Type II Collagen Accumulation in Overlying Dermo-Epidermal Junction of Pilomatricoma Is Mediated by Bone Morphogenetic Protein 2 and 4
2004; Elsevier BV; Volume: 122; Issue: 4 Linguagem: Inglês
10.1111/j.0022-202x.2004.22417.x
ISSN1523-1747
AutoresHideki Mieno, Kei Kuroda, Hiroshi Shinkai, Hidekatsu Yoshioka, Shingo Tajima,
Tópico(s)Genetic and rare skin diseases.
ResumoPilomatricoma consists of the cells differentiating towards hair matrix cells. Immunohistochemical study revealed the deposition of type II collagen in the overlying dermo-epidermal junction (DEJ) of this benign tumor. Proα1(II) mRNA was detected by RT-PCR in the overlying epidermal layer but not in the dermal layer prepared from the lesional skin of pilomatricoma. The neutral salt-soluble proteins extracted from the tumor of pilomatricoma induced proα1(II) mRNA in the cultured human keratinocytes but not in the cultured dermal fibroblasts. Bone morphogenetic protein 2 or 4 (BMP2 or 4) was immunohistochemically detected in some shadow cells of pilomatricoma. Recombinant BMP2 and BMP4 were found to induce proα1(II) mRNA concentration dependently in the cultured human keratinocytes but not in the cultured fibroblasts. Proα1(II) mRNA induced by BMP2 and in cultured keratinocytes contained exon 2, indicating that the mRNA species is non-chondrogenic type IIA form. The results strongly suggest that BMP2 or 4 expressed in pilomatricoma is responsible for the induction of proα1(II) collagen mRNA in the overlying epidermal cells resulting in the deposition of type II collagen in the DEJ. When human keratinocytes were cultured on type II collagen substratum in vitro, the cell proliferation was accelerated at the early period of culture but was inhibited at the late period of culture, whereas the cell proliferation was persistently accelerated by type I or IV collagen substratum. Type II collagen deposition in the DEJ may potentially exert profound effects on keratinocyte proliferation and differentiation. Pilomatricoma consists of the cells differentiating towards hair matrix cells. Immunohistochemical study revealed the deposition of type II collagen in the overlying dermo-epidermal junction (DEJ) of this benign tumor. Proα1(II) mRNA was detected by RT-PCR in the overlying epidermal layer but not in the dermal layer prepared from the lesional skin of pilomatricoma. The neutral salt-soluble proteins extracted from the tumor of pilomatricoma induced proα1(II) mRNA in the cultured human keratinocytes but not in the cultured dermal fibroblasts. Bone morphogenetic protein 2 or 4 (BMP2 or 4) was immunohistochemically detected in some shadow cells of pilomatricoma. Recombinant BMP2 and BMP4 were found to induce proα1(II) mRNA concentration dependently in the cultured human keratinocytes but not in the cultured fibroblasts. Proα1(II) mRNA induced by BMP2 and in cultured keratinocytes contained exon 2, indicating that the mRNA species is non-chondrogenic type IIA form. The results strongly suggest that BMP2 or 4 expressed in pilomatricoma is responsible for the induction of proα1(II) collagen mRNA in the overlying epidermal cells resulting in the deposition of type II collagen in the DEJ. When human keratinocytes were cultured on type II collagen substratum in vitro, the cell proliferation was accelerated at the early period of culture but was inhibited at the late period of culture, whereas the cell proliferation was persistently accelerated by type I or IV collagen substratum. Type II collagen deposition in the DEJ may potentially exert profound effects on keratinocyte proliferation and differentiation. dermo-epidermal junction ethylenediaminetetraacetic acid immunoglobulin N-ethylmaleimide phenylmethylsulfonylfluoride sodium dodecylsulfate 0.15 M sodium chloride and 0.015 M sodium citrate Pilomatricoma or calcifying epithelioma of Malherbe is a benign cutaneous epithelial tumor, and morphologically composed of small basophilic cells and large, probably non-viable shadow cells as well as transitional cells between basophilic and shadow cells (Forbis and Helwig, 1961Forbis R. Helwig E.B. Pilomatrixoma (calcifying epithelioma).Arch Dermatol. 1961; 83: 606-618Crossref PubMed Scopus (351) Google Scholar). Previous immunohistochemical and histochemical studies demonstrated that majority of pilomatricoma cells are considered to be differentiating cells towards hair-forming cells, particularly hair cortex cells, of hair follicles (Moll et al., 1988Moll I. Heid H. Moll R. Cytokeratin analysis of pilomatrixoma: Change in cytokeratin-type expression during differentiation.J Invest Dermatol. 1988; 91: 251-257Abstract Full Text PDF PubMed Google Scholar;Watanabe et al., 1994Watanabe S. Wagatsuma K. Takahashi H. Immunohistochemical localization of cytokeratins and involucrin in calcifying epithelioma: Comparative studies with normal skin.Br J Dermatol. 1994; 131: 506-513Crossref PubMed Scopus (47) Google Scholar). The mechanism of epidermal cell differentiation towards hair matrix cells remains to be determined. Recently β-catenin, a participant in the Wnt signaling, has been shown to play an important role in the morphogenesis of hair follicles and the tumorigenesis of fair follicle-related tumors including pilomatricomas. β-catenin is essential for fate decisions of stem cells between the formation of follicular and epidermal keratinocyte lineages (Gat et al., 1998Gat U. DasGupta R. Degenstein L. Fuchs E. De novo hair follicle morphogenesis and hair tumors in mice expressing a truncated β-catenin in skin.Cell. 1998; 95: 605-614Abstract Full Text Full Text PDF PubMed Scopus (933) Google Scholar;Huelsken et al., 2001Huelsken J. Vogel R. Erdmann B. Cotsarelis G. Birchmeier W. β-catenin controls hair follicle morphogenesis and stem cell differentiation in the skin.Cell. 2001; 105: 533-545Abstract Full Text Full Text PDF PubMed Scopus (1035) Google Scholar), and 26% or 75% of pilomatricomas possess activating mutations in exon 3 of β-catenin gene (Chan et al., 1999Chan E.F. Gat U. McNiff J.M. Fuchs E. A common human skin tumor is caused by activating mutations in beta-catenin.Nat Genet. 1999; 21: 410-413Crossref PubMed Scopus (516) Google Scholar;Moreno-Bueno et al., 2001Moreno-Bueno G. Gamallo C. Perez-Gallego L. Contreras F. Palacios J. Beta-catenin expression in pilomatricomas. Relationship with beta-catenin gene mutations and comparision with beta-catenin expression in normal hair follicles.Br J Dermatol. 2001; 145: 576-581Crossref PubMed Scopus (87) Google Scholar). Bone morphogenetic proteins (BMPs) are considered to be members of transforming growth factor β superfamily based on their amino acid sequence homology. BMPs exhibit chondrogenic and osteogenic properties in vivo and in vitro (Wozney et al., 1988Wozney J.M. Rosen V. Celeste A.J. Novel regulators of bone formation: Molecular clones and activities.Science. 1988; 242: 1528-1534Crossref PubMed Scopus (3229) Google Scholar). For examples, implantation of recombinant human BMP in rat induces bone formation (Wang et al., 1990Wang E.A. Rosen V. D'Alessandro J.S. Recombinant human bone morphogenetic protein induces bone formation.Proc Natl Acad Sci USA. 1990; 87: 2220-2224Crossref PubMed Scopus (1325) Google Scholar). Incubation of bone marrow-derived mesenchymal progenitor cells with BMP induces chondrogenic differentiation in vitro (Johnstone et al., 1998Johnstone B. Hering T.M. Caplan A.I. Goldberg V.M. Yoo J.U. In vitro chondrogenesis of bone marrow-derived mesenchymal progenitor cells.Exp Cell Res. 1998; 238: 265-272Crossref PubMed Scopus (1930) Google Scholar;Sekiya et al., 2001Sekiya I. Colter D.C. Prockop D.J. BMP-6 enhances chondrogenesis in a subpopulation of human marrow stromal cells.Biochem Biophys Res Commun. 2001; 284: 411-418Crossref PubMed Scopus (268) Google Scholar). Besides the chondrogenic or osteogenic activity, BMP2 and BMP4 are potential hair follicle initiation factors and regulators of hair cycle (Blessing et al., 1993Blessing M. Nanney L.B. King L.E. Jones C.M. Hogan B.L. Trasgenic mice as a model to study the role of TGF-beta-related molecules in hair follicles.Genes Dev. 1993; 7: 204-215Crossref PubMed Scopus (149) Google Scholar;Botchkarev, 2003Botchkarev V.A. Bone morphogenetic proteins and their antagonists in skin and hair follicle biology.J Invest Dermatol. 2003; 120: 36-47Crossref PubMed Scopus (142) Google Scholar). Increased evidences suggest that the expression of BMP depends on β-catenin during follicular formations and hair cycling, although whether BMP is a direct or indirect target gene of β-catenin signaling in hair formation is currently unknown (Huelsken et al., 2001Huelsken J. Vogel R. Erdmann B. Cotsarelis G. Birchmeier W. β-catenin controls hair follicle morphogenesis and stem cell differentiation in the skin.Cell. 2001; 105: 533-545Abstract Full Text Full Text PDF PubMed Scopus (1035) Google Scholar). Type II collagen is the major and unique collagenous component of cartilage and plays a critical role in chondrogenesis during embryonic development (Mayne and von der Mark, 1983Mayne R. von der Mark K. Collagens of cartilage.in: Hall B.K. Cartilage. Vol 1. Academic Press, New York1983: 181-214Google Scholar). It has been demonstrated that type II collagen is much more widely distributed during early development than previously thought. Type II collagen is detected in non-chondrogenic tissues including embryonic notochord (Linsenmayer et al., 1973Linsenmayer T.F. Trelstad R.L. Gross J. The collagen of chick embryonic notochord.Biochem Biophys Res Commun. 1973; 53: 39-45Crossref PubMed Scopus (75) Google Scholar;Miller, 1974Miller E.J. Characterization of notochord collagen as a cartilage-type collagen.Biochem Biophys Res Commun. 1974; 60: 424-430Crossref PubMed Scopus (44) Google Scholar), corneal epithelium (Linsenmayer et al., 1977Linsenmayer T.F. Smith G.N. Hay E.D. Synthesis of two collagen types by embryonic chick corneal epithelium in vitro.Proc Natl Acad Sci USA. 1977; 74: 39-43Crossref PubMed Scopus (97) Google Scholar), neural retina (Smith et al., 1976Smith G.N. Linsenmayer T.F. Newsome D.A. Synthesis of type II collagen in vitro by embryonic chick neural retina tissue.Proc Natl Acad Sci USA. 1976; 73: 4420-4423Crossref PubMed Scopus (62) Google Scholar), ectodermal-mesenchymal interfaces throughout the trunk of the chick embryo at the very early stages (stages 14–19; days 2–3) (Thorogood et al., 1986Thorogood P. Bee J. von der Mark K. Transient expression of collagen type II at epitheliomesenchymal interfaces during morphogenesis of the cartilaginous neurocranium.Dev Biol. 1986; 116: 497-509Crossref PubMed Scopus (97) Google Scholar;Kosher and Solursh, 1989Kosher R.A. Solursh M. Widespread distribution of type II collagen during embryonic chick development.Dev Biol. 1989; 131: 558-566Crossref PubMed Scopus (93) Google Scholar) and the subepidermal matrix of mouse hindlimb (stage 31; 7.5 d) which subsequently disappears with development (Fitch et al., 1989Fitch J.M. Mentzer A. Mayne R. Linsenmayer T.F. Independent deposition of collagen type I and II at epithelial-mesenchymal interfaces.Development. 1989; 105: 85-95PubMed Google Scholar,Fitch et al., 1990Fitch J.M. Gordon M.K. Marchant J. Gibney E. Linsenmayer T.F. Avian epidermis express collagen type I and II but not type TX in a restricted spatial pattern during feather morphogenesis.Ann NY Acad Sci. 1990; 580: 492-495Crossref Scopus (2) Google Scholar). We have previously demonstrated that type II collagen is detected in the dermo-epidermal junction (DEJ) excluding the region at which the hair follicles are forming in the human fetal scalp skin at the stages of 17–23 fetal weeks (Azuma et al., 1994Azuma N. Izumi T. Tajima S. Nishikawa T. Ohshima A. Expression of type II collagen at the middle stages of chick embryonic and human fetal skin development.J Invest Dermatol. 1994; 102: 958-962Abstract Full Text PDF PubMed Google Scholar), implying that type II collagen expression in the DEJ precedes the formation of hair follicles in the fetal skin development. In the course of the studies exploring the expression of type II collagen in the postnatal skin, we found the expression of type II collagen in the overlying DEJ of pilomatricoma and demonstrated that the induction was mediated by BMP2 or BMP4 detected in the tumor cells. We further studied the influence of type II collagen deposition on keratinocyte growth and differentiation in the cell culture system. The results in this study indicate that keratinocytes in the postnatal skin is still pluripotential and can express type II collagen under the pathological condition. Immunohistochemical study of pilomatricoma using anti-type II collagen antibody demonstrated that four cases of pilomatricoma exhibited a strong staining in the overlying DEJ. The immunoreactivity was restricted in DEJ, and no apparent immunoreactivity was found in the dermal matrices Figure 1a and b. Normal skins obtained from benign skin tumors including epidermal cyst (n=5) and nevus cell nevus (n=5) showed negative reaction with this antibody Figure 1c. Localization of proα1(II) mRNA was studied by RT-PCR using the RNA prepared from the lesional epidermis and dermis of cases 1 and 2 pilomatricomas. Specific 369 bp fragment of proα1(II) mRNA was found to be detected preferentially in the overlying epidermis of pilomatricoma not in the dermis in both cases (data not shown). Paraffin-embedded sections were incubated with anti-BMP2/4 antibody. BMP2/4 was detected in some shadow cells of pilomatricoma in both cases 1 and 2 but not in the basophilic or transitional cells Figure 2, which was consistent with the previous report (Kurokawa et al., 2000Kurokawa I. Kusumoto K. Bessho K. Okubo Y. Senzaki H. Tsubura A. Immunohistochemical expression of bone morphogenetic protein-2 in pilomatricoma.Br J Dermatol. 2000; 143: 754-758Crossref PubMed Scopus (23) Google Scholar). There was no immunoreactivity of BMP2/4 in the overlying epidermal cells and dermal cells (not shown). We first speculated that some soluble factor(s) may be secreted from the tumor and act on basal keratinocytes to induce the expression of type II collagen. In order to prove this, tumor mass of pilomatricoma was homogenized and sonicated in the neutral buffer (0.1 M Tris-HCl, pH 7.5, containing protease inhibitor cocktail). Human keratinocytes and fibroblasts were treated with the homogenate of pilomatricoma at the dose of 10 μg per mL medium for 24 h. RT-PCR clearly showed that the homogenate induced 369 bp fragment of proα1(II) mRNA in the cultured keratinocytes but not in the cultured fibroblasts Figure 3a. To identify the factor(s) in the homogenate, ELISA was done using the antibodies for various factors including tumor necrosis factor-α (TNF-α), tumor growth factor-β (TGF-β), and interleukin-1β (IL-1β). But these factors gave negative results (not shown). Because the result of immunohistochemistry of shadow cells of piloatricoma with the antibody for BMP2/4 strongly suggested that the candidate factors were BMP2 and BMP4, we continued the further study focusing on BMP2 and BMP4. To prove the type II collagen-inducible activity of BMP2 and BMP4 directly, we studied whether proα1(II) mRNA is induced in the cultured human keratinocytes by the treatment of various amount of BMP2 or BMP4. RT-PCR demonstrated that specific 369 bp fragment of proα1(II) transcript was detected after the treatment of BMP2 and BMP4 at the concentrations higher than 100 ng per mL Figure 3b. By contrast, dermal fibroblasts did not induce proα1(II) transcript by BMP2 and BMP4 (not shown) at the concentrations tested here. Pre-treatment of the cells with recombinant human noggin (1.6 ng per mL) inhibited BMP2-inducible proα1(II) transcript enhancement Figure 3c. Since it has been demonstrated that type II procollagen is expressed in two structurally different mRNAs by alternative splicing of the primary gene transcript that either include (type IIA) or exclude (type IIB) an exon 2 encoding the major portion of the amino terminal propeptide (Ryan and Sandell, 1990Ryan M.C. Sandell L.J. Differential expression of a cysteine-rich domain in the amino-terminal propeptide of type II (cartilage) procollagen by alternative splicing of mRNA.J Biol Chem. 1990; 265: 10334-10339Abstract Full Text PDF PubMed Google Scholar;Sandell et al., 1991Sandell L.J. Morris N. Robbins J.R. Goldring M.B. Alternatively spliced type II procollagen mRNAs define distinct populations during vertebral development: Differential expression of the amino-propeptide.J Cell Biol. 1991; 114: 1307-1319Crossref PubMed Scopus (290) Google Scholar), we performed the structural analysis of proα1(II) mRNA induced by BMP2 and BMP4 in the cultured keratinocytes. One single fragment of 377 bp in both BMP2- and BMP4-treated keratinocytes was detected by RT-PCR (Figure 4a, lanes 2 and 4), whereas 171 bp fragment was detected in the auricular cartilage (Figure 4a, lane 5), indicating that proα1(II) mRNA species induced by BMP2 and BMP4 is type IIA form and those of auricular cartilage is type IIB form. To confirm this result, RT-PCR products were blotted onto the filters and hybridized with the oligonucleotide probe spanning 3′ end of exon 1 and 5′ end of exon 2. A 377 bp fragment was detected by autoradiography in both BMP2- and BMP4-treated keratinocytes but 171 bp fragment in the ear cartilage was not Figure 4b, indicating that proα1(II) mRNA species was type IIA form. In order to investigate the biological significance of type II collagen accumulated in the DEJ, effect of type II collagen on keratinocyte proliferation was studied in vitro. Cells were plated at a relatively low density of 1 × 103 per 32 mm diameter dishes to observe the change of cell growth for longer periods. Type II collagen substratum as well as types I and IV collagen substrata showed a growth-promoting effect at the early phase in culture (days 4 and 6 in culture) but type II collagen substratum exhibited a significant inhibiting effect at the late phase of culture (days 12 and 14 in culture) (p<0.01) when type I collagen and type IV collagen still showed growth-stimulating effect Figure 5. A considerable number of cells on type II collagen substratum on days 12 and 14 was found to float in the medium under a microscope (not shown). We have found type II collagen deposition in overlying DEJ of pilomatricoma. Overlying epidermal keratinocytes will be responsible for the expression of type II collagen because type II collagen mRNA was detected in the overlying epidermis not in the dermal tissue, and type II collagen mRNA was induced by BMP2 and BMP4 treatments in the cultured keratinocytes but not in the cultured fibroblasts. This is an interesting observation because extracellular matrix proteins in DEJ including type IV collagen are considered to be produced through epidermal–dermal interaction (Fleischmajer et al., 1993Fleischmajer R. MacDonald II., E.D. Contard P. Perlish J.S. Immunochemistry of a keratinocyte–fibroblast co-culture model for reconstruction of human skin.J Histochem Cytochem. 1993; 41: 1359-1366Crossref PubMed Scopus (65) Google Scholar;Marinkovich et al., 1993Marinkovich M.P. Keene D.R. Rimberg C.S. Burgeson R.E. Cellular origin of the dermal–epidermal basement membrane.Dev Dyn. 1993; 197: 255-267Crossref PubMed Scopus (202) Google Scholar). In addition, transient expression of type II collagen in the ectodermal–mesenchymal interfaces during the early stage of chick embryonic development is considered to be mediated by epithelial–mesenchymal interaction (Thorogood et al., 1986Thorogood P. Bee J. von der Mark K. Transient expression of collagen type II at epitheliomesenchymal interfaces during morphogenesis of the cartilaginous neurocranium.Dev Biol. 1986; 116: 497-509Crossref PubMed Scopus (97) Google Scholar;Kosher and Solursh, 1989Kosher R.A. Solursh M. Widespread distribution of type II collagen during embryonic chick development.Dev Biol. 1989; 131: 558-566Crossref PubMed Scopus (93) Google Scholar). We have found that the homogenate extracted from pilomatricoma by neutral salt is capable of inducing type II collagen expression in cultured keratinocytes. In the course of searching the candidate growth factor or cytokines responsible for the type II collagen induction, we found BMP2/4 expression in the shadow cells of pilomatricoma but not in the other cell components of pilomatricoma. On the basis of the hypothesis that BMP2 and BMP4 are potent inducible factor of type II collagen, we treated the cultured keratinocytes with BMP2 and BMP4 and found their type II collagen-inducing activity. This was confirmed by the experiment that the induction of type II collagen by BMP2 was abolished in the presence of noggin, a specific agonist of BMP2 and BMP4. We have tested some other cytokines and growth factor such as TGF-β1, TNF-α, or IL-1β but failed to find such activities (not shown). This suggests that BMP2 and BMP4 produced and secreted by shadow cells of pilomatricoma act on overlying epidermis to induce type II collagen. In fact, in the normal-appearing skin external from pilomatricoma type II collagen deposition in DEJ was not be able to be detected (not shown). These results are consistent with the previous studies that the expression of BMPs have an essential role in cutaneous tumorigenesis (Blessing et al., 1995Blessing M. Nanney L.B. King L.E. Hogan B.L. Chemical skin carcinogenesis is prevented in mice by the induced expression of a TGF-beta related transgene.Teratog Carcinog Mutagen. 1995; 15: 11-21Crossref PubMed Scopus (40) Google Scholar,Blessing et al., 1996Blessing M. Schirmacher P. Kaiser S. Overexpression of bone morphogenetic protein-6 (BMP-6) in the epidermis of transgenic mice: Inhibition or stimulation of proliferation depending on the pattern of transgene expression and formation of psoriatic lesions.J Cell Biol. 1996; 135: 227-239Crossref PubMed Scopus (119) Google Scholar;Wach et al., 2001Wach S. Schirmacher P. Protschka M. Blessing M. Overexpression of bone morphogenetic protein-6 (BMP-6) in murine epidermis suppresses skin tumor formation by induction of apoptosis and downregulation of fos/jun family members.Oncogene. 2001; 20: 7761-7769Crossref PubMed Scopus (46) Google Scholar). The question whether type II expression in DEJ is the cause or result of pilomatricoma is still unclear. We initially thought that some factors involving in the differentiation of stem cells towards hair cortex cells in the tumorigenesis of pilomatricoma are also involved in the induction of type II collagen expression in the DEJ. But the presence of detectable BMP2/4 in the shadow cells of pilomatricoma and type II collagen-inducing potential of BMP2 and BMP4 in cultured keratinocytes may suggest that pilomatricoma itself potentially expresses BMP2/4 which induces the expression of type II collagen in the DEJ. Therefore we at present think that type II collagen expression is the result rather than cause of pilomatricoma. Type II collagen has been reported to be expressed in two mRNAs by differential splicing of the primary gene transcript that either includes (type IIA) or excludes (type IIB) an exon 2 encoding the major portion (69 amino acids) of the amino terminal propeptide (Ryan and Sandell, 1990Ryan M.C. Sandell L.J. Differential expression of a cysteine-rich domain in the amino-terminal propeptide of type II (cartilage) procollagen by alternative splicing of mRNA.J Biol Chem. 1990; 265: 10334-10339Abstract Full Text PDF PubMed Google Scholar). Each procollagen mRNA has a distinct tissue distribution during chondrogenesis, with type IIB expressed in chondrocyte and type IIA expressed in the chondroprogenitor cells surrounding the cartilage (Sandell et al., 1991Sandell L.J. Morris N. Robbins J.R. Goldring M.B. Alternatively spliced type II procollagen mRNAs define distinct populations during vertebral development: Differential expression of the amino-propeptide.J Cell Biol. 1991; 114: 1307-1319Crossref PubMed Scopus (290) Google Scholar). Proα1(II) mRNA induced by BMP2 and BMP4 was found to contain exon 2 (type IIA), indicating that type II collagen in DEJ of pilomatricoma is non-chondrogenic from (type IIA). Physiological significance of type II collagen in the DEJ is uncertain at present. Accumulated type II collagen in the DEJ may exert a profound influence on the stability of DEJ as well as the differentiation of keratinocytes and the structural organization of adjacent extracellular matrix components (such as anchoring fibrils) in the subepidermal region since in the normal condition type II collagen never exists in this area except a very short duration of chick embryonic and human fetal skin development (Azuma et al., 1994Azuma N. Izumi T. Tajima S. Nishikawa T. Ohshima A. Expression of type II collagen at the middle stages of chick embryonic and human fetal skin development.J Invest Dermatol. 1994; 102: 958-962Abstract Full Text PDF PubMed Google Scholar). Transient expression of type II collagen in chick embryonic and human fetal scalp skins is thought to be related to the development of feather and hair follicle, because this collagen starts to diminish at the restricted sites where feather buds and hair follicles are being developed. If type II collagen induction in the overlying DEJ of pilomatricoma occurs by the same mechanism as seen in the chick embryonic and human fetal skin development, type II collagen in pilomatricoma induced by BMP2/4, which are potent hair follicle initiation factors (Blessing et al., 1993Blessing M. Nanney L.B. King L.E. Jones C.M. Hogan B.L. Trasgenic mice as a model to study the role of TGF-beta-related molecules in hair follicles.Genes Dev. 1993; 7: 204-215Crossref PubMed Scopus (149) Google Scholar), will be a preceding marker protein of the induction of hair follicles. Keratinocyte differentiation to hair matrix occurs in the lesions other than pilomatricomas such as follicular cysts and cutaneous mixed tumors (LeBoit et al., 1987LeBoit P.E. Parslow T.G. Choy S.-H. Hair matrix differentiation. Occurrence in lesions other than pilomatricoma.Am J Dermatopathol. 1987; 9: 399-405Crossref PubMed Scopus (44) Google Scholar). We found type II collagen accumulation in the overlying DEJ of some cases of trichilemmoma and cutaneous mixed tumor as well (manuscript in preparation). This will also support the idea that type II collagen expression in the DEJ is related to follicular differentiation of keratinocytes. Previous experiments have shown that extracellular matrices like types I and IV collagens, fibronectin, and laminin influence keratinocyte adhesion, spreading, proliferation, and differentiation (Kubo et al., 1987Kubo M. Kan M. Isemura M. Yamane I. Tagami H. Effects of extracellular matrices on human keratinocyte adhesion and growth and on its secretion and deposition of fibronectin in culture.J Invest Dermatol. 1987; 88: 594-601Abstract Full Text PDF PubMed Google Scholar;Adams and Watt, 1989Adams J.C. Watt F.M. Fibronectin inhibits the terminal differentiation of human keratinocytes.Nature. 1989; 340: 307-309Crossref PubMed Scopus (306) Google Scholar;Guo and Grinnell, 1989Guo M. Grinnell F. Basement membrane and human epidermal differentiation in vitro.J Invest Dermatol. 1989; 93: 372-378Abstract Full Text PDF PubMed Google Scholar;Woodley et al., 1990Woodley D.T. Wynn K.C. O'Keefe E.J. Type TW collagen and fibronectin enhance human keratinocyte thymidine incorporation and spreading in the absence of soluble growth factors.J Invest Dermatol. 1990; 94: 139-143Abstract Full Text PDF PubMed Google Scholar). In fact, type I or IV-, fibronectin-, and laminin-coated plastic dishes have been used to efficiently maintain keratinocyte growth in culture. The interaction between keratinocytes and type II collagen has never been studied because expression of type II collagen has never been expected in DEJ or subepidermal area of the skin. It is noted that type II collagen, like other extracellular matrices, promoted keratinocyte proliferation at the early stage of culture, but unlike other extracellular matrices, inhibited keratinocyte proliferation. It is uncertain that the biphasic effect of type II collagen substratum on keratinocyte proliferation is due to the difference of cell density during the cell proliferation or due to the switching of cell signaling pathway mediated by the interaction between keratinocyte and type II collagen. The histological changes in the overlying epidermis of pilomatricoma such as the formation of horn cyst, proliferation of basalioma-like cells (Aso et al., 1990Aso K. Hashimoto H. Kondo S. Watanabe S. A case of eruptive pilomatricoma: Clinical survey of 21 cases of pilomatricoma during the past 13 years.Rinsho Derma. 1990; 44: 977-981Google Scholar), or abnormal elongation of epidermal rete ridge (histologically seen in case 1 in Figure 1) may be related to the altered keratinocyte differentiation induced by the accumulation of type II collagen in the DEJ. Skin samples were obtained from four cases of pilomatricoma under the permission of the patients and subjected to histological and biological analysis. Case 1: 22-y-old Japanese woman noticed skin nodule on the right arm for 1 y. Physical examination showed 21 × 25 mm intradermal tumor with clear border. Case 2: 15-y-old Japanese girl presented with asymptomatic dermal tumor with the diameter of 12 mm on the neck. She noticed the tumor 3 y before. Case 3: 17-y-old Japanese boy presented with asymptomatic dermal tumor with the diameter of 15 mm on the back. He noticed the tumor 3 y before. Case 4: 16-y-old Japanese girl presented with asymptomatic dermal tumor with the diameter of 30 × 25 mm on the back. She noticed the tumor 2 y before. The skin tumors were resected by a surgical operation. Histological diagnosis of pilomatricoma was done with hematoxylin–eosin stain. Normal skins (n=10) were obtained from normal-appearing area of benign skin tumors (epidermal cyst and nevus cell nevus). Skin samples were fixed in 10% buffered formalin, embedded in paraffin and cut into 5 μm sections. The sections were pre-treated with pronase (0.1% in phosphate-buffered saline) (type XX TV, Sigma, St Louis, Missouri) for 30 min, then incubated for 24 h at 4°C with monoclonal anti-human type II collagen antibody (Fuji Chemical Industries, Toyama, Japan) at 1:40 dilution, anti-human polyclonal BMP2/4 antibody (Genzyme/Techne, Minneapolis, Minnesota) at 1:100 dilution. The sections were incubated with biotin-conjugated anti-mouse immunoglobulin (Ig) antibody or anti-goat Ig antibody (Dako, Glostrup, Denmark) at 1:500 dilution for 2 h at room temperature. Antigen–antibody complex was reacted with peroxidase-labeled avidin–biotin complex (Dako) for 30 min. The reaction was visualized with 3-amino-9-ethylcarbazole. The sections were counterstained with hematoxylin. The specificity of anti-type II collagen antibody has been previously demonstrated by western blot assay (Azuma et al., 1994Azuma N. Izumi T. Tajima S. Nishikawa T. Ohshima A. Expression of type II collagen at the middle stages of chick embryonic and human fetal skin development.J Invest Dermatol. 1994; 102: 958-962Abstract Full Text PDF PubMed Google Scholar) and immunohistochemical studies using the specimen of human auricular cartilage (not shown). Normal human keratinocytes (NHKs) were purchased from Sanko-Junyaku (Tokyo, Japan) and cultured in serum-free, low-calcium (0.09 mM), modified MCDB 153 keratinocyte basal media containing the following growth factors, insulin, epidermal growth factor (EGF), hydrocortisone and bovine pituitary extract (BPE) (designated as KGM). NHKs at third or fourth culture were used in this study according to supplier's recommendation. Normal human skin fibroblast culture was established by explant method. Fibroblasts were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS). Cells were plated at a density of 1 × 104 in 35 mm diameter dishes and grown for 6 d. Tumor mass (∼1 g wet weight) was homogenized in 2 mL of 50 mM Tris-HCl, pH 7.4, containing protease inhibitor cocktail (1 mM EDTA, NEM, and PMSF) with a dounce homogenizer, and sonicated with five 30 s bursts (Handy Sonic UR-20P, Tomy Seiko Co., Tokyo, Japan) at 4°C. The homogenate was cleared by centrifugation at 15,000 g for 30 min at 4°C. The protein content of the supernatant was measure by absorbance at 280 Å. Cells were treated with the homogenate (final concentration; 10 μg per mL medium) for 24 h or human recombinant BMP2 and BMP4 (Pharma Biotechnology, Hannover, Germany) for 24 h at the doses of 0–250 ng per mL. To study the BMP-specific induction of type II collagen expression, cells were treated for 24 h with the combination of BMP2 (100 ng per mL) and recombinant human noggin (1.6 ng per mL) (Pepro Tech EC Ltd, London, UK), a potent agonist for BMP2 and BMP4 (Zimmerman et al., 1996Zimmerman L.B. Jesus-Escobar J.M. Harland R.M. The Spemann organizer signal noggin binds and inactivates bone morphogenetic protein 4.Cell. 1996; 86: 599-606Abstract Full Text Full Text PDF PubMed Scopus (1288) Google Scholar). To study the effect of type II collagen on keratinocyte proliferation, cells were plated at a density of 1 × 103 per 35 mm diameter matrix-coated dishes containing either type I collagen from rat tail tendon, type II collagen from bovine cartilage or type IV collagen from human placenta (Beckton-Dickinson Labware, Bedford, MA). The matrix-coated dishes were prepared by the incubation of tissue culture dishes with these matrix proteins at the concentration of 10 μg per mL for 24 h at 4°C, followed by rinse with KGM twice before use (Kubo et al., 1987Kubo M. Kan M. Isemura M. Yamane I. Tagami H. Effects of extracellular matrices on human keratinocyte adhesion and growth and on its secretion and deposition of fibronectin in culture.J Invest Dermatol. 1987; 88: 594-601Abstract Full Text PDF PubMed Google Scholar). Cells were grown in KGM for 14 d. On days 4, 6, 8, 10, 12, and 14 in culture, the cells were trypsinized and cell number was counted with a Coulter counter. Assays were performed twice in triplicate. Values were expressed by mean±SEM. Statistical significance was calculated by one-sided Student's t test and p-value below 0.05 was considered to be significant. Total cellular RNA was isolated from the cultured keratinocytes or dermal fibroblasts with guanidine thiocyanate (Chomczynski and Sacchi, 1987Chomczynski P. Sacchi N. Single step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction.Anal Biochem. 1987; 162: 156-159Crossref PubMed Scopus (62148) Google Scholar), and adjusted to a concentration of 1 μg per μL. In an experiment, to separate the epidermis from dermis, the lesional skins of the pilomatricomas were cut into small pieces (∼1 cm3) and treated with 0.5% dispase (Sanko-Junyaku) at 4°C for 2 h. Epidermal layer was peeled off from the dermal layer by sterile forceps, then RNA was immediately isolated from both layers. For a control experiment, RNA was prepared from auricular cartilage obtained from surgical operation of squamous cell carcinoma on the ear. The exon-specific primer pairs; 5′ACACTCAAGTCCCTCAACAACCAGAT3′ for upstream primer and 5′GACGTCCAGATGACCTTCCTGCGCCTG3′ for downstream primer at the C-terminal propeptide domain of human proα1(II) chain (Sangiorgi et al., 1985Sangiorgi F.O. Benson-Chanda V. de Wet W.J. Sobel M.E. Tsipouras P. Ramirez F. Isolation and partial characterization of entire human proα1 II collagen gene.Nucleic Acid Res. 1985; 13: 2207-2225Crossref PubMed Scopus (52) Google Scholar) were synthesized. For internal standard, the upstream primer 5′TTAATGTCACGCACGATTTCCC3′ and the downstream primer 5′GTGATGGTGGCATGGGTCA3′ of β-actin cDNA (Ponte et al., 1984Ponte P. Ng S.Y. Engel J. Gunning P. Kedes L. Evolutionary conservation in the untranslated regions of actin mRNAs: DNA sequence of a human beta-actin cDNA.Nucleic Acid Res. 1984; 12: 1687-1696Crossref PubMed Scopus (889) Google Scholar) were also synthesized. Total RNA (1 μg) was used for first strand cDNA synthesis. RT-PCR was performed in the presence of 3 pmol of a 3′-oligonucleotide and 35 U of reverse transcriptase from avian myoblastosis virus (Takara Shuzo Co., Otsu, Shiga, Japan) in a total reaction volume of 20 μL (Kawasaki, 1990Kawasaki E.S. Amplification of RNA.in: Innis M.A. Gelfand D.H. Sninsky J.J. White T.J. PCR Protocols: a Guide to Method and Applications. Academic Press, New York1990: 21-27Crossref Google Scholar). The resulting cRNA was then subjected to the first PCR using 25 pmol 5′-oligonucleotide and additional 3′-oligonucleotide in a total volume of 100 μL. The cDNA was amplified for 35 cycles at 94°C per 1 min, 58°C per 1 min, and 72°C per 2 min. Analysis of all the PCR products was performed by agarose (1.5%) gel electrophoresis followed by ethidium bromide staining. The PCR products were subcloned into a TA cloning vector (Invitrogen Corp., San Diago, California). Plasmid DNA was isolated (Wizard plus Miniprep, Promega Corp., Madison, Wisconsin), and subjected to nucleotide sequencing by dideoxy chain termination method (Sanger et al., 1977Sanger F. Nicklen S. Coulson A.R. DNA sequencing with chain-terminating inhibitors.Proc Natl Acad Sci USA. 1977; 74: 5463-5467Crossref PubMed Scopus (51478) Google Scholar) to identify the cloned cDNA. To analyze the splicing of exon 2 of proα1(II) mRNA, the upstream primer 5′ATGATTCG3′ corresponding to the exon 1 sequences of proα1(II) cDNA and downstream primer 5′AGGCCCAGGAGGTCCTTTGGG3′ corresponding to exon 5 sequences were synthesized (Sangiorgi et al., 1985Sangiorgi F.O. Benson-Chanda V. de Wet W.J. Sobel M.E. Tsipouras P. Ramirez F. Isolation and partial characterization of entire human proα1 II collagen gene.Nucleic Acid Res. 1985; 13: 2207-2225Crossref PubMed Scopus (52) Google Scholar). RNA was isolated from cultured keratinocytes which had been treated with BMP2 and BMP4. RT-PCR was performed as described above. Result of RT-PCR is expected to give PCR product with 377 bp in type IIA mRNA and 171 bp in type IIB mRNA. To confirm the expression of proα1(II) mRNA species in which exons 1 and 2 are contiguous, a 24 bp oligonucleotide probe (5′TGCCAGCCTCCTGGACATCCTGGC3′) corresponding to 12 nucleotides at 3′ end of exon 1 and 12 nucleotides at the 5′ end of exon 2 was synthesized as previously described (Ryan and Sandell, 1990Ryan M.C. Sandell L.J. Differential expression of a cysteine-rich domain in the amino-terminal propeptide of type II (cartilage) procollagen by alternative splicing of mRNA.J Biol Chem. 1990; 265: 10334-10339Abstract Full Text PDF PubMed Google Scholar), then radioactively labeled at 5′ end with [γ-32P]ATP (210 TBq per mmol, Amersham, Biosciences Corp., Piscataway, New Jersey) and T4 polynucleotide kinase. RT-PCR products which had been resolved on 1.5% agarose gel were capillary-transferred onto nylon filters, then hybridized at 27°C overnight with the exons 1–2-specific oligonucleotide probe in a solution of 50 mM Tris-HCl, pH 7.5, 1 M NaCl, 10% dextran sulfate, 1% SDS, and 100 μg per mL tRNA. The filters were washed twice with 1 × SSC at 55°C for 2 h, then subjected to autoradiography.
Referência(s)