Bone morphogenetic protein signaling regulates ... - Semantic Scholar

2 downloads 0 Views 1MB Size Report
Nov 28, 2006 - ... Andrei N. Mardaryev*, Alice Tommasi di Vignano†, Ruzanna Atoyan*, ...... Hopping SB, Brady JN, Udey MC, Vogel JC (2006) J Clin Invest ...
Bone morphogenetic protein signaling regulates the size of hair follicles and modulates the expression of cell cycle-associated genes Andrey A. Sharov*, Tatyana Y. Sharova*, Andrei N. Mardaryev*, Alice Tommasi di Vignano†, Ruzanna Atoyan*, Lorin Weiner†, Shi Yang‡, Janice L. Brissette†, G. Paolo Dotto†§, and Vladimir A. Botchkarev*¶储 Departments of *Dermatology and ‡Pathology and Laboratory Medicine, Boston University School of Medicine, Boston, MA 02118; †Cutaneous Biology Research Center, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA 02129; §Department of Biochemistry, Lausanne University, CH-1066 Lausanne, Switzerland; and ¶Medical Biosciences, School of Life Sciences, University of Bradford, Bradford BD7 1DP, United Kingdom Communicated by Elaine Fuchs, The Rockefeller University, New York, NY, October 8, 2006 (received for review February 7, 2006)

Bone morphogenetic protein (BMP) signaling is involved in the regulation of a large variety of developmental programs, including those controlling organ sizes. Here, we show that transgenic (TG) mice overexpressing the BMP antagonist noggin (promoter, K5) are characterized by a marked increase in size of anagen hair follicles (HFs) and by the replacement of zig-zag and auchen hairs by awl-like hairs, compared with the age-matched WT controls. Markedly enlarged anagen HFs of TG mice show increased proliferation in the matrix and an increased number of hair cortex and medulla cells compared with WT HFs. Microarray and real-time PCR analyses of the laser-captured hair matrix cells show a strong decrease in expression of Cdk inhibitor p27Kip1 and increased expression of selected cyclins in TG vs. WT mice. Similar to TG mice, p27Kip1 knockout mice also show an increased size of anagen HFs associated with increased cell proliferation in the hair bulb. Primary epidermal keratinocytes (KC) from TG mice exhibit significantly increased proliferation and decreased p27Kip1 expression, compared with WT KC. Alternatively, activation of BMP signaling in HaCaT KC induces growth arrest, stimulates p27Kip1 expression, and positively regulates p27Kip1 promoter activity, thus further supporting a role of p27Kip1 in mediating the effects of BMP signaling on HF size. These data suggest that BMP signaling plays an important role in regulating cell proliferation and controls the size of anagen HFs by modulating the expression of cell-cycleassociated genes in hair matrix KC. Noggin 兩 proliferation 兩 skin

S

kin morphogenesis leads to the formation of an organ that covers and protects the body from external insults and also results in the development of the hair follicles (HFs) that generate hairs, whose phenotype (length, thickness, shape, and color) varies substantially between distinct anatomical sites of the mammalian body (1–5). Hair fiber production is regulated by epithelial–mesenchymal interactions between the HF epithelium and mesenchyme and occurs only during the growing phase of the hair cycle (anagen), whereas it ceases during HF involution (catagen) and relative resting (telogen; refs. 2 and 6–9). Previous observations link variability in hair length and thickness with such parameters as duration of the anagen phase of the hair cycle, as well as with the size of the hair bulb and follicular papilla (FP; refs. 10–12). In particular, it was shown that the size of human HFs and hair diameter strongly correlates with the volume and number of cells in the FP, thus suggesting an important role for the FP in the control of HF size and hair phenotype (10, 12). In murine dorsal skin, HFs are grouped into four principal types, each characterized by distinct size of the hair bulb and phenotype of the hairs generated (4, 5). Guard or tylotrich HFs that represent ⬇5–10% of the HFs have the largest hair bulb compared with other HF types and produce long straight hairs. Awl and auchen HFs show intermediate-sized hair bulbs; however, awl HFs produce straight hairs, whereas auchen 18166 –18171 兩 PNAS 兩 November 28, 2006 兩 vol. 103 兩 no. 48

hairs have a single contraction. Zig-zag HFs that comprise ⬇50–70% of the HFs have the smallest hair bulbs and produce relatively short hairs with two or three contractions. Hair-fiber thickness and shape strongly correlate with the number and size of keratinocytes (KC) in the hair medulla, which in straight hairs (guard and awl) remains constant over the entire hair length, except where reduced at its proximal and distal ends (4, 5, 13, 14). In auchene and zig-zag hairs, the number and size of the medullar KC are substantially reduced at the contraction areas, thus resulting in local hair shaft thinning and bending (Fig. 1J). The thickness of the hair medulla is controlled by the activity of the Edar, IGF, and FGF signaling pathways (14–19); however, the downstream targets and mechanisms involved in regulating hair shape and formation of hair shaft bending remain largely unknown. Bone morphogenetic protein (BMP) signaling plays important roles in the regulation of the size and phenotype of ectodermal organs (20–23). BMPs exert their effects by binding to specific BMP receptors, which transduce signals to the nucleus by recruiting Smad1/5 transcription regulators or components of the MAPK pathway (24). BMP signaling is regulated by extracellular antagonists, including noggin, which modulate binding of the selected BMPs to BMP receptors (25). BMP signaling is critically important for proper control of HF initiation and KC differentiation into the follicle-specific cell lineages, as well as for hair pigmentation (26–33). In this paper, we provide evidence that BMP signaling also plays an important role in the control of HF size and hair phenotype by regulating cell proliferation and modulating the expression of cell-cycle-associated genes in hair matrix KC. Results Noggin Overexpression Results in an Increase of HF Size and Leads to Transformation of Zig-Zag and Auchene Hairs into Awl-Like Hairs. To

explore the role of BMP signaling in the regulation of HF size, K5-Noggin (TG) mice were generated, as described (33, 34). Together with phenotypes described previously, such as retardation of the eyelid opening and hair darkening, TG mice showed progressive increase in the size of anagen HFs, which reach their maximum in 5- to 6-month-old animals (Fig. 1 A–F). Consistently with previous observations (4), anagen VI skin in WT mice contained ⬇65% HFs with small hair bulb diameters Author contributions: A.A.S. and T.Y.S. contributed equally to this work; V.A.B. designed research; A.A.S., T.Y.S., A.N.M., A.T.d.V., R.A., and L.W. performed research; A.A.S., T.Y.S., A.N.M., A.T.d.V., L.W., S.Y., J.L.B., and G.P.D. contributed new reagents/analytic tools; A.A.S., T.Y.S., A.N.M., R.A., and V.A.B. analyzed data; and V.A.B. wrote the paper. The authors declare no conflict of interest. Abbreviations: BMP, bone morphogenetic protein; FP, follicular papilla; HF, hair follicle; KC, keratinocyte; TG, transgenic; Cdk, cyclin-dependent kinase. 储To

whom correspondence should be addressed. E-mail: [email protected].

© 2006 by The National Academy of Sciences of the USA

www.pnas.org兾cgi兾doi兾10.1073兾pnas.0608899103

(60–89 ␮m, presumably those that generate zig-zag hairs), ⬇30% HFs with intermediate-sized hair bulbs (90–129 ␮m in diameter, presumably generating awl and auchen hairs), and ⬇5% HFs with large hair bulbs (130–180 ␮m in diameter), most likely producing guard hairs (Fig. 1G). In contrast to WT mice, TG mice showed lack of HFs with small-sized hair bulbs and a marked increase in the number of intermediate HFs (89.1 ⫾ 8.3%), whereas the proportion of large HFs was not changed significantly (Fig. 1G). Furthermore, TG mice were characterized by remarkable changes in hair phenotype and showed a replacement of the zig-zag and auchene hairs by awl-like hairs, in which the lack of contractions was associated with significantly increased thickness and number of the cortex and medulla cells compared with WT zig-zags (Fig. 1 H–L; Fig. 5A, which is published as supporting information on the PNAS web site). However, cell number and thickness of the inner root sheath cells were similar in WT and TG HFs (Fig. 5 A, D, and E). In contrast to zig-zag and auchene hairs, guard hairs of TG and WT mice did not show differences in their diameters or in the number of the cortex or medulla cells (Fig. 1 H–L; data not shown). Interestingly, the marked enlargement of the hair bulb seen in the majority of the HFs in TG mice was not accompanied by any changes in overall size of the FP or in number of the FP cells, as well as in size of the hair matrix KC (Fig. 5 B, C, and F–L). In large HFs of WT and TG mice, size and number of the FP cells were also quite similar, whereas they remained significantly higher than the corresponding parameters in the small/ intermediate HFs (Fig. 5 B, C, and F–I). These data suggested that transformation of the small HFs into the intermediate type in TG mice is associated with a marked increase of cell number in the hair matrix region, hair cortex, and medulla, and is not accompanied by any changes in size or cell number in the FP. HF in K5-Noggin Mice Show Increased KC Proliferation and Markedly Decreased Expression of Cyclin-Dependent Kinase (Cdk) Inhibitor p27Kip1. To elucidate the molecular mechanisms underlying the

marked increase of HF size in TG mice, and to define whether similar mechanisms may contribute to regulation of the HF size in WT mice, laser-capture microdissection of the HF matrix and Sharov et al.

FP areas from small or intermediate anagen VI HFs of WT mice and from intermediate HFs of TG mice were performed (Fig. 2 A and B). After laser-capture microdissection, RNA samples were processed for microarray and real-time PCR analyses, as described (35). Consistently with data published previously (36, 37), RNA samples obtained by laser-capture microdissection showed expression of the hair matrix and FP-specific genes, which was not changed after two rounds of RNA amplification (Fig. 2C; Fig. 6A, which is published as supporting information on the PNAS web site,). In concordance with our previous observations (33), anagen VI HFs in TG mice showed decreased expression of phospho-Smad1/5 in the outer root sheath and in the lower parts of the hair matrix and FP compared with the small HFs of WT mice (Fig. 2H), thus suggesting that transgenederived noggin may indeed influence the magnitude of BMPSmad signaling in hair matrix KC leading to its moderate decrease, compared with WT mice. Interestingly, intermediate HFs in WT mice also showed reduced pSmad1/5 expression in the hair matrix compared with small HFs (Fig. 2H). Microarray analysis of the hair matrix and FP of small and intermediate HFs of WT mice and intermediate HFs of TG mice showed 2-fold or higher changes in expression of 175 genes that encode adhesion/extracellular matrix molecules; cytoskeleton/ cell motility markers; and molecules involved in the control of cell differentiation, metabolism, signaling, and transcription (Fig. 2 D–J; Tables 1 and 2, which are published as supporting information on the PNAS web site). Most remarkable differences between small HFs of WT mice and intermediate HFs of both WT and TG mice were seen in the expression of genes involved in the control of cell proliferation (Fig. 2D and Table 1). By real-time PCR, expression of Cdk inhibitor p27Kip1 showed a ⬎10-fold decrease in hair matrix of TG HFs, whereas expression of the selected cyclins (A2, D1, E2, and H) was moderately increased compared with the small HFs of WT mice (Fig. 2D). Similarly to TG HFs, intermediate HFs of WT mice showed decreased expression of Cdk inhibitors p21Cip1, p27Kip1, and p57 and increased expression of cyclins, compared with small WT HFs (Fig. 2D). Consistently with these results, intermediate HFs of both WT and TG mice showed a significantly (P ⬍ 0.05) increased number PNAS 兩 November 28, 2006 兩 vol. 103 兩 no. 48 兩 18167

DEVELOPMENTAL BIOLOGY

Fig. 1. Increase of HF size in K5-Noggin mice is accompanied by increase of cell number in the hair matrix, cortex, and medulla. At distinct time points of postnatal development (postnatal day 6.5, 10 –12 weeks, 6 months), hairs and dorsal skin of WT and TG mice were harvested. Skin cryosections were processed for histoenzymatic visualization of alkaline phosphatase (A–F), and histomorphometric analyses of the HFs and plucked hairs were performed (G–L). Statistical analyses were performed by using Student’s t test, *, P ⬍ 0.05; **, P ⬍ 0.01; ***, P ⬍ 0.001. (A–F) Progressive increase of the hair bulb size (arrows) in TG vs. WT mice. (G) Lack of small HFs and significant (P ⬍ 0.001) increase in proportion of intermediate HFs in TG mice versus WT mice. (H and I) Replacement of zig-zag and auchene hairs in TG mice by awl-like hairs. (J, K, and L) Awl-like hairs of TG mice show significantly increased diameter (P ⬍ 0.01) and number of medulla cells (P ⬍ 0.05), compared with zig-zag hairs of WT mice. G, guard; SC, subcutis; Zz, zig-zag.

Fig. 2. Molecular analyses of anagen HFs from WT and TG mice suggest differences in expression of cell-cycle-associated genes. Skin of 12-week-old WT and TG mice was synchronized at anagen VI stage of the hair cycle by depilation. Skin was harvested, and cryosections were processed for laser-capture microdissection, RNA isolation and amplification, microarray and real-time PCR analyses (A–G), and for immunohistochemistry (H–J). (A and B) Anagen VI HF before and after laser-capture microdissection. (C) RT-PCR of the hair matrix- and FP-specific genes in RNA samples obtained from the corresponding areas of anagen VI HFs of WT and TG mice. (D–G) Real-time PCR of genes expressed in the hair matrix or FP that show differences between intermediate HFs of TG mice and intermediate or small HFs in WT mice. (H) pSmad1/5 in the hair matrix (large arrows), hair shaft (small arrows), and outer root sheath (arrowheads) in small and intermediate HFs of WT and TG mice. (I) Ectopic appearance of BrdU⫹ (green fluorescence, arrowhead) and Ki-67⫹ cells (red fluorescence, arrowhead) in differentiating cells of the hair shaft in TG mice. (J) Decrease in p27Kip1 expression in hair matrix of intermediate HFs in WT and TG mice (arrows).

of BrdU⫹ cells in the hair bulb associated with the strong decrease of p27Kip1 protein expression compared with small HFs of WT animals (Fig. 2 I and J; Fig. 5M). Also, intermediate HFs of TG mice showed ectopic expression of Ki-67 among differentiating cells located far above the hair matrix (Fig. 2I). Thus, these data revealed that intermediate HFs in both WT and TG mice are characterized by significantly increased proliferation in the hair matrix and by reduced expression of Cdk inhibitors, compared with small WT HFs, suggesting a role for cell cycleassociated proteins in regulating HF size and hair phenotype. In addition to changes in expression of cell cycle-associated genes and in concordance with previous observations (14, 18, 19, 38), our data revealed a decrease in the expression of Igfbp-3/5 and Krox20 in the hair matrix of intermediate HFs of WT mice, 18168 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0608899103

compared with small HFs (Figs. 2 E and F and 6C). Despite a phenotypic similarity to intermediate HFs of WT mice, intermediate HFs in TG mice showed only moderate/minor differences in the expression of genes encoding the components of Edar, FGF, IGF, Shh, and Wnt signaling pathways in the hair matrix, compared with small HFs of WT mice (Figs. 2 E–G and 6 D and E). However, similar to WT mice, expression of the transcription factor Krox-20 showed ⬎6-fold decrease in hair matrix of intermediate HFs of TG mice vs. small HFs of WT mice (Fig. 6C). Interestingly, despite the lack of change in noggin expression, intermediate and small HFs of WT mice showed differences in expression of other components of BMP-signaling pathways: BMP and Wnt antagonist Sostdc1 or Wise (39) increased, Sharov et al.

whereas BMP receptor ligand Gdf5, Bmpr-Ib, and Gdf10 decreased in the matrix and FP of intermediate HFs, compared with small HFs (Figs. 2 F and G and 6B). Together with data demonstrating a decreased expression of pSmad1/5 and p27Kip1 in the hair matrix of intermediate HFs vs. small HFs (Fig. 2 D, F, H, and J), these results suggested that Sostdc1, Gdf5/10, and Bmpr-Ib may contribute to the control of cell proliferation and regulating size in these HF types of WT mice. Similarly to K5-Noggin Mice, p27Kip1 Knockout Mice Show Increased Size of Anagen HF and Increased KC Proliferation in the Hair Matrix.

To further explore the role for p27Kip1 in the control of HF size, hair bulb diameter and cell proliferation were compared between anagen VI HFs of p27Kip1 knockout (⫺/⫺) mice and corresponding WT mice. It was shown previously that p27Kip1 knockout mice are characterized by hyperplasia, increased size, and cell number in many organs (40). Despite apparently normal HF morphogenesis and lack of significant alterations in HF cycling (data not shown), p27Kip1 knockout mice contained significantly (P ⬍ 0.05) fewer numbers of small HFs and increased numbers of intermediate HFs compared with WT skin (Fig. 3 A, B, and E). Similar to TG mice, the increased HF size in these mice was associated with ectopic localization of Ki-67positive cells in the precortex area above the hair matrix (Fig. 3 C and D). Furthermore, the proportion of zig-zag hairs in p27Kip1 knockout mice was significantly reduced (40.7 ⫾ 5.9% vs. 63.1 ⫹ 7.3% in WT mice; P ⬍ 0.05), whereas awl hairs were increased (32.3 ⫾ 4.5% vs. 15.9 ⫹ 2.4% in WT mice; P ⬍ 0.05), and the number of cells in their medullae was also significantly higher (P ⬍ 0.05) compared with WT mice (Fig. 3F). Thus, the hair phenotype in p27Kip1 knockout mice showed similarity to K5Noggin mice (Fig. 1), further suggesting that p27Kip1 may indeed play a role in mediating the effects of BMP signaling on cell proliferation and size of anagen HFs. p27Kip1 Expression and Promoter Activity Are Positively Regulated by BMP Signaling. To determine whether BMP signaling is capable

of regulating KC proliferation/differentiation by a p27Kip1dependent mechanism, the proliferation rate was compared between primary epidermal KC from WT and TG mice, as well as between HaCaT cells cultured with BMP-4 or vehicle control. Primary KC from TG mice showed significantly increased proliferation (P ⬍ 0.01) compared with WT KC (Fig. 4A). Alternatively, HaCaT cells after BMP-4 treatment showed significant (P ⬍ 0.01) increase in proportion of cells in G0/G1 phase and decrease of S-phase cells, compared with vehicle control (Fig. 4B). By real-time PCR, expression of Cdk inhibitors p21Cip1 and p27Kip1 significantly decreased (P ⬍ 0.05) in primary KC of TG mice, whereas expression of Cdk2 and cyclin M3 increased, compared with KC of WT mice (Fig. 4C). In HaCaT cells, Sharov et al.

expression of p27Kip1 transcripts substantially increased after BMP-4 treatment, and these BMP-4 effects were blocked by addition of recombinant noggin protein (Fig. 4D). By Western blot analysis, the expression of p27Kip1, as well as phosophoSmad1/5, also increased after incubation of HaCaT cells with BMP-4, whereas expression of the phospo-Rb, a major positive regulator of S-phase entry, decreased compared with the controls (Fig. 4E). Furthermore, cotransfection of HaCaT cells with vectors containing one of the constitutively active BMP receptors and/or pCMVSmad1/Smad5 vectors (41), as well as with p27PF-Luc reporter plasmid containing human p27 promoter (42), resulted in 16- to 17-fold increase of p27PF-Luc activity (Fig. 4F). These data suggest that BMP signaling positively regulates activity of the p27Kip1 promoter and p27Kip1 expression at both mRNA and protein levels in vitro and may indeed promote proliferation/differentiation transition in hair matrix KC in vivo by up-regulating p27Kip1 expression. Discussion Control of organ size is a complex process that requires coordination of intrinsic signals regulating intraorgan proliferation and extrinsic factors, such as nutritional and hormonal influences (43, 44). BMP signaling plays an important role in regulating size and phenotypic variability of ectodermal organs (20–22, 45). Here, we show that BMP signaling is involved in the control of HF size, hair shape, and thickness, at least in part, by regulating the expression of Cdk inhibitor p27Kip1 and cell number in the hair bulb and hair fiber (Figs. 1–4). We show that, in contrast to complete inactivation of BMP receptor IA (BMPR-IA) signaling resulting in a failure of the inner root sheath and hair shaft formation (29–32), moderate inhibition of BMP signaling in K5-Noggin mice causes an increase in cell proliferation and size of the hair bulb, as well as transformation of bended zig-zag and auchene hairs into straight awl-like hairs (Figs. 1 and 2). In zig-zag and auchen hairs of WT mice, number and size of medulla cells are markedly reduced at the areas of contractions, resulting in the local hair shaft thinning and bending (Fig. 1J). It has recently been shown that in the hair bulb, proliferative activity of precursors for distinct cell lineages appears to be different; in contrast to the progenitors forming the inner root sheath or hair cortex/cuticle, medulla progenitor cells divide more than once before their switch to terminal differentiation (46). These data may suggest that proliferative activity of the medulla progenitors in zig-zag and auchen HFs may not be constant, decreasing at certain time points of anagen and leading to the formation of contractions in hair fiber. It was shown previously that medulla progenitor cells express BMPR-IA and pSmad1/5 (28–30), whereas the number of the medulla cells increased in awl-like hairs of K5-Noggin mice (Figs. 1 and 2). Because intermediate (awl) and small (zig-zag) HFs PNAS 兩 November 28, 2006 兩 vol. 103 兩 no. 48 兩 18169

DEVELOPMENTAL BIOLOGY

Fig. 3. Increased size and cell proliferation in anagen HFs of p27Kip1 knockout mice. Dorsal skin of 12-week-old WT and p27Kip1 knockout (⫺/⫺) mice was harvested 12 days after depilation. Skin cryosections were processed for either histoenzymatic visualization of alkaline phosphatase (A and B) or immunofluorescent stainings with anti-Ki-67 antiserum (C and D). Histomorphometric analyses of the HFs and plucked hairs were performed (E and F). Statistical analysis was performed by using Student’s t test, *, P ⬍ 0.05. (A and B) Anagen VI HFs in p27Kip1 ⫺/⫺ mice (arrows) and WT mice. (C and D) Proliferating cells in the matrix (arrows) and ectopic appearance of Ki-67 in differentiating cells above the FP (arrowhead) in the HF of p27Kip1 ⫺/⫺ mice. (E) Decrease of small HFs and increase of intermediate HFs in p27Kip1 ⫺/⫺ mice. (F) Increase in number of medulla cells in awl hairs of p27Kip1 ⫺/⫺ mice.

Fig. 4. Effects of BMP signaling on KC growth arrest, p27Kip1 expression and promoter activity. Primary mouse KC were obtained from newborn WT and TG mice. Human HaCaT KC were incubated for 24 h with rhBMP4, and a combination of BMP4 and noggin or diluent control and were harvested 6 –24 h after beginning the experiment. Flow cytometry of cell cycle (A and B), real-time PCR (C and D), Western blot (E), and promoter assay (F) were performed. (A) A significant increase in number of cells in S phase and decrease of G0/G1 cells in TG vs. WT mice. (B) Significant increase in number of cells in G0/G1 phase and decrease in number of proliferating cells after BMP-4 treatment, compared with control. (C) Real-time PCR of the Cdk2, cyclin M3, p21Cip1, and p27Kip1 transcripts in primary KC of WT and TG mice. (D) Increase in p27Kip1 transcripts in HaCaT cells 6 –24 h after BMP-4 treatment and inhibition of these effects by noggin. (E) Increase in p27Kip1 protein expression, Smad1/5 phosphorylation and decrease of pRb protein in HaCaT cells after BMP-4 treatment. (F) Effects of cotransfection with vectors containing constitutively active BMPR-IA (Alk3QD), BMPR-IB (Alk6QD), pCMVmSmad1, and/or pCMVSmad5 on transcription of the p27PF-Luc reporter plasmid containing human p27 promoter region and pGVB2L-Luc (negative control). (G) Scheme illustrating mechanisms of the effects of BMP signaling on proliferation and differentiation of hair matrix KC.

show differences in expression of endogenous Sostdc1, Gdf5/10 and Bmpr-Ib (Figs. 2 and 6), BMP signaling may contribute to regulation of the hair bulb size, as well to the formation of hair shaft bending in WT mice by controlling fluctuations of proliferative activity of the medulla progenitors in zig-zag and auchen HFs during anagen. Several lines of evidence suggest that the number of cells in the hair medulla is also controlled by the Edar, FGF, and IGF signaling pathways, and that transcription factor Krox20 may be involved in mediating the effects of FGFs and IGFs on formation of hair shaft bending in zig-zag HFs (14–19). Our data demonstrating a decreased expression of Krox20 in the matrix of intermediate HFs of K5-Noggin mice (Figs. 2 and 6) suggest that BMP signaling may also be involved in the control of Krox20 expression in the HF. However, mechanisms that underlie a crosstalk between the BMP and IGF, FGF, or Edar signaling pathways in controlling HF size and hair shaft bending require further elucidation. Also, it is unclear whether a large number of genes encoding adhesion/extracellular matrix molecules, metabolic and proteolytic enzymes, and signaling and transcription regulators that show differences in expression between hair matrix and/or FP of WT and K5-Noggin mice (Tables 1 and 2) may be considered BMP targets and may somehow contribute to remarkable changes in hair phenotype seen in noggin TGs. In particular, genes such as Ptch2, Lef-1, Wnt inhibitor Wif1, which link BMP signaling with Shh and Wnt signaling pathways, as well as Msx-2 transcription factor (47) may certainly play a role in mediating the effects of BMP signaling on KC proliferation/differentiation in the HF. Taken together, our data provide compelling evidence that BMP signaling controls HF size and phenotype of the produced hairs, at least in part by regulation of the expression of cell 18170 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0608899103

cycle-dependent kinase inhibitor p27Kip1 in hair matrix KC (Fig. 4G). These data may be important for further understanding mechanisms of the HF miniaturization seen in a number of hair growth disorders (androgenetic alopecia and alopecia areata), as well as for studying the pathobiology of the HF response to external insults targeting replicating KC, such as chemotherapy and ionizing radiation. Materials and Methods Animals, Tissue Collection, and Morphometric Analyses. K5-Noggin

overexpressing (TG) mice were generated as described (34). Homozygous p27Kip1 knockout mice (B6.129S4-Cdkn1btm1Mlf/J; ref. 40) were purchased from The Jackson Laboratory (Bar Harbor, ME). Hairs were plucked for further analyses from dorsal skin of 10-week-old telogen mice. Skin samples were collected from neonatal mice (postnatal day 6.5), from 12-weekold mice at day 12 after depilation (anagen VI), as well as from 6-month-old animals (n ⫽ 5–7 per time points for each mouse strain), as described elsewhere (48). Laser-Capture Microdissection, RNA Amplification, and Microarray Analysis. The distinct areas from WT and K5-Noggin anagen HF

(hair matrix, FP) were dissected by Laser Capture microdissection system (Arcturus, Mountain View, CA), as described (35, 49). Total RNAs isolated after two rounds of linear amplification were processed through microarray analyses by using 41K Whole Mouse Genome 60-mer oligo-microarray (Agilent Technologies, Mogen, St. Louis, MO). All microarray data on gene expression were normalized to the corresponding data obtained from the reference RNA, as described (50). Two independent data sets for each HF areas (hair matrix, FP) were obtained from WT and TG mice, and P values were calculated by the Agilent feature Sharov et al.

Extraction software (version 7.5) using distribution of the background intensity values to signal intensity and Student’s t test.

cent microscope (Nikon, Florham Park, NJ), in combination with SPOT digital camera and image analysis software (Diagnostic Instruments, Sterling Heights, MI).

Western Blot Analysis and Semiquantitative and Real-Time PCR.

Immunohistochemistry and in Situ Hybridization. Frozen skin samples embedded into Tissue-Tek medium were processed for immunohistochemical protocols, as described (33, 34). Primary antisera used for this study are listed in Table 4, which is published as supporting information on the PNAS web site. In situ hybridization with digoxigenin-labeled riboprobe for Sostdc1 was performed as described (33, 34). Image preparation and analyses were performed by using bright-field and fluores1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

27. 28. 29.

Fuchs E, Raghavan S (2002) Nat Rev Genet 3:199–209. Fuchs E, Merrill BJ, Jamora C, DasGupta R (2001) Dev Cell 1:13–25. Millar SE (2002) J Invest Dermatol 118:216–225. Galbraith DB (1969) Nature 222:288–290. Sundberg JP, Hogan ME (1994) in Handbook of Mouse Mutations with Skin and Hair Abnormalities (CRC, Boca Raton, FL), pp 57–68. Panteleyev AA, Jahoda CA, Christiano AM (2001) J Cell Sci 114:3419–3431. Stenn KS, Paus R (2001) Physiol Rev 81:449–494. Botchkarev VA, Kishimoto J (2003) J Invest Dermatol Symp Proc 8:46–55. Christiano AM (2004) Cell 118:530–532. Van Scott EJ, Ekel TM (1958) J Invest Dermatol 31:281–287. Hebert JM, Rosenquist T, Gotz J, Martin GR (1994) Cell 78:1017–1025. Elliott K, Stephenson TJ, Messenger AG (1999) J Invest Dermatol 113:873–877. Mustonen T, Pispa J, Mikkola ML, Pummila M, Kangas AT, Pakkasjarvi L, Jaatinen R, Thesleff I (2003) Dev Biol 259:123–136. Schlake T (2005) Mech Dev 122:988–997. Mustonen T, Ilmonen M, Pummila M, Kangas AT, Laurikkala J, Jaatinen R, Pispa J, Gaide O, Schneider P, Thesleff I, Mikkola ML (2004) Development (Cambridge, UK) 131:4907–4919. Zhang M, Brancaccio A, Weiner L, Missero C, Brissette JL (2003) Genesis 37:30–37. Schlake T (2005) Development (Cambridge, UK) 132:2981–2990. Weger N, Schlake T (2005) J Invest Dermatol 125:873–882. Weger N, Schlake T (2005) J Invest Dermatol 125:847–849. Tucker AS, Matthews KL, Sharpe PT (1998) Science 282:1136–1138. Jiang, T-X, Jung, H-S, Widelitz RB, Chuong, C-M (1999) Development (Cambridge, UK) 126:4997–5009. Plikus M, Wang WP, Liu J, Wang X, Jiang TX, Chuong CM (2004) Am J Pathol 164:1099–1113. Kassai Y, Munne P, Hotta Y, Penttila E, Kavanagh K, Ohbayashi N, Takada S, Thesleff I, Jernvall J, Itoh N (2005) Science 309:2067–2070. Miyazono K, Maeda S, Imamura T (2005) Cytokine Growth Factor Rev 16:251–263. Yanagita M (2005) Cytokine Growth Factor Rev 16:309–317. Botchkarev VA, Botchkareva NV, Roth W, Nakamura M, Chen, L-H, Herzog W, Lindner G, McMahon JA, Peters C, Lauster R, et al. (1999) Nat Cell Biol 1:158–164. Kulessa H, Turk G, Hogan BL (2000) EMBO J 19:6664–6674. Botchkarev VA, Botchkareva NV, Nakamura M, Huber O, Funa K, Lauster R, Paus R, Gilchrest BA (2001) FASEB J 15:2205–2214. Kobielak K, Pasolli HA, Alonso L, Polak L, Fuchs E (2003) J Cell Biol 163:609–623.

Sharov et al.

Cell Culture and Transfection Experiments. Mouse primary KC from newborn WT and TG mice were prepared as described (29). HaCaT KC were grown in DMEM supplemented by 4% FBS until 60–70% confluence. Cells were incubated for 6–24 h with 200 ng/ml rhBMP4 (R & D Systems, Minneapolis, MN) or diluent control, as described (33, 34). The immunofluorescent staining of incorporated BrdU was performed by using the BrdU Flow Kit (BD Pharmingen, San Diego, CA). The proliferation rate of primary KC and HaCaT cells was assessed by using a FACScan flow cytometer equipped with CellQuest PRO software (BD Biosciences, Mountain View, CA). The p27PF-Luc reporter plasmid containing the human p27 promoter region (42), vectors containing constitutively active BMP receptors, Smad1/5-expressing plasmids under the pCMV promoter (41) (kindly provided by K. Funa), and the pGVB2L-Luc control vector were used for transfection of HaCaT cells, as described (33). We thank Profs. B. A. Gilchrest and M. J. O’Brien for their continuous support and Prof. K. Funa (University of Gothenburg, Gothenburg, Sweden) and Prof. T. Sakai (Kyoto University, Kyoto, Japan) for providing reagents for this study. This study was supported by National Institutes of Health Grants AR48573, AR49778, and CA092090 (to V.A.B.). 30. Andl T, Ahn K, Kairo A, Chu EY, Wine-Lee L, Reddy ST, Croft NJ, Cebra-Thomas JA, Metzger D, Chambon P, et al. (2004) Development (Cambridge, UK) 131:2257–2268. 31. Yuhki M, Yamada M, Kawano M, Iwasato T, Itohara S, Yoshida H, Ogawa M, Mishina Y (2004) Development (Cambridge, UK) 131:1825–1833. 32. Ming Kwan K, Li AG, Wang XJ, Wurst W, Behringer RR (2004) Genesis 39:10–25. 33. Sharov AA, Fessing M, Atoyan R, Sharova TY, Haskell-Luevano C, Weiner L, Funa K, Brissette JL, Gilchrest BA, Botchkarev VA (2005) Proc Natl Acad Sci USA 102:93–98. 34. Sharov AA, Weiner L, Sharova TY, Siebenhaar F, Atoyan R, Reginato AM, McNamara CA, Funa K, Gilchrest BA, Brissette JL, Botchkarev VA (2003) EMBO J 22:2992–3003. 35. Mammucari C, Tommasi di Vignano A, Sharov AA, Neilson J, Havrda MC, Roop DR, Botchkarev VA, Crabtree GR, Dotto GP (2005) Dev Cell 8:665–676. 36. Rendl M, Lewis L, Fuchs E (2005) PLoS Biol 3:e331. 37. Osawa M, Egawa G, Mak SS, Moriyama M, Freter R, Yonetani S, Beermann F, Nishikawa S (2005) Development (Cambridge, UK) 132:5589–5599. 38. Schlake T (2006) Mech Dev 123:641–648. 39. O’Shaughnessy RF, Yeo W, Gautier J, Jahoda CA, Christiano AM (2004) J Invest Dermatol 123:613–621. 40. Fero ML, Rivkin M, Tasch M, Porter P, Carow CE, Firpo E, Polyak K, Tsai LH, Broudy V, Perlmutter RM, et al. (1996) Cell 85:733–744. 41. Kusanagi K, Inoue H, Ishidou Y, Mishima HK, Kawabata M, Miyazono K (2000) Mol Biol Cell 11:555–565. 42. Minami S, Ohtani-Fujita N, Igata E, Tamaki T, Sakai T (1997) FEBS Lett 411:1–6. 43. Conlon I, Raff M (1999) Cell 96:235–244. 44. Potter CJ, Xu T (2001) Curr Opin Genet Dev 11:279–286. 45. Plikus MV, Zeichner-David M, Mayer JA, Reyna J, Bringas P, Thewissen JG, Snead ML, Chai Y, Chuong CM (2005) Evol Dev 7:440–457. 46. Legue E, Nicolas JF (2005) Development (Cambridge, UK) 132:4143–4154. 47. Ma L, Liu J, Wu T, Plikus M, Jiang TX, Bi Q, Liu YH, Muller-Rover S, Peters H, Sundberg JP, et al. (2003) Development (Cambridge, UK) 130:379–389. 48. Paus R, Muller-Rover S, Van Der Veen C, Maurer M, Eichmuller S, Ling G, Hofmann U, Foitzik K, Mecklenburg L, Handjiski B (1999) J Invest Dermatol 113:523–532. 49. Ohyama M, Terunuma A, Tock CL, Radonovich MF, Pise-Masison CA, Hopping SB, Brady JN, Udey MC, Vogel JC (2006) J Clin Invest 116:249–260. 50. Hamatani T, Carter MG, Sharov AA, Ko MS (2004) Dev Cell 6:117–131.

PNAS 兩 November 28, 2006 兩 vol. 103 兩 no. 48 兩 18171

DEVELOPMENTAL BIOLOGY

Western blot analysis of total tissue proteins obtained from the extracts of full-thickness skin of TG and WT mice was performed rabbit polyclonal antisera against p27Kip1, pRb (Zymed, South San Francisco, CA), and pSmad1/5 (gift of K. Funa, University of Gothenburg, Gothenburg, Sweden), as described (33, 34). Total RNAs were isolated from captured hair matrix and FP areas followed by two rounds of linear amplification, as described above. PCR primers were designed on Beacon Designer software (Premier Biosoft, Palo Alto, CA; Table 3, which is published as supporting information on the PNAS web site). PCR and real-time PCR were performed by using iCycler Thermal Cycler (Bio-Rad, Hercules, CA) and MyiQ SingleColor Real-Time PCR Detection System (Bio-Rad), as described (33, 34).