Skip Navigation


Carcinogenesis Advance Access originally published online on May 10, 2007
Carcinogenesis 2007 28(10):2074-2081; doi:10.1093/carcin/bgm112
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/10/2074    most recent
bgm112v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Honda, K.
Right arrow Articles by Sakai, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Honda, K.
Right arrow Articles by Sakai, T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Epidermal hyperplasia and papillomatosis in mice with a keratinocyte-restricted deletion of csk

Kazuhisa Honda1,5,{dagger}, Takehisa Sakaguchi1,{dagger}, Keiko Sakai1, Christian Schmedt2, Angel Ramirez3, Jose Luis Jorcano3, Alexander Tarakhovsky4, Hiroshi Kamisoyama1,5 and Takao Sakai1,*

1 Department of Biomedical Engineering and Orthopaedic Research Center/ND20, Lerner Research Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA
2 Genomics Institute of the Novartis Research Foundation, San Diego, CA 92121 USA
3 Epithelial Damage, Repair and Tissue Engineering Project, CIEMAT, Madrid, 28040, Spain
4 Laboratory of Lymphocyte Signaling, The Rockefeller University, New York, NY 10021, USA
5 Department of Animal Science, Faculty of Agriculture, Kobe University, Kobe 657-8501, Japan

* To whom correspondence should be addressed. Tel: +1 216 445 3223; Fax: +1 216 444 9198; Email: sakait{at}ccf.org


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 References
 
The Src family kinases (SFKs) are believed to play critical roles in malignant transformation, as well as in growth, invasion and dissemination of neoplastic tissue. Inhibition of SFK-mediated signal transduction and activation of downstream targets inhibits tumor progression. To determine whether constitutive activity of SFK per se is sufficient to induce tumorigenesis in vivo, we have generated a mouse model with a keratinocyte-restricted deletion of the SFK-negative regulator csk (Csk-K5 mice). Even though expression levels of SFKs were lower in C-terminal Src kinase (Csk)-null keratinocytes, activity levels were higher than in control keratinocytes. At the age of 3 months, all Csk-K5 mice displayed signs of chronic inflammation in dermis and epidermal hyperplasia. About 19% of Csk-K5 mice (7 out of 36) developed papillomatous lesions. However, these lesions did not show any signs of neoplastic transformation over the next 8 months. Epidermal hyperplasia and hyperkeratosis in Csk-K5 mice were associated with an increased number of stem cells in the interfollicular epidermis, an increased proliferation of basal keratinocytes and a delayed terminal differentiation of the suprabasal keratinocytes. Our results clearly demonstrate that even though SFK-mediated signaling promotes tumor progression, elevated activity of SFKs in vivo alone is not sufficient to induce neoplastic transformation.

Abbreviations: Csk, C-terminal Src kinase; DEJ, dermal–epidermal junction; FAK, focal adhesion kinase; mAb, monoclonal antibody; MEM, modified Eagle’s medium; PBS, phosphate-buffered saline; SFK, Src family kinase


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 References
 
Src family kinases (SFKs) are widely distributed non-receptor type tyrosine kinases that are involved in multiple signaling pathway downstream of various types of cell surface receptors. SFKs are believed to play critical roles in malignant transformation, as well as in growth, invasion and dissemination of neoplastic tissue (1,2). Inhibiting SFK-mediated signal transduction and activation of downstream targets suppress tumor progression, suggesting that negative regulatory mechanisms for SFKs should protect cells from carcinogenesis. C-terminal Src kinase (Csk) is a cytoplasmic tyrosine kinase that controls the activity of SFKs through phosphorylation of their C-terminal tyrosine. Phosphorylation of this tyrosine results in the suppression of SFK kinase activity and the down-regulation of downstream signaling pathways (3). The complete knockout mice of the csk gene have demonstrated that Csk is essential for embryonic development (4,5). Mice lacking csk die around embryonic day 10, and the lack of Csk causes an increase in activity of SFKs and hyperactivation of numerous signaling pathways that regulate cell proliferation, differentiation and migration. However, because of embryonic lethality, this knockout model cannot be used to address the role of Csk in adult physiology and pathology.

SFKs, particularly Src and Fyn, are known to play critical roles not only in the proliferation and differentiation of keratinocytes but also in the malignant transformation of the skin. Activation of Fyn controls cell–cell adhesion during normal keratinocyte differentiation, whereas increased Fyn activity suppresses growth of keratinocytes (6,7). Keratin 14-Fyn transgenic mice exhibit a phenotype of thickened and hyperplastic epidermis (8). Over-expression of a constitutively active form of Src in epidermis induces hyperplasia and spontaneous squamous cell carcinoma, suggesting that activation of Src can substitute for an initiating event during skin carcinogenesis (9,10). These experimental models using exogenously expressed proteins demonstrate that an increased expression/activity of Src or Fyn induces epidermal hyperplasia. Given the important signaling functions determined for SFKs in the epidermis, we hypothesized that a naturally increased activity of all members of SFKs expressed in epidermis due to the loss of Csk would induce neoplastic transformation in keratinocytes in vivo. To address this hypothesis, we have generated a mouse model with a keratinocyte-restricted deletion of Csk (Csk-K5 mice) by crossing mice carrying a floxed csk gene (11) with transgenic mice that express Cre recombinase under the control of the keratin 5 promoter (12). In the skin, this promoter is active in basal cells of stratified epithelia and in the outer root sheath of hair follicles. We report here that the loss of Csk leads to epidermal hyperplasia and hyperkeratosis. At the age of 3 months, 19% of the Csk-K5 mice developed papillomatosis. However, malignant transformation of these lesions was not observed during an 8-month follow-up period. We found that an increased number of stem cells in the interfollicular epidermis, an increased proliferation of basal keratinocytes and a delayed terminal differentiation of the suprabasal keratinocytes underlay epidermal hyperplasia of the Csk-K5 mice. Our results provide the first evidence that elevated activity of SFKs in vivo by itself is not sufficient to induce neoplastic transformation.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 References
 
Mice
All mice were maintained and bred at the Cleveland Clinic’s Biological Research Unit in accordance with the institutional guidelines and National Institutes of Health standards. This study was approved by the Institutional Animal Care and Use Committee. Mice were regularly monitored and had free access to standard mice chow and water.

Antibodies
The following antibodies were used for the analyses: mouse monoclonal antibody (mAb) against Csk (clone 52, BD Transduction Laboratories, Lexington, KY); rabbit antibody against Csk (a kind gift of Dr David Schlaepfer, Department of Immunology, The Scripps Research Institute, La Jolla, CA); mouse mAb against active SFKs (clone 28, which recognizes active form of SFKs, including Src, Fyn, Yes and Fgr) (13); mouse mAb against Src (clone 184Q20, BioSource/Invitrogen, Camariuo, CA); mAbs against Fyn (clone 25, BD Transduction Laboratories; clone 1S, Biosource/Invitrogen and clone y303, Epitomics, Burlingame, CA) and rabbit antibody against Fyn (Fyn3, Santa Cruz Biotechnology, Santa Cruz, CA); rabbit antibody against SFKs (SRC2, Santa Cruz Biotechnology); rabbit antibodies against Srcp418 and focal adhesion kinase (FAK)p397 (BioSource/Invitrogen); rabbit antibodies against keratin 6, 10, 14 and loricrin (Covance Res., Princeton, NJ); rat mAbs against integrin ß1 [clone 9EG7 (BD PharMingen, San Jose, CA), which recognizes only activated integrin ß1 (14,15) and clone MB1.2, which reacts with all forms of integrin ß1 (16) (the latter a kind gift of Dr Bosco Chan, Department of Microbiology and Immunology, University of Western Ontario, London, Canada)] and integrin ß4 [clone 346-11A (BD PharMingen)]; fluorescein isothiocyanate-conjugated mAb against human integrin {alpha}6 (clone GoH3, BD PharMingen); rabbit mAb against Ki67 (clone SP6, Lab Vision Corp, Fremont, CA); rat mAb against mouse F4/80 (macrophage/monocyte-specific antigen; clone A3-1, Serotec, Raleigh, NC) and mouse mAbs against {alpha}-tubulin (clone B-5-1-2) and ß-actin (clone AC15) (Sigma, St. Louis, MO). Rabbit antibodies against nidogen and laminin-5 were the kind gift of Dr Takako Sasaki (Department of Biochemistry and Molecular Biology, Oregon Health & Science University, Portland, OR).

Real-time polymerase chain reaction
Total RNA was isolated from primary keratinocytes using TRIzol reagent (Invitrogen Corporation, Carlsbad, CA). DNase I (Ambion, Austin, TX)-treated total RNA (0.5 µg) was reverse transcribed using the SuperScriptTM III First-Strand Synthesis System for Reverse Transcriptase–PCR (Invitrogen Corporation) with random primers according to the supplier’s protocol. Then synthesized cDNA of Src, Fyn and 18S rRNA was amplified using the SYBR Green PCR Master Mix (Applied Biosystems, FosterCity, CA) and 7500 Real-Time PCR System (Applied Biosystems) in a 25 µl reaction volume. The following primers were used: Src forward, 5'-TCCCGCACCCAGTTCAAC-3' and Src reverse, 5'-GACACAGGCCATCAGCATGT-3'; Fyn forward, 5'-AGGTGCGAAGTTTCCCATTAAG-3' and Fyn reverse, 5'-TCAGACTTGATTGTGAACCTTCCA-3' and 18S rRNA forward, 5'-GGCGACGACCCATTCG-3' and 18S rRNA reverse, 5'-ACCCGTGGTCACCATGGTA-3'. PCRs were performed as follows: 95°C for 10 min, 35 cycles of 95°C for 15 s and 60°C for 1 min. The amplicons were purified after gel electrophoresis using a QIAquick Gel Extraction Kit (Qiagen, Valencia, CA). The DNA concentration of purified amplicons was spectrophotometrically quantified at 260 nm. These amplicons were used as a standard for real-time PCR analysis. Real-time PCR was performed under the same condition as described above. All samples were analyzed in triplicate. After the reactions, the specificity of amplifications in each sample was confirmed by dissociation analysis showing that each sample gave a single melting peak [Src (79.96 ± 0.07°C), Fyn (79.02 ± 0.06°C) and 18s rRNA (80.73 ± 0.07°C): mean ± SE, n = 8]. Then the standard curve (17) was prepared for each gene using serial dilutions of amplicons (0.2 fg to 200 ng), and the amounts of cDNA in each gene were calculated using sequence detection software (Applied Biosystems). In each cDNA sample, the amount of Src and Fyn was normalized to that of 18S rRNA.

Isolation of primary keratinocytes
Murine primary epidermal keratinocytes from 3- or 4-day old mice were isolated according to the method described previously (18), with a slight modification. Briefly, the mouse skin was removed and incubated overnight in defined keratinocytes-SFM (Gibco/Invitrogen, Grand Island, NY) with 2.5 U/ml dispase II (Roche Diagnostics, Indianapotis, IN) and antibiotics at 4°C. The epidermis was separated from the dermis with fine forceps, cut into small pieces and trypsinized (0.25% trypsin/0.1% ethylenediaminetetraacetic acid) for 10 min at 37°C. Then the cells were washed twice with defined keratinocytes-SFM and used for analysis. For adult mouse skin, primary keratinocyte culture was carried out as described (19). Briefly, after shaving hairs on the back, the skin was removed and subcutaneous tissue was scraped off. Then the skin was cut into several small pieces and incubated in phosphate-buffered saline (PBS) containing 0.8% trypsin (Invitrogen Corporation) for 45 min at 37°C. Epidermis was then separated from dermis with fine forceps and incubated in modified Eagle’s medium (MEM) without calcium (Sigma) for 30 min at 37°C with constant shaking. After the filtration of samples through a 70 µm cell strainer (Beckton Dickinson, Bedford, MA), the cells were washed, suspended in MEM without calcium containing 5 µg/ml insulin, 10 ng/ml epidermal growth factor, 10 µg/ml transferrin, 0.36 µg/ml hydrocortisone, 10 µM phosphoethanolamine, 10 µM ethanolamine, 8% chelated fetal bovine serum and 45 µm Ca2+ and plated on a culture dish coated with fibronectin plus collagen. When the cells became subconfluent, they were used for the analysis.

Wounding and preparation of wound tissue
Skin wound healing analysis was performed as described previously (20). Briefly, full-thickness excisional skin wounds (6 mm in diameter) were made in 6- to 8-week old Csk-K5 mice and control Csk(flox/flox) mice, and the rates of wound closure in each group were monitored. Mice were killed 12 h, 24 h, 3 days, 5 days, 7 days and 12 days after wounding (animals: n = 4 in each time point), and 8–10 mm area, including the complete epithelial margins, was collected and used for histopathological analysis.

Histological analysis, immunohistochemistry and immunofluoresecence
For histological analyses, skin samples were either directly frozen in OCT compound (Tissue-Tek, Sakura Finetek, USA Inc., Torrance CA) or fixed in 4% fresh paraformaldehyde in PBS, pH 7.2, overnight and dehydrated in a graded alcohol series and embedded in paraffin. Hematoxylin–eosin staining of 6 µm thick paraffin sections was performed according to the standard protocols. Immunohistochemistry and immunofluorescence studies were performed as described previously (20,21). For immunocytochemical studies, primary keratinocytes were seeded on chamber slides (Nalge Nunc International, Rochester, NY) coated with fibronectin (Sigma) and collagen type I (Upstate Biotechnology, Charlottesville, VA) (10 µg/ml) or Matrigel (growth factor reduced, BD Biosciences, Bedford, MA) (100 µg/ml) and were incubated 18 h. Then the cells were fixed in 4% fresh paraformaldehyde in PBS permeabilized in 0.1% Triton X-100 in PBS, and incubated with the indicated antibodies. Assessment of proliferation in cultured primary keratinocytes was carried out with the Cell Proliferation Enzyme-Linked ImmunoSorbent Assay, 5-bromo-2-deoxyuridine (colorimetric) kit according to the manufacture’s protocol (Roche Diagnostics, Indianapolis, IN).

Fluorescence-activated cell-sorting analysis
Fluorescence-activated cell-sorting analysis was carried out as described previously (22).

Cell lysis and immunoblotting
Freshly isolated primary keratinocytes were used for the analysis. Cell lysis and immunoblotting were performed as described previously (21). Briefly, keratinocytes were lysed with radioimmunoprecipitation, buffer, and total cell lysates were resolved in sodium dodecyl sulfate–polyacrylamide gel electrophoresis gel, transferred onto Immobilon-P polyvinylidene fluoride membrane (Millipore Corp., Billerica, MA) and probed with primary and peroxidase-conjugated secondary antibodies (Jackson ImmunoResearch, West Grove, PA), and then immunoreactive bands were detected using an Enhanced Chemiluminescence Plus detection kit (Amersham/GE Healthcare, Piscataway, NJ). In some immunoblotting analyses, samples were transferred onto Immobilon-FL polyvinylidene fluoride membrane (Millipore Corp.) and probed with primary and IRDye 800CW- or IRDye 680-conjugated secondary antibodies (LI-COR Biosciences, Lincolon, NE). Then immunoreactive bands were detected using the Odyssey Infrared Imaging System (LI-COR Biosciences).

Statistical analysis
The frequency of papillomatous lesions in Csk-K5 mice was analyzed using the two-sided Mann–Whitney U-test. The rest of the statistical data analyses were performed using the unpaired Student’s t-test.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 References
 
Csk expression in the skin and keratinocyte-specific deletion of the csk gene
We generated mice with a keratinocyte-specific deletion of the csk gene using Cre-loxP recombination technology. To obtain homozygously floxed mice expressing cre recombinase [Csk(flox/flox)/K5-Cre+, hereafter called ‘Csk-K5’ or ‘mutant’], we crossed Csk(flox/flox) mice, which were viable and fertile and showed no obvious phenotype (11), with mice expressing the cre gene under the control of the keratin 5 promoter. This promoter has been shown to target the expression of a transgene to basal cells of stratified squamous epithelia and hair follicles (12).

The first step in evaluating offspring of the mating was to perform genotype analysis of genomic DNA prepared from tail biopsies of 7-day-old mice by PCR (Figure 1A). In crossing male Csk(flox/flox)/K5-Cre+ mice and female Csk(flox/flox) mice, a total of 331 pups showed a Mendelian distribution: 91 (27%) Csk(flox/+)/K5-Cre–, 75 (23%) Csk(flox/+)/K5-Cre+, 85 (26%) Csk(flox/flox)/K5-Cre– and 80 (24%) Csk(flox/flox)/K5-Cre+ offspring, suggesting that deletion of Csk in keratin 5-positive cells did not result in embryonic or perinatal lethality, or was there any evidence of an obvious phenotype in newborns.


Figure 1
View larger version (30K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. Disruption of Csk in Csk keratinocyte-specific knockout mice. (A) PCR genotyping performed on tail DNA from offspring of the crossing of male Csk/(flox+)/K5-Cre+ and female Csk(flox/flox) mice. Using 5'-TGGGGTTGAATGGTATGAACACTC-3' and 5'-TGCCATGTGGAGAAGAGAATCAGC-3' primers, PCR may give three products: 579bp for the wild-type allele (WT), 705 bp for the floxed allele (Flox) and 201 bp for the knockout allele (KO). The presence of a Csk-deleted band confirms that Cre-mediated recombination has occurred. Keratin 5-Cre PCR using 5'-GAACATTCCACAGACCTGCA-3' and 5'-CCTTTGAATTGCTGGAACCC-3' primers results in a 292 bp product. (B) Western blot analysis of Csk expression in freshly isolated primary keratinocytes from the back skin of 4-day-old control [Csk(flox/flox) and Csk(flox/+)/K5-Cre+] and mutant [Csk(flox/flox)/K5-Cre+] mice. The primary keratinocytes were lysed in radioimmunoprecipitation buffer, then total cell lysates were resolved in a sodium dodecyl sulfate–polyacrylamide gel electrophoresis, transferred to polyvinylidene fluoride membranes and probed with anti-Csk and anti-ß-actin antibodies. (C) Triple immunofluorescence staining for Csk, integrin {alpha}6 and 4'6-diamidino-2-phenylindole (DAPI, Molecular Probes, Eugene, OR) of the sections from the back skin from 2-week-old control Csk(flox/flox) and Csk-K5 mice. Whereas Csk is expressed in basal keratinocytes and occasionally suprabasal keratinocytes of control epidermis, the epidermis of Csk-K5 mice completely lacked Csk expression. Staining for integrin {alpha}6 indicates the DEJ. Merge, merged image. Bar, 30 µm.

 
We confirmed the absence of Csk protein in epidermis using two different methods: western blot and immunohistochemistry. Csk, detected as a 50 kDa band in keratinocyte lysates from 4-day-old control Csk(flox/flox) mice, was absent in keratinocyte lysates from their mutant (Csk-K5) littermates (Figure 1B). Immunolocalization of Csk in the skin of 2-week-old control mice showed that Csk was expressed in all basal keratinocytes and occasionally suprabasal keratinocytes of interfollicular epidermis and outer root sheath cells in the hair follicle. In contrast, Csk was not detectable in epidermis of Csk-K5 mice (Figure 1C). Integrin {alpha}6 displayed thicker and an irregular staining pattern in the Csk-K5 epidermis compared with the control epidermis (Figure 1C). Fluorescence-activated cell-sorting analysis revealed that the expression of integrin {alpha}6 in Csk-K5 primary keratinocytes was up-regulated compared with control keratinocytes (data not shown). There was no change of the expression pattern of Csk in dermal layers of Csk-K5 skin (data not shown).

Csk-K5 keratinocytes significantly elevates SFK activity in vitro
Since Csk is known as a cytoplasmic tyrosine kinase that negatively regulates the activity of SFKs through phosphorylation of their C-terminal tyrosine (3), to confirm whether Csk indeed plays this role in keratinocytes, we next investigated the expression levels of SFKs and their activation levels. Using antibody SRC2, which detects Src, Fyn, Yes and Fgr (23), we found significantly lower protein levels of SFKs in primary keratinocytes from 4-day-old Csk-K5 mice (Figure 2A). Real-time PCR analysis also showed a significant down-regulation of Src and Fyn mRNA levels in mutant keratinocytes (Figure 2B), suggesting that the lower protein level results from the down-regulation at promoter level and is not a consequence of increased degradation. In contrast, the level of the active form of SFKs [active SFKs, including Src, Fyn, Yes and Fgr; (13)] detected by antibody clone 28 was significantly higher in mutant keratinocytes (Figure 2A). These results are consistent with previous reports that embryonic tissues and fibroblasts from Csk-null mice exhibit elevated kinase activity of Src and Fyn, although their protein levels were lower than in control mice (4,5).


Figure 2
View larger version (49K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Characterization of Csk-K5 primary keratinocytes. (A) Upper panels: western blot analysis of Src, active SFK and total SFK expression in primary keratinocytes from 4-day-old control Csk(flox/flox) and Csk-K5 mice. Anti-Src antibody clone 184Q20, antibody clone 28 (which recognizes active form of SFKs (13)), antibody SRC2 [which recognizes total SFKs (23)] and anti-{alpha}-tubulin were used as the probes. After the intensity of the Src and total SFK bands was normalized to that of the {alpha}-tubulin bands, the levels of Src and total SFKs were found to be significantly lower in Csk-K5 keratinocytes (P < 0.05, Student’s t-test). Lower panel: relative amount of active SFKs in primary keratinocytes from control Csk(flox/flox) (white bar) versus Csk-K5 (black bar) mice. Intensity of the active SFK bands was normalized to that of the total SFK bands, and the relative protein expression levels are shown relative to the control value of 100 (percent of control). Error bars represent standard deviation. *P < 0.05 (Student’s t-test). Note that the level of the active form of SFKs is significantly higher in mutant keratinocytes. (B) Real-time PCR analysis of Src and Fyn mRNA expression in primary keratinocytes from control Csk(flox/flox) (white bars) versus Csk-K5 (black bars) mice. The relative mRNA expression levels are shown relative to the control value of 100 (percent of control). Error bars represent standard deviation. **P < 0.01 (Student’s t-test). (C) Western blot analysis of FAKp397 expression in primary keratinocytes from 16-day-old control Csk(flox/flox) and Csk-K5 mice. Primary keratinocytes were plated on fibronectin/collagen type I (Fn/ColI) or Matrigel (growth factor reduced). One hour after plating, cells were lysed with radioimmunoprecipitation buffer and subsequently used for the analysis. Note that Csk-K5 keratinocytes demonstrate highly up-regulated FAKp397 after the intensity of the FAKp397 bands is normalized to that of the total FAK bands: a 4.5-fold increase on fibronectin/collagen type I and a 5.1-fold increase on Matrigel. (D) Localization of SFK activity in Csk-null keratinocytes. Triple immunofluorescence staining of control Csk(flox/flox) and Csk-K5 primary keratinocytes with Srcp418/F-actin (Alexa488-conjugated phalloidin, Molecular, Probes)/DAPI and FAKp397/F-actin/DAPI 18 h after adhesion to the plates coated with fibronectin/collagen type I or Matrigel showed that active SFK are localized and active in focal contacts. Merge, merged image. Bar, 20 µm.

 
Focal contacts are sites where cells interact with extracellular matrix through integrin receptors. Focal contacts mediate strong adhesion to the substrate and they anchor bundles of oriented actin filaments (stress fibers) through a plaque that consists of many different proteins (24). Cells that have an increased Src activity form atypical focal contacts called podosomes (25). Dynamic changes in podosomes are regulated by both the scaffolding and catalytic activities of Src. Since tyrosine 397 in FAK is a direct phosphorylation site as a substrate of Src (26,27), an increased pY397 FAK staining at podosomes should serve as a marker of higher Src activity. To evaluate the phosphorylation level of tyrosine 397 in FAK during the initial process of focal contact formation, primary keratinocytes from 16-day-old mice were plated on fibronectin/collagen type I or Matrigel (growth factor reduced) and incubated for 1 h. Western blot analysis demonstrated that Csk-K5 keratinocytes displayed highly up-regulated FAKp397 compared with control keratinocytes: a 4.5-fold increase on fibronectin/collagen type I and a 5.1-fold increase on Matrigel (Figure 2C). To determine intracellular localization of the active SFKs, we next performed comparative immunocytochemical analyses of primary keratinocytes from 16-day-old Csk-K5 and control mice; 18 h after cells were seeded on plates coated with fibronectin/collagen type I or Matrigel (Figure 2D). Src autophosphorylated on tyrosine Y418, which indicates an active form of Src kinase, was localized primarily in focal contacts. The main Src substrate in focal contacts is FAK (26,27). Immunocytochemical staining with an antibody that recognizes an active form of FAK, FAKpY397, showed an increased number of focal contacts containing active FAK. These results suggest that activated SFKs are primarily localized in focal contacts.

Csk-K5 skin shows epidermal hyperplasia and papillomatosis, but the lack of Csk does not induce neoplastic changes
The overall appearance of newborn Csk-K5 pups was indistinguishable from that of their Csk(flox/flox) littermates. However, hematoxylin–eosin staining of the skin from 2-week-old mutant mice revealed marked acanthosis with multiple cell layers and hyperkeratosis in the epidermal layer (Figure 3A). At higher magnification, ovoid and irregularly arranged keratinocytes were observed in Csk-K5 epidermis (Figure 3A, arrowheads), whereas basal keratinocytes in control mice showed uniform pattern of cuboidal cells. Although all Csk-K5 mice were still alive 3 months after birth, their skin displayed epidermal hyperplasia and formed multiple skin ulcers (Figure 3B). Infiltration of macrophages/monocytes and deposition of fibronectin in the dermal layer without ulcerative change were also increased in the skin of the mutant mice (Figure 3C and data not shown), indicating that the chronic inflammatory changes had occurred. However, neither foreign body macrophages nor granulomas were found in Csk-K5 dermis (data not shown). There is evidence that transgenic mice that express the constitutively active form of Src in epidermis showed spontaneous induction of squamous cell carcinoma (10). Our in vitro results confirmed that the lack of Csk caused a significant up-regulation of active form of SFKs in keratinocytes (Figure 2A) and raised expectations that the Csk-null epidermis would undergo spontaneous neoplasia, as we hypothesized. In fact, 19% Csk-K5 mice (7 out of 36) at 3 months old developed small tumors up to 1.5 mm in diameter, whereas control mice did not show any tumors. Histological examination revealed that in spite of marked hyperplasia, there was no evidence for neoplastic transformation (skin papillomatosis; Figure 3D). According to the two-sided Mann–Whitney U-test, the occurrence of those papillomatous lesions in Csk-K5 mice were significantly higher than in control mice (P = 0.003). Importantly, the activity of SFKs was high in keratinocytes of these lesions (Figure 3E). Interestingly, Csk-K5 mice did not develop any neoplastic lesions spontaneously up to 8 months after the formation of papillomatous lesions, and those lesions remained of similar size. From these findings, we conclude that, although the lack of Csk caused an up-regulation of SFK activity, it is not sufficient to initiate neoplastic conversion of keratinocytes in vivo.


Figure 3
View larger version (102K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. Csk-K5 mice phenotype. (A) Upper panels: hematoxylin–eosin-stained sections in back skin of 2-week-old mice show epidermal hyperplasia and hyperkeratosis but no obvious abnormality of hair follicles in Csk-K5 skin. Bar, 115 µm. Lower panels: higher magnification of hematoxylin–eosin-stained skin sections from 2-week-old mice clearly shows epidermal hyperplasia with multiple cell layers and hyperkeratosis, and ovoid arranged keratinocytes in Csk-K5 mice (arrowheads). Bar, 30 µm. (B) Epidermal hyperplasia (asterisk) and ulceration in hematoxylin–eosin-stained section of the skin from 3-month-old Csk-K5 mice. Arrow points to the epithelial border of an ulcer tissue. g, granulation tissue in ulcer bed. Bar, 170 µm. (C) Immunostaining for F4/80 (macrophage/monocyte-specific antigen) in back skin of 3-month-old control and mutant mice. Infiltration is prominent in the dermal layer of mutant mice [staining of the sebaceous glands (arrows) is the background staining]. Bar, 40 µm. (D) Hematoxylin–eosin-stained section of papillomatous lesion from the Csk-K5 mice. Bar, 280 µm. (E) Immunohistochemical staining for the active form of SFKs in the papillomatous lesion from the Csk-K5 mice and normal back skin from the control Csk(flox/flox) mice using antibody clone 28 which recognizes only active forms of SFKs. Note that keratinocytes in the papillomatous lesion highly express the active form of SFKs (active SFKs, upper left panel), whereas keratinocytes in the control skin do not show any prominent expression of these kinases (upper right panel). To prevent binding of the anti-mouse secondary antibody to endogenous tissue immunoglobulin, a mouse M.O.MTM kit (Vector, Burlingame, CA) was used for the immunostaining procedures. No binding of the anti-mouse secondary to tissue immunoglobulin is confirmed by the immunohistochemical staining of the same lesion without primary antibody (reactions with biotinylated anti-mouse IgG and avidin biotin complex) (lower left panel). Bar, 30 µm.

 
An increased proliferation of basal keratinocytes and a delay in terminal differentiation of suprabasal keratinocytes underlie hyperplasia and papillomatosis of Csk-K5 skin
To address the mechanisms underlying epidermal hyperplasia and papillomatosis in Csk-K5 epidermis, we first addressed whether the thickened epidermis in Csk-K5 mice had resulted from hyperproliferation of keratinocytes. Comparison of mitotically active Ki67-positive cells in the basal layer of epidermis from 2-week-old Csk-K5 and control mice showed a significantly higher number of proliferating Csk-null basal keratinocytes (P = 0.0004; Figure 4A), suggesting an abnormal proliferation rate. In vitro proliferation assays using primary keratinocytes revealed that the proliferation rate in Csk-K5 keratinocytes was significantly faster than in control keratinocytes (P = 0.0003; data not shown). To distinguish whether Ki67-positive keratinocytes divide in vivo at a similar pace in Csk-K5 as in normal littermates, we conducted wound-healing experiments and found no significant difference in wound closure between control and Csk-K5 mice wounds (Figure 4B). Three days after wounding, there was no obvious difference in re-epithelialization of the wounded skin. Seven days after wounding, re-epithelialization was completed, the epidermis of each group was of similar thickness, and granulation tissue was formed underneath in both control and Csk-K5 wounds. These data suggested that the proliferation rate of Csk-null keratinocytes is not elevated in vivo. Therefore, the higher number of Ki67-positive basal keratinocytes and the thicker epidermis in Csk-K5 mice might be a result of altered differentiation.


Figure 4
View larger version (76K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Keratinocyte proliferation and differentiation are altered in epidermis of Csk-K5 mice. (A) The number of Ki67-positve basal keratinocytes was significantly higher in back skin of 2-week-old mutant (black bar) than control (white bar) mice. Data are presented as number of positive cells (%, calculated out of 100 basal keratinocytes) from the sections immunostained for Ki67. Error bars represent standard deviation. *P < 0.01 (Student’s t-test). (B) Skin wound healing analysis. Hematoxylin–eosin-stained sections from wounded areas 24 h, 3 days and 7 days after wounding. Skin wounds healed normally in mice without Csk. Arrows indicate the epithelial border. es, eschar; g, granulation tissue; re, re-epithelialization. Bar, 115 µm. (C) Double immunofluorescence staining for nidogen and activated integrin ß1 using antibody clone 9EG7 (upper panels) and nidogen and total integrin ß1 using antibody clone MB1.2 (lower panels) in back skin from 2-week-old control and mutant mice. Note that activated integrin ß1-positive cells are found not only in the basal layer but also in the suprabasal layer of Csk-K5 epidermis (arrow heads). Bar, 30 µm. (D) Double immunofluorescence staining for keratin 14 (K14), keratin 10 (K10), keratin 6 (K6), loricrin and {alpha}6 integrin ({alpha}6) in back skin from 2-week-old control and mutant mice. Bar, 30 µm. (E) Double immunofluorescence staining for integrin ß4 and DAPI and staining for laminin-5 (Ln-5) in back skin from 2-week-old control and mutant mice. Bar, 30 µm.

 
Epidermis is renewed from epidermal stem cells. Although the specific surface markers are not yet known, these stem cells express high levels of active integrin ß1 and are confined to the basal layer of epidermis (28–30). Immunostaining with antibody 9EG7, which recognizes only activated integrin ß1 (14,15), showed that integrin ß1-positive cells in the mutant epidermis were present not only in the basal but also in the suprabasal layer with an increase in number (Figure 4C), suggesting that the loss of Csk might affect differentiation at the level of stem cells in the interfollicular epidermis. To determine whether the abnormal distribution of activated integrin ß1-positive cells in mutant epidermis is linked to an abnormal differentiation of keratinocytes, we next examined the distribution of several differentiation markers by immunohistochemistry (Figure 4D). Keratin 14 was present in all epidermal layers of the mutant skin. Keratin 10, which is a marker of the suprabasal keratinocytes, was expressed in all hyperthickened suprabasal layers of mutant skin (Figure 4D), suggesting that hyperplasia is probably associated with an enlarged suprabasal keratinocyte compartments. The terminal differentiation marker loricrin was detected in more layers in Csk-K5 than in control littermates (Figure 4D), suggesting abnormal keratinocyte differentiation. Keratin 6, which is normally not expressed in the interfollicular epidermis, is considered a marker for pathological conditions such as inflammation and epidermal hyperproliferation (31,32). Whereas completely absent in control littermates, keratin 6 was strongly expressed throughout the epidermis of Csk-K5 mice (Figure 4D).

Since SFKs are involved in signaling from adhesive structures at the dermal–epidermal junction (DEJ) (33), we examined the DEJ in Csk-K5 mice. The expression of integrins {alpha}6 and ß4 (Figure 4D and E) DEJ markers nidogen and laminin-5 (Figure 4C and E) displayed an unusual broad zigzag staining pattern in CSK-K5 mice, whereas the expression pattern of these molecules was restricted to the DEJ in control epidermis.


   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
References
 
The role of Csk in epithelial cells has not been well studied. In the present study, we generated mice lacking Csk in all layers of epidermis. Mice lacking Csk in keratin 5-expressing cells were viable and showed no increased mortality. We found the following: (i) Csk-null keratinocytes show high levels of SFK activity both in vivo and in vitro, (ii) all Csk-K5 mice displayed epidermal hyperplasia and hyperkeratosis, as well as signs of chronic inflammation in dermis, (iii) although 19% of the Csk-K5 mice developed papillomatous lesions, none of them showed signs of malignant transformation in the 8-month follow-up period and (iv) high numbers of cells expressing activated ß1-integrin, a stem cell marker, and their unusual presence in suprabasal layers as well as their abnormal pattern of differentiation markers suggest that abnormal differentiation starting at the level of interfollicular epidermis may account for the hyperplastic phenotype.

The complete knockout of the csk gene in mice was embryonic lethal due to an increase in Src family activity and hyperactivation of many signaling pathways (4,5), suggesting that hyperactivation of these pathways in our Csk-K5 mice leads to epidermal hyperproliferation. In fact, we found high numbers and abnormal localization of cells expressing activated ß1-integrin, known as one of the stem cell markers, in epidermis of Csk-K5 mice (Figure 4C). Epithelial skin stem cells found in the bulge are able to self-renew and are thought to contribute to all lineages, including epidermis (28–30). Thus, our findings suggest a regulatory role of Csk in the differentiation of epidermal stem cells at the level of interfollicular epidermis.

We found an unusual broad zigzag staining pattern of the DEJ markers nidogen and laminin-5 in CSK-K5 mice (Figure 4C and E). There is evidence that SFKs are involved in signaling from adhesive structures at the DEJ (33). Thus, an abnormally assembled basement membrane suggests that Csk and/or SFK activity play a role not only in outside in but also in inside out signaling at DEJ adhesive structures. This idea is supported with our finding that active SFKs are localized predominantly in focal adhesions of Csk-null keratinocytes (Figure 2C).

Deletion of Csk in the epidermal keratinocytes induced an influx of inflammatory cells (mainly macrophages) in the dermal layer even when ulcers did not form. In parallel, or more likely caused by the inflammation, increased deposition of fibronectin was observed in the dermal layer of mutant mice. Since we need to find out multinucleated foreign body giant cells as phagocytotic responses to chronic inflammation, we found no evidence for bacterial or fungal infection, suggesting that the inflammation is not caused by an infection. The chronic inflammatory phenotype observed in the dermal layer of Csk-K5 mice mimics inflammatory disorders such as psoriasis. We also found strong induction of keratin 6, which is normally not expressed in the interfollicular epidermis and is considered a marker for pathological conditions such as inflammation and epidermal hyperproliferation (31,32), throughout the epidermis of Csk-K5 mice (Figure 4D). Induction of keratin 6 is also reported in a keratin 14-Fyn transgenic mice epidermis (8). Keratin 14-Fyn transgenic mice demonstrate a thickened, hyperplastic and scaly epidermis dependent on increased Fyn expression. Taken together, these findings suggest that elevated Fyn activity may account for the hyperplasia in the Csk-null epidermis. However, keratin 14-Fyn mice do not develop any spontaneous epidermal tumors followed for 6 months.

The absence of spontaneous neoplasia in Csk-K5 mice is unexpected. Over-expression of a constitutively active form of Src in skin epidermis can induce spontaneous squamous cell carcinoma (9,10). Increased activity of SFKs is present in many human neoplasms, including colonic and breast carcinomas (34–36), and such an increased activity in tumors could theoretically result from activating mutations, impairment of down-regulatory mechanisms or both. However, activating mutations of SFKs are rare in these carcinomas (37,38). This evidence raises an argument that the spontaneous neoplasia observed in the mouse model over-expressing Src with constitutively activating mutations may be due to the supraphysiological levels of the kinase activity per se and that the impaired down-regulation of activated SFKs could account for increased tumoral SFK activities. More precise characterization of the negative regulatory mechanisms of activated SFKs may provide insights into carcinogenesis. Therefore, to address this, we generated an in vivo mouse model with a keratinocyte-restricted deletion of csk. However, it should be noted that we have not found any experimental evidence so far (up to 11 months) that Csk deficiency itself results in spontaneous development of either benign or malignant skin tumors, although Csk was believed to play an important role in the process of oncogenic transformation by negative regulation of SFKs (3). This finding suggests that increased SFK activity, and not expression, is not the genetic basis for the spontaneous development of either benign or malignant skin tumors.

Our current results advance the understanding of the in vivo complexity of the regulatory mechanisms of SFKs. The keratinocyte-specific deletion of Csk could be a promising model system to investigate further the molecular mechanisms underlying keratinocyte proliferation and differentiation by regulating SFK activity in vivo.


   Footnotes
 
{dagger} These authors contributed equally to this work. Back


   Acknowledgments
 
We thank Dr David Schlaepfer, Dr Bosco Chan and Drs Takako Sasaki and Rupert Timpl for antibodies, Drs Edward Maytin and Melissa Piliang for their valuable discussions, Dr Shin Hasegawa for his generous support and Ms Christine Kassuba for editorial assistance. We gratefully acknowledge the support from The Cleveland Clinic to Dr T.S.

Conflict of Interest statement: None declared.

Note Added in Proof

When this manuscript was under review, another group published a report further supporting our findings (Yagi et al (2007) EMBO (Eur. Mol. Biol. Organ.) J., 26, 1234–1244).


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 

  1. Brown MT, et al. Regulation, substrates and functions of Src. Biochim. Biophys. Acta (1996) 1287:121–149.[Medline]
  2. Thomas SM, et al. Cellular functions regulated by Src family kinases. Annu. Rev. Cell Dev. Biol. (1997) 13:513–609.[CrossRef][Web of Science][Medline]
  3. Yeatman TJ. A renaissance for SRC. Nat. Rev. Cancer (2004) 4:470–480.[CrossRef][Web of Science][Medline]
  4. Imamoto A, et al. Disruption of the csk gene, encoding a negative regulator of Src family tyrosine kinases, leads to neural tube defects and embryonic lethality in mice. Cell (1993) 73:1117–1124.[CrossRef][Web of Science][Medline]
  5. Nada S, et al. Constitutive activation of Src family kinases in mouse embryos that lack Csk. Cell (1993) 73:1125–1135.[CrossRef][Web of Science][Medline]
  6. Cabodi S, et al. A PKC-{eta}/Fyn-dependent pathway leading to keratinocyte growth arrest and differentiation. Mol. Cell (2000) 6:1121–1129.[CrossRef][Web of Science][Medline]
  7. Calautti E, et al. Fyn tyrosine kinase is a downstream mediator of Rho/PRK2 function in keratinocyte cell-cell adhesion. J. Cell Biol. (2002) 156:137–148.[Abstract/Free Full Text]
  8. Li W, et al. Srcasm corrects Fyn-induced epidermal hyperplasia by kinase-downregulation. J. Biol. Chem. (2007) 282:1161–1169.[Abstract/Free Full Text]
  9. Matsumoto T, et al. Overexpression of a constitutively active form of c-src in skin epidermis increases sensitivity to tumor promotion by 12-O-tetradecanoylphorbol-13-acetate. Mol. Carcinog. (2002) 33:146–155.[CrossRef][Web of Science][Medline]
  10. Matsumoto T, et al. Targeted expression of c-Src in epidermal basal cells leads to enhanced skin tumor progression, promotion, malignant progression, and metastasis. Cancer Res. (2003) 63:4819–4828.[Abstract/Free Full Text]
  11. Schmedt C, et al. Csk controls antigen receptor-mediated development and selection of T-lineage cells. Nature (1998) 394:901–904.[CrossRef][Medline]
  12. Ramirez A, et al. A keratin K5Cre transgenic line appropriate for tissue-specific or generalized Cre-mediated recombination. Genesis (2004) 39:52–57.[CrossRef][Web of Science][Medline]
  13. Kawakatsu H, et al. A new monoclonal antibody which selectively recognizes the active form of Src tyrosine kinase. J. Biol. Chem. (1996) 271:5680–5685.[Abstract/Free Full Text]
  14. Lenter M, et al. A monoclonal antibody against an activation epitope on mouse integrin chain ß1 blocks adhesion of lymphocytes to the endothelial integrin {alpha}6ß1. Proc. Natl Acad. Sci. USA (1993) 90:9051–9055.[Abstract/Free Full Text]
  15. Von Ballenstrem CG, et al. VLA-ß1 integrin subunit-specific monoclonal antibodies MB1.1 and MB1.2: binding to epitopes not dependent on thymocyte development or regulated by phorbol ester and divalent cations. Hybridoma (1996) 15:125–132.[Web of Science][Medline]
  16. Ilic D, et al. FAK promotes organization of fibronectin matrix and fibrillar adhesions. J. Cell Sci. (2004) 117:177–187.[Abstract/Free Full Text]
  17. Giulietti A, et al. An overview of real-time quantitative PCR: applications to quantify cytokine gene expression. Methods (2001) 25:386–401.[CrossRef][Web of Science][Medline]
  18. Calderai R, et al. Long-term culture of murine epidermal keratinocytes. J. Invest. Dermatol. (2000) 114:1064–1065.[CrossRef][Web of Science][Medline]
  19. Romero MR, et al. Analysis of cultured keratinocytes from a transgenic mouse model of psoriasis: effects of suprabasal integrin expression on keratinocyte adhesion, proliferation and terminal differentiation. Exp. Dermatol. (1999) 8:53–67.[Web of Science][Medline]
  20. Sakai T, et al. Plasma fibronectin supports neuronal survival and reduces brain injury following transient focal cerebral ischemia but is not essential for skin wound healing and hemostasis. Nat. Med. (2001) 7:324–330.[CrossRef][Web of Science][Medline]
  21. Sakai T, et al. Integrin-linked kinase (ILK) is required for polarizing the epiblast, cell adhesion and controlling actin accumulation. Genes Dev. (2003) 17:926–940.[Abstract/Free Full Text]
  22. Sakai T, et al. Modulation of ß1A integrin function by tyrosine residues in the ß1 cytoplasmic domain. J. Cell Biol. (1998) 141:527–538.[Abstract/Free Full Text]
  23. Liang Y, et al. State and role of SRC family kinases in replication of herpes simplex virus 1. J. Virol. (2006) 80:3349–3359.[Abstract/Free Full Text]
  24. Geiger B, et al. Transmembrane extracellular matrix-cytoskeleton crosstalk. Nat. Rev. Mol. Cell Biol. (2001) 2:793–805.[CrossRef][Web of Science][Medline]
  25. Frame MC. Newest findings on the oldest oncogene; how activated src does it. J. Cell Sci. (2004) 117:989–998.[Abstract/Free Full Text]
  26. Schlaepfer DD, et al. Control of motile and invasive cell phenotypes by focal adhesion kinase. Biochim. Biophys. Acta (2004) 1692:77–102.[Medline]
  27. Mitra SK, et al. Focal adhesion kinase: in command and control of cell motility. Nat. Rev. Mol. Cell Biol. (2005) 6:56–68.[CrossRef][Web of Science][Medline]
  28. Alonso L, et al. Stem cells of the skin epithelium. Proc. Natl Acad. Sci. USA (2003) 100:11830–11835.[Abstract/Free Full Text]
  29. Essayem S, et al. Hair cycle and wound healing in mice with a keratinocyte-restricted deletion of FAK. Oncogene (2006) 25:1081–1089.[CrossRef][Web of Science][Medline]
  30. Watt FM, et al. Epidermal stem cells: an update. Curr. Opin. Genet. Dev. (2006) 16:518–524.[CrossRef][Web of Science][Medline]
  31. Dominey AM, et al. Targeted overexpression of transforming growth factor {alpha} in the epidermis of transgenic mice elicits hyperplasia, hyperkeratosis and spontaneous, squamous papillomas. Cell Growth Differ. (1993) 4:1071–1082.[Abstract]
  32. Stoler A, et al. Use of monospecific antisera and cRNA probes to localize major changes in keratin expression during normal and abnormal epithelial differentiation. J. Cell Biol. (1988) 107:427–446.[Abstract/Free Full Text]
  33. Mariotti A, et al. EGF-R signaling through Fyn kinase disrupts the function of integrin {alpha}6ß4 at hemidesmosomes: role in epithelial cell migration and carcinoma invasion. J. Cell Biol. (2001) 155:447–458.[Abstract/Free Full Text]
  34. Sakai T, et al. An epitope localized in c-Src negative regulatory domain is a potential marker in early stage of colonic neoplasms. Lab. Invest. (1998) 78:219–225.[Web of Science][Medline]
  35. Ito Y, et al. Activation of c-Src gene product in hepatocellular carcinoma is highly correlated with the indices of early stage phenotype. J. Hepatol. (2001) 35:68–73.[CrossRef][Web of Science][Medline]
  36. Ito Y, et al. Activation of c-Src is inversely correlated with biological aggressiveness of breast carcinoma. Breast Cancer Res. Treat. (2002) 76:261–267.[CrossRef][Web of Science][Medline]
  37. Daigo Y, et al. Absence of genetic alteration at codon 531 of the human c-src gene in 479 advanced colorectal cancers from Japanese and Caucasian patients. Cancer Res. (1999) 59:4222–4224.[Abstract/Free Full Text]
  38. Irby RB, et al. Activating SRC mutation in a subset of advanced human colon cancers. Nat. Genet. (1999) 21:187–190.[CrossRef][Web of Science][Medline]
Received January 10, 2007; revised April 26, 2007; accepted April 30, 2007.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 CarcinogenesisHome page
B. Serrels, A. Serrels, S. M. Mason, C. Baldeschi, G. H. Ashton, M Canel, L. J. Mackintosh, B. Doyle, T. P. Green, M. C. Frame, et al.
A novel Src kinase inhibitor reduces tumour formation in a skin carcinogenesis model
Carcinogenesis, February 1, 2009; 30(2): 249 - 257.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
28/10/2074    most recent
bgm112v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Honda, K.
Right arrow Articles by Sakai, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Honda, K.
Right arrow Articles by Sakai, T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?