:43-49
7
37. Zheng Y, Danilenko DM, Valdez P, Kasman I, Eastham-Anderson J, Wu J, Ouyang W:
8
Interleukin-22, a TH17 cytokine, mediates IL-23-induced dermal inflammation and
9
acanthosis. Nature 2007 445:648-651
10
11
12
38. Werner S, Grose R: Regulation of wound healing by growth factors and cytokines. Physiol
Rev 2003, 83:835-870
39. Hosokawa R, Urata MM, Ito Y, Bringas P Jr, Chai Y: Functional significance of Smad2 in
13
regulating basal keratinocyte migration during wound healing. J Invest Dermatol 2005,
14
125:1302-1309
15
40. Han G, Li AG, Liang YY, Owens P, He W, Lu S, Yoshimatsu Y, Wang D, ten Dijke P, Lin X,
16
Wang XJ: Smad7-induced β-catenin degradation alters epidermal appendage development.
17
Dev Cell 2006, 11:301-312
30
1
41. Wolk K, Haugen HS, Xu W, Witte E, Waggie K, Anderson M, Vom Baur E, Witte K,
2
Warszawska K, Philipp S, Johnson-Leger C, Volk HD, Sterry W, Sabat R: IL-22 and IL-20
3
are key mediators of the epidermal alterations in psoriasis while IL-17 and IFN-γ are not. J
4
Mol Med 2009 87:523-536
5
31
1
Table 1. List of Primary Antibodies
Antibodies to
Clone
Source
Application
Dilution
Pretreatment
CD109
C-9 / mouse monoclonal
Santa Cruz Biotechnology
Immunoblotting
1:1000
–
CD109
M-250 / rabbit polyclonal
Santa Cruz Biotechnology
Immunoperoxidase
1:500
Autoclave
Smad7
H-79 / rabbit polyclonal
Santa Cruz Biotechnology
Immunofluorescence
1:20
Water bath
Sigma-Aldrich
Immunoblotting
1:5000
–
Dako
Immunoperoxidase
1:500
Water bath
eBioscience
Immunoperoxidase
1:3200
Water bath
eBioscience
Immunoperoxidase
1:100
Sigma-Aldrich
Immunoperoxidase
1:100
Water bath
β-actin
CD3
CD45R
Gr-1
p63
AC-74 / mouse
monoclonal
Rabbit polyclonal
RA3-6B2 / rat
monoclonal
RB6-8C5 / rat
monoclonal
4A4 / mouse
monoclonal
Proteinase K
RT, 10 min
CK10
Rabbit polyclonal
Covance Inc
Immunoperoxidase
1:1000
Water bath
CK14
AF64 / rabbit polyclonal
Covance Inc
Immunoperoxidase
1:1000
Water bath
Filaggrin
Rabbit polyclonal
Covance Inc
Immunoperoxidase
1:500
Water bath
BD Biosciences
Immunoperoxidase
1:100
Rabbit polyclonal
Cell Signaling Technology
Immunoperoxidase
1:100
Water bath
Rabbit polyclonal
Cell Signaling Technology
Immunofluorescence
1:100
Water bath
Immunoblotting
1:1000
–
Mouse monoclonal
Cell Signaling Technology
Immunoblotting
1:1000
–
Rabbit monoclonal
Cell Signaling Technology
Immunofluorescence
1:20
Water bath
Immunoblotting
1:2000
–
Rabbit monoclonal
Cell Signaling Technology
Immunoblotting
1:2000
–
BrdU
C-caspase-3
pSmad2
(Ser465/467)
Smad2
pStat3
(Tyr705)
Stat3
B44 / mouse
monoclonal
2
32
2N HCl
RT, 30 min
1
Figure Legends
2
Figure 1. Generation of Mice Lacking CD109.
3
A. Schematic representation of the targeting vector. A portion of exon 1 and all of exon 2
4
were replaced by the lacZ-PGK-neo cassette, which was placed in-frame with the start codon of
5
CD109 (triangle). DTa, diphtheria toxin A gene; neo, neomycin-resistance gene. B. Mouse
6
genotyping by Southern blotting. SpeI-digested genomic DNA isolated from the tails of wild-type
7
(+/+), heterozygous (+/–) and homozygous (–/–) CD109 mice were hybridized with the 5′ probe
8
shown in A. SpeI digestion yielded a 3.6 kb fragment for the wild-type allele and a 6.4 kb
9
fragment for the mutant allele. C. Mouse genotyping by PCR. Primer positions are indicated by
10
half arrows in A. PCR products amplified with P1 and P2 (wild-type allele) are 205 bp; those with
11
P3 and P4 (mutant allele) are 605 bp.
12
Figure 2. CD109 Is Expressed in the Skin and Testis in Mice.
13
A. Western blotting for CD109 expression in various tissues of adult CD109+/+ and CD109–/–
14
sibling mice. CD109-specific bands of ~150 and ~180 kDa were detected in the skin and testis of
15
CD109+/+ mice (arrows), but were undetectable in CD109–/– mice. Blots probed with anti-β-actin
16
antibody are shown as a loading control. B. In situ hybridization for CD109 mRNA expression in
17
the skin and testis. 33P-labeled antisense ribonucleotide probes for CD109 were hybridized to
18
frozen sections of the skin and the testis of 6-week-old CD109+/+ and CD109–/– mice. CD109
19
mRNA was detected in the epidermis of the skin and the seminiferous tubules of the testis of
33
1
CD109+/+ mice, but not in CD109–/– mice. Arrowheads indicate the boundary of the epidermis.
2
Signals were captured by dark-field microscopy. Scale bars: 100 µm. C. Immunohistochemical
3
analysis of CD109 expression in mice. CD109 protein is expressed in the squamous epithelia of
4
the skin and tongue, and the seminiferous tubules of the testis of adult CD109+/+ mice (upper
5
panel), but not in CD109–/– mice (lower panel). Scale bars: 100 µm.
6
Figure 3. CD109–/– Mice Develop Skin Abnormalities.
7
A. Macroscopic images of hair growth impairment of a CD109–/– mouse. Images of the
8
dorsal skin of a CD109–/– mouse (lower panels) and its CD109+/+ sibling (upper panels) were
9
taken every 7 days (not all images shown). Impairment of hair growth in the CD109–/– mouse was
10
apparent from P7 to P28, but then hair growth recovered and no severe impairment was
11
observed after P35. B. Microscopic images of skin abnormalities in CD109–/– mice. H-E sections
12
of the dorsal skin were prepared from CD109+/+ (upper panels) and CD109–/– (lower panels) mice
13
at each age. Hair follicles of CD109–/– mice exhibited ectasia from P14 to P21 (arrowheads) and
14
hair shafts were kinked at P14 (arrow). Hyperplasia of the epidermis and sebaceous glands was
15
first observed at P7 and remained at P70. Scale bar: 100 µm. C. Oil-red-O staining of the skin of
16
CD109+/+ (upper panels) and CD109–/– (lower panels) mice. Accumulation of sebum in the ectatic
17
hair follicles was apparent in CD109–/– mice at P14 (arrowheads). Scale bar: 50 µm. D.
18
Comparison of epidermal thickness between CD109+/+ and CD109–/– mice. H-E sections of the
19
dorsal skin were prepared from CD109+/+ and CD109–/– mice at each age (n = 3) and epidermal
34
1
thickness of interfollicular epidermis was measured as described in Materials and Methods
2
(Values are means ± SD). * P < 0.05; ** P < 0.01.
3
Figure 4. CD109 Deficiency Causes Thickening of the Epidermal Basal / Suprabasal
4
Layer.
5
A, B. Immunohistochemical analysis of the epidermis of CD109+/+ and CD109–/– mice.
6
Dorsal (A) and sole (B) skin sections from CD109+/+ (left) and CD109–/– (right) mice were
7
immunostained with antibodies against p63, cytokeratin 14 (CK14), cytokeratin 10 (CK10), and
8
filaggrin. Thickening of the basal / suprabasal layer, which is positive for p63 and CK14, was
9
observed in CD109–/– mice compared with CD109+/+ mice. No apparent difference was observed
10
in the spinous and granular layers, which are positive for CK10 and filaggrin, respectively,
11
between CD109+/+ and CD109–/– mice. Scale bars: 100 µm. C. The ratios of BrdU-positive cells
12
or cleaved caspase-3-positive cells in epidermis of CD109+/+ and CD109–/– mice. While no
13
significant difference was detected in cleaved caspase-3-positive ratio between CD109+/+ and
14
CD109–/– mice, the BrdU-positive ratio was significantly increased in the epidermis of
15
CD109–/–mice compared with CD109+/+mice at P14 and P28 (n = 3). * P < 0.05; ** P < 0.01.
16
Figure 5. TGF-β Signal Activation Was Undetectable, but Stat3 Phosphorylation Was
17
Enhanced in the Epidermis of CD109-Deficient Mice.
18
19
A. Fluorescence immunostaining of phosphorylated Smad2 in the epidermis of CD109+/+
and CD109–/– mice. Dorsal skin sections from CD109+/+ (upper panels) and CD109–/– (lower
35
1
panels) mice at P14 were immunostained with an antibody against phospho-Smad2 (pSmad2;
2
green). A similar nuclear staining was observed in epidermis of both CD109+/+ and CD109–/–
3
mice. Nuclear counterstain was performed with DAPI (blue). Scale bar: 100 µm. B. Fluorescence
4
immunostaining of phosphorylated Stat3 in the epidermis of CD109+/+ and CD109–/– mice. Dorsal
5
skin sections from CD109+/+ (upper panels) and CD109–/– (lower panels) mice at P14 were
6
immunostained with an antibody against phospho-Stat3 (pStat3; green). The level of Stat3
7
phosphorylation was elevated in the epidermis of CD109–/– mice compared with CD109+/+ mice.
8
Nuclear counterstain was performed with DAPI (blue). Scale bar: 100 µm. C. Time course of
9
Stat3 and Smad2 phosphorylation after TGF-β1 (0.1 nM) stimulation in CD109+/+ and CD109–/–
10
keratinocytes. Phosphorylation of Stat3 and Smad2 was determined by western blotting. Levels
11
of Smad2 phosphorylation induced by TGF-β1 in CD109+/+ and CD109–/– keratinocytes were
12
almost the same at each time point, whereas levels of Stat3 phosphorylation in CD109–/–
13
keratinocytes were significantly elevated under growth factor-free conditions compared with
14
those in CD109+/+ keratinocytes. Expression of total Stat3 and Smad2 is also shown. Expression
15
of β-actin is shown as a loading control. D. In vivo wound healing assay in CD109+/+ and
16
CD109–/– mice. The assay was performed as described in Materials and Methods. No significant
17
difference was observed between CD109+/+ and CD109–/– mice.
36
1
Figure S1. Low-Magnified Images of the Skin of CD109+/+ and CD109–/– Mice.
Low-magnified microscopic appearance of the skin of CD109+/+ (left panels) and CD109–/–
2
3
mice (right panels) at P5 and P7. H-E sections of the dorsal skin were prepared. Scale bars: 400
4
µm.
5
Figure S2. Inflammatory Cell Infiltration in the Skin of CD109+/+ and CD109–/– Mice.
6
A. H-E staining showed inflammatory cell infiltration in the dermis of CD109–/– mice at P14.
7
B-D. Immunohistochemical analysis showed CD3+ cells (T lymphocytes), Gr-1+ cells
8
(neutrophils) and a small number of CD45R+ cells (B lymphocytes) at P14. Dark blue arrowheads
9
indicate T-lymphocytes (B), B-lymphocytes (C) or neutrophils (D) in the dermis. The cells
10
indicated by black arrows in B are dendritic cells in the epidermis. Scale bar: 100 µm.
11
Figure S3. BrdU Incorporation Analysis in the Epidermis of CD109+/+ and CD109–/– Mice.
12
Dorsal skin sections from CD109+/+ (left panels) and CD109–/– (right panels) mice from P14
13
and P28. Skin tissues were obtained 2 h after intrapenitoneal injection of BrdU. Prepared slides
14
were immunostained with anti-BrdU antibody. Scale bar: 200 µm.
15
Figure S4. Positive Ratios of pSmad2 in the Epidermis of CD109+/+ and CD109–/– Mice.
16
No significant difference was detected in the positive ratio between CD109+/+ and CD109–/–
17
mice from P7 to P28 (4 mice per group at each time point).
18
Figure S5. Fluorescence Immunostaining of pStat3 in the Epidermis of CD109+/+ and
19
CD109–/– Mice.
1
1
Dorsal skin sections from CD109+/+ (upper panels) and CD109–/– (lower panels) mice from
2
P0 to P70 were immunostained with an antibody against pStat3 (green). Stat3 phosphorylation
3
was increased in the epidermis of CD109–/– mice on and after P7. Nuclear counterstain was
4
performed with DAPI (blue). Scale bars: 100 µm.
2
A
205 bp
P1 P2
ATG
probe
exon1
CD109
exon2
Genomic Locus
3.6 kb
SpeI
SpeI
short arm
DTa
exon1
long arm
lacZ
PGK neo
Targeting Vector
PGK neo
Targeted Locus
SpeI
603 bp
ATG
probe
exon1
P3 P4
lacZ
6.4 kb
SpeI
SpeI
B
C
6.4 kb
603 bp
205 bp
3.6 kb
CD109
+/+
+/-
-/-
CD109
+/+
+/-
-/-
Fig. 1
CD109
reb
ce rum
reb
he ellum
ar
lun t
g
live
r
sp
lee
pa n
nc
thy reas
m
sa us
liv
kid ary g
lan
ne
d
tes y
tis
ov
ary
do
rs
es al sk
op
in
sto hagu
ma s
ch
kDa
250
+/+
CD109
150
ce
A
CD109
250
150
-/-
B
dorsal skin
CD109
+/+
testis
CD109
-/-
CD109
+/+
CD109
-/-
C
CD109
CD109
+/+
-/-
dorsal skin
sole skin
tongue
testis
Fig. 2
A
CD109
CD109
P3
CD109
CD109
CD109
P21
P28
P35
P56
P70
-/-
P0
P3
P5
P7
P14
P21
P28
P35
P56
P70
+/+
-/-
+/+
-/-
C
CD109
P14
+/+
B
CD109
P7
D
+/+
50
CD109
*
40
*
30
20
* **
CD109
(n=3)
NS NS
**
*
+/+
-/-
*
10
CD109
-/-
0
0
20
40
60
80
100
Fig. 3
A
CD109
+/+
CD109
-/-
B
CD109
p63
p63
CK14
CK14
CK10
CK10
Filaggrin
Filaggrin
C
10
*
8
**
6
4
2
0
P14
P28
Cleaved Caspase-3positive ratio [%]
H-E
BrdU-positive ratio [%]
H-E
0.12
0.1
0.08
0.06
0.04
0.02
0
+/+
-/-
CD109
NS
NS
P14
CD109
CD109
(n=3)
+/+
-/-
P28
Fig. 4
A
+/+
CD109
-/CD109
pSmad2
DAPI
Merge
pStat3
DAPI
Merge
B
+/+
CD109
-/CD109
1
2
4
1
[h]
4
[kDa]
180
2
CD109
pStat3
86
Stat3
86
pSmad2
60
Smad2
60
46
CD109
+/+
CD109
-/-
D
CD109
wound diameter [%]
C
100
80
CD109
(n=8)
+/+
-/-
60
40
20
0
0 1 2 3 4 5 6 7 8
days after wounding [day]
Fig. 5
Fig. S1
CD109
P5
P7
+/+
CD109
-/-
Fig. S2
A
B
C
D
CD109
+/+
CD109
-/-
Fig. S3
CD109
P14
P28
+/+
CD109
-/-
pSmad2-positive ratio [%]
Fig. S4
CD109
100
CD109
(n=4)
80
60
40
20
0
P7
P14
P21
P28
+/+
-/-
Fig. S5
P0
P5
P7
P14
P21
P28
P56
P70
+/+
CD109
-/CD109
+/+
CD109
-/CD109