Changing Pattern of Deiminated Proteins in Developing Human Epidermis
2003; Elsevier BV; Volume: 120; Issue: 5 Linguagem: Inglês
10.1046/j.1523-1747.2003.12138.x
ISSN1523-1747
AutoresYukiko Tsuji, Masashi Akiyama, Ken Arita, Tatsuo Senshu, Hiroshi Shimizu,
Tópico(s)Nail Diseases and Treatments
ResumoPeptidylarginine deiminases are widely distributed, calcium-ion-dependent enzymes that convert arginine residues of proteins into citrulline residues. This reaction, deimination, is thought to be an important event during the final stage of epidermal differentiation, possibly associated with integration and disintegration of keratin filaments. To elucidate the possible roles of protein deimination during human epidermal development we investigated localization of deiminated proteins using anti-citrulline peptide antibody, which preferentially recognizes citrulline residues in the V subdomains of keratin 1, and anti-chemically modified citrulline antibody, which enables detection of citrulline residues independent of amino acid sequences. Anti-chemically modified citrulline antibody, but not anti-citrulline peptide antibody stained the periderm in two-layered epidermis of 49 d and 57 d estimated gestational age. In the stratified epidermis of 88 d, 96 d, and 108 d estimated gestational age fetal skin, anti-citrulline peptide antibody and anti-chemically modified citrulline antibody staining was seen in the periderm and intermediate cell layers. After periderm cells regressed and keratinization began in the interfollicular epidermis, anti-citrulline peptide antibody and anti-chemically modified citrulline antibody were restricted to the cornified cell layers of the interfollicular epidermis, similar to the distribution patterns of that in adult epidermis. Postembedding immunoelectron microscopy showed anti-citrulline peptide antibody immunogold labeling over the cytoplasmic intermediate filament network in the periderm and the intermediate cell layers. These results demonstrate an orderly formation of deiminated proteins in different layers of embryonic epidermis and suggest important roles for peptidylarginine deiminases in human epidermal morphogenesis. Peptidylarginine deiminases are widely distributed, calcium-ion-dependent enzymes that convert arginine residues of proteins into citrulline residues. This reaction, deimination, is thought to be an important event during the final stage of epidermal differentiation, possibly associated with integration and disintegration of keratin filaments. To elucidate the possible roles of protein deimination during human epidermal development we investigated localization of deiminated proteins using anti-citrulline peptide antibody, which preferentially recognizes citrulline residues in the V subdomains of keratin 1, and anti-chemically modified citrulline antibody, which enables detection of citrulline residues independent of amino acid sequences. Anti-chemically modified citrulline antibody, but not anti-citrulline peptide antibody stained the periderm in two-layered epidermis of 49 d and 57 d estimated gestational age. In the stratified epidermis of 88 d, 96 d, and 108 d estimated gestational age fetal skin, anti-citrulline peptide antibody and anti-chemically modified citrulline antibody staining was seen in the periderm and intermediate cell layers. After periderm cells regressed and keratinization began in the interfollicular epidermis, anti-citrulline peptide antibody and anti-chemically modified citrulline antibody were restricted to the cornified cell layers of the interfollicular epidermis, similar to the distribution patterns of that in adult epidermis. Postembedding immunoelectron microscopy showed anti-citrulline peptide antibody immunogold labeling over the cytoplasmic intermediate filament network in the periderm and the intermediate cell layers. These results demonstrate an orderly formation of deiminated proteins in different layers of embryonic epidermis and suggest important roles for peptidylarginine deiminases in human epidermal morphogenesis. peptidylarginine deiminase anti-citrulline peptide antibody anti-chemically modified citrulline antibody Peptidylarginine deiminases (PAD) are a group of enzymes that convert protein arginine residues to citrulline residues. This reaction is called "deimination". There are five known PAD subtypes so far reported: type I in the epidermis (Takahara et al., 1989Takahara H. Tsuchida M. Kusubata M. Akutsu K. Tagami S. Sugawara K. Peptidylarginine deiminase of the mouse. Distribution, properties, and immunocytochemical localization.J Biol Chem. 1989; 264: 13361-13368Abstract Full Text PDF PubMed Google Scholar;Ishigami et al., 1998Ishigami A. Kuramoto M. Yamada M. Watanabe K. Senshu T. Molecular cloning of two novel types of peptidylarginine deiminase cDNAs from retinoic acid-treated culture of a newborn rat keratinocyte cell line.FEBS Lett. 1998; 433: 113-118Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar;Rus'd et al., 1999Rus'd A.A. Ikejiri Y. Ono H. Yonekawa T. Shiraiwa M. Kawada A. Takahara H. Molecular cloning of cDNAs of mouse peptidylarginine deiminase type I, type III and type IV, and the expression pattern of type I in mouse.Eur J Biochem. 1999; 259: 660-669Crossref PubMed Scopus (65) Google Scholar); (ii) type II in the brain and muscle (Takahara et al., 1989Takahara H. Tsuchida M. Kusubata M. Akutsu K. Tagami S. Sugawara K. Peptidylarginine deiminase of the mouse. Distribution, properties, and immunocytochemical localization.J Biol Chem. 1989; 264: 13361-13368Abstract Full Text PDF PubMed Google Scholar;Watanabe and Senshu, 1989Watanabe K. Senshu T. Isolation of cDNA clones encoding rat skeletal muscle peptidylarginine deiminase.J Biol Chem. 1989; 264: 15255-15260Abstract Full Text PDF PubMed Google Scholar); (iii) type III in the hair follicles (Nishijyo et al., 1997Nishijyo T. Kawada A. Kanno T. Shiraiwa M. Takahara H. Isolation and molecular cloning of epidermal-and hair follicle-specific peptidylarininge deiminase (type III) from rat.J Biochem (Tokyo). 1997; 121 (875): 968Crossref Scopus (43) Google Scholar;Rogers et al., 1997Rogers G. Winter B. McLaughlan C. Powell B. Nesci T. Peptidylarignine deiminase of the hair follicle: Characterization, localization, and function in keratinizing tissues.J Invest Dermatol. 1997; 108: 700-707Crossref PubMed Scopus (67) Google Scholar;Rus'd et al., 1999Rus'd A.A. Ikejiri Y. Ono H. Yonekawa T. Shiraiwa M. Kawada A. Takahara H. Molecular cloning of cDNAs of mouse peptidylarginine deiminase type I, type III and type IV, and the expression pattern of type I in mouse.Eur J Biochem. 1999; 259: 660-669Crossref PubMed Scopus (65) Google Scholar;Kanno et al., 2000Kanno T. Kawada A. Yamanouchi J. et al.Human peptidylarginine deiminase type III. Molecular cloning and nucleotide sequence of the cDNA, properties of the recombinant enzyme, and immunohistochemical localization in human skin.J Invest Dermatol. 2000; 115: 813-823Crossref PubMed Scopus (112) Google Scholar); (iv) type IV originated from rat epidermis (Ishigami et al., 1998Ishigami A. Kuramoto M. Yamada M. Watanabe K. Senshu T. Molecular cloning of two novel types of peptidylarginine deiminase cDNAs from retinoic acid-treated culture of a newborn rat keratinocyte cell line.FEBS Lett. 1998; 433: 113-118Abstract Full Text Full Text PDF PubMed Scopus (42) Google Scholar;Yamakoshi et al., 1998Yamakoshi A. Ono H. Nishijyo T. Shiraiwa M. Takahara H. Cloning of cDNA encoding a novel isoform (type IV) of peptidylarginine deiminase from rat epidermis.Biochim Biophys Acta. 1998; 1386: 227-232Crossref PubMed Scopus (25) Google Scholar); and (v) type V induced in differentiated human promyelocytic leukemia HL-60 cells (Nakashima et al., 1999Nakashima K. Hagiwara T. Ishigami A. et al.Molecular characterization of peptidylarginine deiminase in HL-60 cells induced by retinoic acid and 1a,25-dihidrozyvitamin D3.J Biol Chem. 1999; 274: 27786-27792Crossref PubMed Scopus (157) Google Scholar). All these PAD show a definite requirement for calcium ions. Deimination is one of the most important post-translational modifications during epidermal terminal differentiation. As the positive charge of arginine residues is irreversibly lost, deimination of even a single arginine residue can cause significant changes in the structure and function of target proteins. For this reason, PAD activities are thought to be under strict regulation in vivo. In the human epidermis, deiminated proteins are localized in the cornified cell layers (Senshu et al., 1996Senshu T. Kan S. Ogawa H. Manabe M. Asaga H. Preferential deimination of keratin K1 and filaggrin during the terminal differentiation of human epidermis.Biochem Biophys Res Commun. 1996; 225: 712-719Crossref PubMed Scopus (128) Google Scholar). The major protein substrate of PAD is keratin (K) 1 and minor substrates include filaggrin and K10 (Senshu et al., 1995Senshu T. Akiyama K. Kan S. Asaga H. Ishigami A. Manabe M. Detection of deiminated proteins in rat skin: Probing with a monospecific antibody after modification of citrulline residues.J Invest Dermatol. 1995; 105: 163-169Crossref PubMed Scopus (124) Google Scholar), suggesting deimination of these proteins might be important during the final stages of epidermal differentiation. Virtually, no data are available regarding the possible involvement of protein deimination in the developing human fetal epidermis. The purpose of this study was to immunocytochemically examine when, where, and what kinds of epidermal proteins are deiminated during human epidermal development. We have used (i) anti-citrulline peptide antibody (ACP), which preferentially reacts with citrulline residues in the V subdomains of keratin 1 in human epidermis (Ishida-Yamamoto et al., 2000Ishida-Yamamoto A. Senshu T. Takahashi H. Akiyama K. Nomura K. Iizuka H. Decreased deiminated keratin K1 in psoriatic hyperproliferative epidermis.J Invest Dermatol. 2000; 114: 701-705Crossref PubMed Scopus (90) Google Scholar), and (ii) anti-chemically modified citrulline antibody (AMC), which recognizes chemically modified citrulline residues in proteins regardless of the amino acid sequence. Here we show a highly regulated orderly formation of deiminated proteins in different layers of embryonic epidermis suggesting an important role for deiminated proteins in epidermal morphogenesis. In addition, we discuss similarity of the regression of the embryonic periderm with the keratinization of adult epidermis in terms of protein deimination. Human embryonic and fetal skin specimens were obtained from abortuses of 49 d, 57 d, 88 d, 96 d, 108 d, and 163 d estimated gestational age (EGA) through the Central Laboratory of Human Embryology at the University of Washington, Seattle with the approval of the Human Subjects Review Board and in accordance with the United States Department of Health, Education, and Welfare policies. For immunoelectron microscopic observation, skin specimens from a fetus of 105 d EGA was used. EGA was determined from the maternal history, fetal measurements, and comparative histologic appearance of epidermis (Shepard, 1975Shepard T.H. Normal and abnormal growth patterns. Growth and development of the human embryo and fetus.in: Gardner L.I. Endocrine and Genetic Diseases of Childhood and Adolescence. 108. WB Saunders, Philadelphia1975: 1-8Google Scholar;Holbrook, 1979Holbrook K.A. Human epidermal embryogenesis.Int J Dermatol. 1979; 18: 329-356PubMed Google Scholar;Mercer et al., 1987Mercer B.M. Sklar S. Shariatmadar A. Gillieson M.S. D'Alton M.E. Fetal foot length as a predictor of gestational age.Am J Obstet Gynecol. 1987; 156: 350-355Abstract Full Text PDF PubMed Scopus (97) Google Scholar). Normal adult human skin was obtained at the time of skin surgery. Normal human epidermal keratinocytes (NHEK) and an immortalized keratinocyte cell line (HaCaT cells) were cultured on eight-chamber glass slides each in low concentration calcium medium (Ca2+=0.15 mM in keratinocyte growth medium). In order to induce keratinization, both cells were cultured with high concentration calcium medium (Ca2+=1.2 mM) for 48 h. Cultured epidermal sheets over a fibroblast feeder layer (Japan Tissue Engineering Co. J-TEC, Aichi, Japan) were also used for the study. Immediately before the immunolabeling, all specimens were fixed with precooled 1:1 acetone/methanol for 10 min, and treated with 0.05% Triton X at 37°C for 10 min. Preparation and affinity purification of the rabbit polyclonal antibody (ACP) against a deiminated undecapeptide corresponding to the identified deimination site in the V2 subdomain of mouse K1 (amino acid residues 545GSSGGGRGGSS555) have been described previously (Senshu et al., 1999Senshu T. Akiyama K. Ishigami A. Nomura K. Studies on specificity of peptidylarginine deiminase reactions using an immunochemical probe that recognizes an enzymatically deiminated partial sequence of mouse keratin K1.J Dermatol Sci. 1999; 21: 113-126Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar). ACP was shown to react not only with deiminated mouse K1, but also with deiminated human K1 (Ishida-Yamamoto et al., 2000Ishida-Yamamoto A. Senshu T. Takahashi H. Akiyama K. Nomura K. Iizuka H. Decreased deiminated keratin K1 in psoriatic hyperproliferative epidermis.J Invest Dermatol. 2000; 114: 701-705Crossref PubMed Scopus (90) Google Scholar). A monospecific antibody to chemically modified citrulline residues (AMC) was prepared as described previously (Senshu et al., 1992Senshu T. Sato T. Inoue T. Akiyama K. Asaga H. Detection of citrulline residue in deiminated proteins on polyvinylidene difluoride membrane.Anal Biochem. 1992; 203: 94-100Crossref PubMed Scopus (143) Google Scholar). Other primary antibodies were mouse monoclonal antibody, 34βB4, which recognizes human suprabasal keratins (Novocastra, Newcastle upon Tyne, U.K.), and an anti-human filaggrin antibody (Biomedical Technologies Inc., Stoughten, MA). The secondary antibodies were fluorescein isothiocyanate (FITC)-conjugated goat anti-rabbit IgG (H+L) antibody (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) and tetrarhodamine isothiocyanate-conjugated anti-mouse IgG1 antibody. Cryostat sections of skin specimens and cultured keratinocytes on glass slides were incubated with ACP (3.8 μg per ml) alone or with a mixture of ACP and 34βB4 (1: 10 dilution in phosphate-buffered saline), or a mixture of ACP and anti-human filaggrin (1: 500 dilution in phosphate-buffered saline). After phosphate-buffered saline washing, the bound antibodies were detected using FITC-conjugated goat anti-rabbit IgG (H+L) antibody and tetrarhodamine isothiocyanate-conjugated goat anti-mouse IgG1 antibody. For all labeling, specimens were incubated with the secondary antibody alone, and incubation with irrelevant primary antibodies or nonimmunized sera served as negative controls. In some cases, nuclei were counterstained with propidium iodide (Dojindo Laboratories, Kumamoto, Japan). Image collection of the immunohistochemical staining was performed by a confocal laser scanning microscope with software Fluoview version 2.0 (Olympus America Inc., Melville, NY). AMC staining was performed as previously described (Senshu et al., 1995Senshu T. Akiyama K. Kan S. Asaga H. Ishigami A. Manabe M. Detection of deiminated proteins in rat skin: Probing with a monospecific antibody after modification of citrulline residues.J Invest Dermatol. 1995; 105: 163-169Crossref PubMed Scopus (124) Google Scholar). Briefly, skin cryostat sections were fixed with 5% glutaraldehyde in phosphate-buffered saline for 15 min. Sections were then incubated in 0.0125% FeCl3, 2.3 M H2SO4, 1.5 M H3PO4, 0.25% diacetyl monoxime, and 0.125% anti-pyrine (modification medium) at 37°C for 3 h to modify chemically the citrulline residues. Control sections were incubated in the medium depleted of diacetyl monoxime and anti-pyrine. Immunoperoxidase staining of the modified proteins was performed with AMC and a Histfine SAB-PO kit (Nitirei, Tokyo, Japan) using 3,3′-diaminobenzidine as a chromogenic substrate. Fetal skin samples were cryofixed, cryosubstituted, and embedded in Lowicryl K11M resin (Chemische Werke Lowi, Waldkraiburg, Germany) as previously described (Shimizu et al., 1989Shimizu H. McDonald J.N. Kennedy A.R. Eady R.A.J. Demonstration of intra- and extracellular localization of bullous pemphigoid antigen using cryofixation and freeze substitution for postembedding immunoelectron microscopy.Arch Dermatol Res. 1989; 281: 443-448Crossref PubMed Scopus (113) Google Scholar). Ultrathin sections were cut, collected on formvar-coated nickel grids, and immunostained as described previously (Ishida-Yamamoto et al., 1996Ishida-Yamamoto A. Eady R.A.J. Watt F.M. Roop D.R. Hohl D. Iizuka H. Immunoelectron microscopic analysis of cornified cell envelope formation in normal and psoriatic epidermis.J Histochem Cytochem. 1996; 44: 167-175Crossref PubMed Scopus (73) Google Scholar). ACP was used as the primary antibody and 5 nm gold-conjugated goat anti-rabbit IgG was used as the secondary antibody. Although no ACP antibody staining was observed in the early two-layered epidermis (49 d and 57 d EGA) (Figure 1), significant AMC staining was detectable in the periderm cytoplasm at 49 d EGA (Figure 2). In the three-layered or more stratified epidermis (88 d, 96 d, and 108 d EGA), both the periderm and intermediate cells became ACP and AMC positive for both in the cytoplasm, but the basal cells remained negative for both techniques. After the regression of periderm and the commencement of keratinization in the interfollicular epidermis (163 d EGA), staining of ACP and AMC became restricted to the cornified cells, and were absent from the spinous and basal cells. In adult epidermis, both ACP and AMC staining was restricted to the cornified cell layers as described previously (Ishida-Yamamoto et al., 2000Ishida-Yamamoto A. Senshu T. Takahashi H. Akiyama K. Nomura K. Iizuka H. Decreased deiminated keratin K1 in psoriatic hyperproliferative epidermis.J Invest Dermatol. 2000; 114: 701-705Crossref PubMed Scopus (90) Google Scholar).Figure 2AMC staining was detected in the periderm of two-layered epidermis. In the early two-layered epidermis at 49 d EGA, staining for AMC was observed in the cytoplasm of the periderm (arrowheads; brown, immunoperoxidase). In the stratified epidermis of 88 d, 96 d, and 108 d EGA, staining for AMC was observed both in the cytoplasm of the periderm and intermediate cells. After the regression of periderm cells on 163 d EGA, staining for AMC was restricted to the stratum corneum. In normal adult skin, AMC staining is limited in the stratum corneum. Black dots: basement membrane zone. Scale bars: 50 μm.View Large Image Figure ViewerDownload (PPT) In the ACP-negative two-layered epidermis, neither 34βB4 nor anti-filaggrin staining was observed (data not shown). In the stratified epidermis, ACP-positive intermediate layer cells were also positive for 34βB4 demonstrating the colocalization of ACP-positive proteins with the suprabasal keratins (Figure 3, 108 d EGA). During this period, filaggrin was not yet expressed. After the periderm cells regressed, ACP staining was restricted to cornified cells and partly overlapped with 34βB4 keratin staining (Figure 3, 163 d EGA). At this point, the anti-filaggrin antibody began to stain the upper part of granular cell layers and the cornified cell layers. This staining profile of the keratinized fetal epidermis resembled that of the adult epidermis (Figure 3, adult). In immunoelectron microscopy, ACP immunogold labeling was observed over the cytoplasmic filament network in the periderm (Figure 4a) and the intermediate cells (Figure 4b) at 105 d EGA. No ACP labeling was observed in the basal cells (data not shown). In the negative control specimens, almost no immunogold labeling was seen. Neither ACP nor 34βB4 staining was observed in NHEK or HaCaT cells cultured in low calcium medium. When they were induced to differentiate by culturing in high calcium medium, weak 34βB4 staining only became detectable in large flattened cells, which remained negative for ACP (data not shown). ACP-positive staining, however, was detectable in cells covering the upper surface of human epidermal sheets cultured on a fibroblast feeder layer (Figure 5). These cells were also stained with 34βB4 and anti-filaggrin (data not shown). In this study, we investigated the formation of deiminated proteins in developing human epidermis using two antibodies against deiminated proteins. One is ACP, which preferentially recognizes citrulline residues in the V subdomains of keratin and the other is AMC, which enables the detection of citrulline residues independent of amino acid sequences. The V subdomains of K1 contain only a few arginine residues that are the target sites for PAD in highly glycine-rich sequences containing scattered phenylalanine and tyrosine residues. An important finding in this study was the cytoplasmic staining of the periderm in the two-layered epidermis with AMC, both layers of which were not stained with ACP at all (Figure 6). This suggests some deimination of periderm proteins at this developmental stage. Judging from the specificity of ACP, AMC-positive proteins are probably deiminated at arginine residues with specific amino acid sequences unrelated to the V subdomains of K1. In adult human epidermis, AMC-positive deiminated proteins include K1/K10 and filaggrin (Senshu et al., 1996Senshu T. Kan S. Ogawa H. Manabe M. Asaga H. Preferential deimination of keratin K1 and filaggrin during the terminal differentiation of human epidermis.Biochem Biophys Res Commun. 1996; 225: 712-719Crossref PubMed Scopus (128) Google Scholar). Actually, neither layer stained with the monoclonal antibody 34βB4 against the suprabasal keratins (K1 and K10). In rat newborn epidermis, AMC staining revealed that K6 (56 kDa)/K16 (48 kDa) (the hyperproliferative keratins), trichohyalin, and filaggrin were deiminated (Senshu et al., 1995Senshu T. Akiyama K. Kan S. Asaga H. Ishigami A. Manabe M. Detection of deiminated proteins in rat skin: Probing with a monospecific antibody after modification of citrulline residues.J Invest Dermatol. 1995; 105: 163-169Crossref PubMed Scopus (124) Google Scholar). In the periderm of human fetal epidermis, neither K6/K16 nor filaggrin are expressed, although 40 kDa and 52 kDa simple epithelial keratins are expressed (Moll et al., 1982Moll R. Moll I. Wiest W. Changes in the pattern of cytokeratin polypeptides in epidermis and hair follicles during skin development in human fetuses.Differentiation. 1982; 23: 170-178Crossref PubMed Scopus (176) Google Scholar;Dale et al., 1985Dale B.A. Holbrook K.A. Kimball J.R. Hoff M. Sun T. Expression of epidermal keratins and filaggrin during human fetal skin development.J Cell Biol. 1985; 101: 1257-1259Crossref PubMed Scopus (216) Google Scholar). Thus, AMC-positive, ACP-negative deiminated protein(s) in early periderm are unlikely to be deiminated K6/K16 or filaggrin, however, this remains to be clarified. The staining pattern with AMC (Figure 2) suggested that there was some cornified cell envelope reactivity in addition to cytoplasmic staining. As the AMC antibody is not specific for the V domain of K1, we cannot exclude the possibility that AMC-positive deiminated proteins in developing human epidermis may include certain cornified cell envelope proteins; however, there is no known evidence that cornified cell envelope proteins, including involucrin, are deiminated by PAD. Interestingly, in the three or more layered stratified epidermis, both the cytoplasm of periderm and intermediate cells became AMC and ACP positive. This was confirmed by immunoelectron microscopy showing ACP immunogold particles over the cytoplasmic filament networks of both periderm and intermediate cells. The expression of suprabasal keratins was observed in the intermediate cells as previously reported (Moll et al, 1983;Dale et al., 1985Dale B.A. Holbrook K.A. Kimball J.R. Hoff M. Sun T. Expression of epidermal keratins and filaggrin during human fetal skin development.J Cell Biol. 1985; 101: 1257-1259Crossref PubMed Scopus (216) Google Scholar). Thus, the major substrate of PAD in the intermediate cell layer could be K1 as in adult epidermis. Alternatively, the ACP staining may reflect deimination of other cytoskeletal proteins bearing regional sequence similarities to the V subdomains of K1. The ACP-positive staining of the periderm cells at the three or more layered epidermis may also be accounted for, by such deiminated proteins. Similar, though as yet unidentified ACP-positive proteins were reported in fetal and early neonatal mouse epidermis (Senshu et al., 1999Senshu T. Akiyama K. Ishigami A. Nomura K. Studies on specificity of peptidylarginine deiminase reactions using an immunochemical probe that recognizes an enzymatically deiminated partial sequence of mouse keratin K1.J Dermatol Sci. 1999; 21: 113-126Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar). Using a database search, the only known cytoskeletal or envelope proteins that have any sequence similarity to the K1 V subdomain were K10, K2e, and loricrin, but loricrin has no arginine residues in its amino acid sequence. Thus, as the ACP-positive deiminated protein(s) in the intermediate cells, K10 and K2e could be the candidate molecules. In periderm cells, neither K10 nor K2e (65.5 kDa) are expressed (Dale et al., 1985Dale B.A. Holbrook K.A. Kimball J.R. Hoff M. Sun T. Expression of epidermal keratins and filaggrin during human fetal skin development.J Cell Biol. 1985; 101: 1257-1259Crossref PubMed Scopus (216) Google Scholar;Collin et al., 1992Collin C. Moll R. Kubicka S. Ouhayoun J.P. Franke W.W. Characterization of human cytokeratin 2, an epidermal cytoskeletal protein synthesized late during differentiation.Exp Cell Res. 1992; 202: 132-141Crossref PubMed Scopus (114) Google Scholar) and there are likely to be other candidate substrates yet to be identified in the periderm. After the regression of periderm and the commencement of keratinization in the interfollicular epidermis (163 d EGA), deiminated proteins were localized in the cornified layers as observed in adult human epidermis. The regulated expression of deiminated proteins in different layers of embryonic epidermis reported here represents an interesting feature of developing human epidermis that distinguishes from normal adult epidermis. In the latter, deiminated proteins are only observed in cornified cell layers, highlighting the fact that deimination is under strict regulation during the keratinization of normal adult epidermis. Formation of deiminated proteins in the periderm and intermediate cells of the fetal skin suggests not only gene expression but also activation of the as yet unidentified PAD. Interestingly, this appears to occur in cells partially lacking morphologic features of keratinization. It has recently been shown that several proteins essential for the formation of the cornified cell envelope appear in the periderm and the intermediate layer before the onset of keratinization (Lee et al., 1999Lee S.C. Lee J.B. Kook J.P. Seo J.J. Nam K.I. Park S.S. Kim P.K. Expression of differentiation markers during fetal skin development in humans: Immunohisto-chemical studies on the precursor proteins forming the cornified cell envelope.J Invest Dermatol. 1999; 112: 882-886Crossref PubMed Scopus (29) Google Scholar;Akiyama et al., 1999Akiyama M. Smith L.T. Yoneda K. Holbrook K.A. Hohl D. Shimizu H. Periderm cells from cornified cell envelope in their regression process during human epidermal development.J Invest Dermatol. 1999; 112: 903-909Crossref PubMed Scopus (57) Google Scholar). Cornified cell envelope precursor proteins, involucrin and loricrin are restricted to periderm cells. Periderm cells are destined to regress during the epidermal development. Periderm regression has been postulated to be similar in terms of cell loss and cornified cell envelope formation to epidermal keratinization (Akiyama et al., 1999Akiyama M. Smith L.T. Yoneda K. Holbrook K.A. Hohl D. Shimizu H. Periderm cells from cornified cell envelope in their regression process during human epidermal development.J Invest Dermatol. 1999; 112: 903-909Crossref PubMed Scopus (57) Google Scholar). Protein deimination in the periderm may be an important process in the regression of the periderm cell as it is in the epidermal keratinization. In addition, protein deimination in intermediate cells may be a preparatory process for keratinization as the expression of proteins involved in the formation of cornified cell envelope. Types of PAD expressed and their target proteins remain to be identified and may provide fruitful areas for future research. The present keratinocyte culture system cannot be considered as a model of fetal epidermal morphogenesis and the culture system simply corresponds to near-normal epidermal differentiation. Keratinocytes cultured in low Ca2+ conditions showed neither suprabasal keratin expression nor deiminated protein formation. Cells cultured in the high Ca2+ condition expressed suprabasal keratins, without the formation of deiminated proteins. It has been shown by immunoelectron microscopy that K1 is not deiminated in the first few cornified cell layers of adult human epidermis (Ishida-Yamamoto et al., 2002Ishida-Yamamoto A. Senshu T. Eady R.A.J. Takahashi H. Shimizu H. Akiyama M. Iizuka H. Sequential reorganization of cornified cell keratin filaments involving filaggrin-mediated compaction and K1 deimination.J Invest Dermatol. 2002; 118: 282-287Crossref PubMed Scopus (65) Google Scholar). In this context, keratinized cultured cells under high Ca2+ conditions might be at a similar stage of differentiation to the first few cornified cell layers. Whereas, deiminated K1 was detected in keratinocytes covering the uppermost surface of the epidermal sheet cultured on a fibroblast feeder layer. Thus, the superficial cells in the epidermal sheet are thought to be more differentiated and closer to the in vivo keratinizing epidermal cells, compared with cultured cells without a feeder layer. In conclusion, the results of this study reveal a highly ordered sequence of deiminated proteins in different layers of developing human epidermis. These findings indicate that the keratinization corresponding to that of the adult epidermis occurs in the periderm during the periderm regression phase and suggests that deimination by PAD may be a key event in the development of epidermal differentiation. We thank Prof. John E. Olerud, Ms Marcia L. Usui, and Mr Robert A. Underwood for their generous help in this project; Dr James R. McMillan for his constructive comments; Mr Hideki Nakamura, Ms Megumi Sato, and Ms Kaori Sakai for their fine technical assistance. This work was supported in part by a Grant-in-Aid for Scientific Research (A) (no. 13357008) and (B) (no. 12470175) given to H.S., and (C) (no. 12670839) to M.A. from Japan Society for the Promotion of Science.
Referência(s)