Barrier Formation in the Human Fetus is Patterned
1999; Elsevier BV; Volume: 113; Issue: 6 Linguagem: Inglês
10.1046/j.1523-1747.1999.00800.x
ISSN1523-1747
AutoresMatthew J. Hardman, Mark W. J. Ferguson, Carolyn Byrne, Lynette Moore,
Tópico(s)Dermatology and Skin Diseases
ResumoWe recently demonstrated patterned stratum corneum maturation and skin barrier formation during fetal development in rodents and rabbit. The presence of skin patterning in these mammals led us to predict patterned barrier formation during human infant development. Here we extend our mammalian study and demonstrate patterned stratum corneum development and skin barrier formation in the pre-term human infant. Surprisingly, we show initiation of human barrier regionally as early as 20–24 wk gestational age (22–26 wk menstrual age), bringing barrier formation close to the time of periderm disaggregation. We use the mouse model to show that patterns of periderm disaggregation mirrors barrier formation. Periderm disaggregation follows and recapitulates barrier pattern, suggesting a relationship between the processes. This work reveals regional patterning in skin maturation and barrier formation in the human infant and demonstrates that initiation of human skin barrier formation in utero coincides with the current lower limit of viability of the pre-term infant. We recently demonstrated patterned stratum corneum maturation and skin barrier formation during fetal development in rodents and rabbit. The presence of skin patterning in these mammals led us to predict patterned barrier formation during human infant development. Here we extend our mammalian study and demonstrate patterned stratum corneum development and skin barrier formation in the pre-term human infant. Surprisingly, we show initiation of human barrier regionally as early as 20–24 wk gestational age (22–26 wk menstrual age), bringing barrier formation close to the time of periderm disaggregation. We use the mouse model to show that patterns of periderm disaggregation mirrors barrier formation. Periderm disaggregation follows and recapitulates barrier pattern, suggesting a relationship between the processes. This work reveals regional patterning in skin maturation and barrier formation in the human infant and demonstrates that initiation of human skin barrier formation in utero coincides with the current lower limit of viability of the pre-term infant. estimated gestational age transepidermal water loss The skin permeability barrier is essential for terrestrial life and forms during late gestation (reviewed inRoop, 1995Roop D.R. Defects in the barrier.Science. 1995; 267: 474Crossref PubMed Scopus (167) Google Scholar;Williams et al., 1998Williams M.L. Hanley K. Elias P.M. Feingold K.R. Ontogeny of the epidermal permeability barrier.J Invest Dermatol. 1998; 3S: 75-79Abstract Full Text PDF Scopus (62) Google Scholar). Human infants born before 30-32 wk gestation can suffer from problems related to barrier deficiency including desiccation (Lorenz et al., 1982Lorenz J.M. Kleinman L.I. Kotagal U.R. Reller M.D. Water balance in very low-birth-weight infants: relationship to water and sodium intake and effect on outcome.Fetal Neonatal Med. 1982; 101: 423-432Google Scholar), thermoregulatory problems (Harpin and Rutter, 1982Harpin V.A. Rutter N. Sweating in preterm babies.J Pediatr. 1982; 100: 614-618Abstract Full Text PDF PubMed Scopus (44) Google Scholar;Hammarlund et al., 1986Hammarlund K. Stromberg B. Sedin G. Heat loss from the skin of preterm and fullterm newborn infants during the first weeks after birth.Biol Neonate. 1986; 50: 1-10Crossref PubMed Scopus (46) Google Scholar), microbial infection (Leyden, 1982Leyden J. Neonatal Skin: Structure and Function. Marcel Dekker, New York1982Google Scholar;Askin, 1995Askin D.F.L. Bacterial and fungal infections in the neonate.J Obstet Gynecol Neonatal Nurs. 1995; 24: 635-643Crossref PubMed Scopus (3) Google Scholar;Rowen et al., 1995Rowen J.L. Atkins J.T. Levy M.L. Baer S.C. Baker C.J. Invasive fungal dermatitis in the ≤ 1000-gram neonate.Paediatrics. 1995; 95: 682-687PubMed Google Scholar), and accidental poisoning (reviewed inNachman and Esterly, 1971Nachman R.L. Esterly N.B. Increased skin permeability in preterm infants.J Pediatr. 1971; 79: 628-632Abstract Full Text PDF PubMed Scopus (100) Google Scholar;Goutieres and Alcardi, 1977Goutieres F. Alcardi J. Accidental percutaneous hexachlorophene intoxication in children.Br Med J. 1977; ii: 663-665Crossref Scopus (27) Google Scholar;Wester et al., 1982Wester R. Maibach H.I. Comparative percutaneous toxicity in neonatal skin.in: Maibach H.I. Boisits E.K. Neonatal Skin: Structure and Function. Marcel Dekker, New York1982: 137-148Google Scholar). The skin barrier is conferred by the stratum corneum and skin barrier development correlates with stratum corneum development (Evans and Rutter, 1986Evans N.J. Rutter N. Development of the epidermis in the newborn.Biol Neonate. 1986; 49: 74-80Crossref PubMed Scopus (198) Google Scholar;Azsterbaum et al., 1992Azsterbaum M. Menon G.K. Feingold K.R. Williams M.L. Ontogeny of the epidermal barrier to water loss in the rat: correlation of function with stratum corneum structure and lipid content.Pediatr Res. 1992; 31: 308-317Crossref PubMed Scopus (74) Google Scholar;Hardman et al., 1998Hardman M.J. Sisi P. Banbury D.N. Byrne C. Patterned acquisition of barrier function during development.Development. 1998; 125: 1541-1552Crossref PubMed Google Scholar). Human skin development from single-layered surface ectoderm to multilayered keratinized epidermis has been well documented (reviewed inBreathnach, 1975Breathnach A.S. Aspects of epidermal ultrastructure.J Invest Dermatol. 1975; 65: 2-15Crossref PubMed Scopus (66) Google Scholar;Holbrook and Odland, 1980Holbrook K.A. Odland G.F. Regional development of the human epidermis in the first trimester embryo and the second trimester fetus (ages related to the time of amniocentesis and fetal biopsy).J Invest Dermatol. 1980; 74: 161-168Crossref PubMed Scopus (55) Google Scholar;Dale et al., 1985Dale B.A. Holbrook K.A. Kimball J.R. Hoff M. Sun T.T. Expression of epidermal keratins and filaggrin during human fetal skin development.J Cell Biol. 1985; 101: 1257-1269Crossref PubMed Scopus (221) Google Scholar;Foster et al., 1988Foster C.A. Bertram J.F. Holbrook K.A. Morphometric and statistical analyses describing the in utero growth of human epidermis.Anat Rec. 1988; 222: 201-2–206Crossref PubMed Scopus (7) Google Scholar;Holbrook et al., 1994Holbrook K.A. Ultrastructure of the epidermis.in: Leigh I.M. Birgitte Lane E. Watt F.M. The Keratinocyte Handbook. Cambridge University Press, Cambridge1994: 3-39Google Scholar) and ''keratinization'' or stratum corneum formation reported at 22 wk in epidermis of head/scalp and palmar/plantar skin (Holbrook and Odland, 1975Holbrook K.A. Odland G.F. The fine structure of developing human epidermis: light, scanning and transmission electron microscopy of the periderm.J Invest Dermatol. 1975; 65: 16-38Crossref PubMed Scopus (139) Google Scholar) and by 25 wk in the interfollicular epidermis of the remainder of the body (Foster et al., 1988Foster C.A. Bertram J.F. Holbrook K.A. Morphometric and statistical analyses describing the in utero growth of human epidermis.Anat Rec. 1988; 222: 201-2–206Crossref PubMed Scopus (7) Google Scholar). A transitory fetal layer, the periderm, interfaces between developing epidermis and amniotic fluid (Hashimoto et al., 1966Hashimoto K. Gross B.G. Dibella R.J. Lever W.F. The ultrastructure of the skin of human embryos: IV. the epidermis.J Invest Dermatol. 1966; 106: 317-335Crossref Scopus (58) Google Scholar;Hoyes, 1968Hoyes A.D. Electron microscopy of the surface layer (periderm) of human infant skin.J Anat. 1968; 103: 321-336PubMed Google Scholar). Periderm differentiates in tandem with epidermal development and is sloughed into amniotic fluid at about 24–25 wk, as the skin keratinizes (Holbrook and Odland, 1975Holbrook K.A. Odland G.F. The fine structure of developing human epidermis: light, scanning and transmission electron microscopy of the periderm.J Invest Dermatol. 1975; 65: 16-38Crossref PubMed Scopus (139) Google Scholar;Foster et al., 1988Foster C.A. Bertram J.F. Holbrook K.A. Morphometric and statistical analyses describing the in utero growth of human epidermis.Anat Rec. 1988; 222: 201-2–206Crossref PubMed Scopus (7) Google Scholar). Skin barrier activity has been reported at 34 wk (Evans and Rutter, 1986Evans N.J. Rutter N. Development of the epidermis in the newborn.Biol Neonate. 1986; 49: 74-80Crossref PubMed Scopus (198) Google Scholar). A recent study of ultra-low birth weight infants, however, identified 30 wk post-conception age as a point where fetal barrier assumes functionality of adult barrier (Kalia et al., 1998Kalia Y.N. Nonato L.B. Lund C.H. Guy R.H. Development of skin barrier function in premature infants.J Invest Dermatol. 1998; 111: 320-326Crossref PubMed Scopus (197) Google Scholar). Recently, we developed whole-body assays for skin permeability and verified them as a measure of barrier by comparison with transepidermal water loss (TEWL) assay, a standard barrier assay (Nolte et al., 1993Nolte C.J.M. Oleson M.A. Bilbo P. Parenteau N.L. Development of the stratum corneum and barrier in an organotypic skin culture.Arch Dermatol Res. 1993; 285: 466-474Crossref PubMed Scopus (67) Google Scholar). When applied to developing rodent and rabbit fetuses the new assays showed that skin barrier forms in a patterned manner late in fetal development (Hardman et al., 1998Hardman M.J. Sisi P. Banbury D.N. Byrne C. Patterned acquisition of barrier function during development.Development. 1998; 125: 1541-1552Crossref PubMed Google Scholar). Barrier forms at epidermal initiation sites, then spreads around the body as moving fronts converging on ventral and dorsal midlines (Figure 1). Initiation sites and moving fronts were identified in mouse, rat, and rabbit. As a consequence of these studies we suggest that patterned barrier acquisition may be widespread among mammals. If so, we predict patterned barrier formation in the human infant. Our previous studies of developing rodent skin showed that barrier forms in stages (Hardman et al., 1998Hardman M.J. Sisi P. Banbury D.N. Byrne C. Patterned acquisition of barrier function during development.Development. 1998; 125: 1541-1552Crossref PubMed Google Scholar). The skin permeability assays used in these studies demonstrate a skin change that correlates with maturation of cornified envelopes and an initial fall in TEWL. The assays measure an early stage in barrier formation, whereas other forms of barrier assay (e.g. transepidermal water assay by evaporimeter;Nilsson, 1977Nilsson G.E. Measurement of water exchange through the skin.Med Biol Eng Comput. 1977; 15: 209-218Crossref PubMed Scopus (383) Google Scholar) measure a late stage of barrier formation. The skin permeability assays, therefore, provide a new tool for precise in situ identification of initiation of barrier formation in the human infant, a finding that could have significant clinical relevance. In this study we extend our analysis of whole-body barrier formation in mammals and use the new assays to (i) ask whether skin barrier formation is patterned in the human infant; (ii) find the precise stage in human skin development correlating with initiation of barrier formation; and (iii) investigate the pattern of periderm disaggregation and its temporal relationship to barrier formation. ICR mice were time-mated within a 2 h mating window and the mid-point of the window designated gestational age 0. Sprague–Dawley rats were time mated within a 12 h mating window and the mid-point of the window designated 0. Pouch young marsupial Monodelphis domestica were killed at defined times after birth (birth occurs at 13.5 d). Estimated gestational age (EGA) was calculated from the time designated 0. For example, 16 d 5 h after time 0 was called E16/5 or 16/5 d EGA. Human fetal ages are EGA unless otherwise specifically indicated. EGA = menstrual age minus 14 d. Human skin samples (minimum 1 cm in one dimension) were taken post mortem from spontaneously aborted infants of 17–30 wk EGA (19 infants) or supplied by Medical Research Council Tissue Bank, London (five infants, source of circumferential skin strips). Significantly growth retarded or macerated infants, or those who died more than 3 h after birth, were excluded from the study. EGA was calculated from multiple parameters including last menstrual period, crown–rump length, crown–heel length, foot length, head circumference, and organ weights. Research with human material was performed according to the Polkinghorne Code of Practice, U.K., after approval of protocols by the local Research Ethics Committee. Skin permeability assay was performed using the novel dye exclusion technique described in detail inHardman et al., 1998Hardman M.J. Sisi P. Banbury D.N. Byrne C. Patterned acquisition of barrier function during development.Development. 1998; 125: 1541-1552Crossref PubMed Google Scholar. Briefly, embryonic skin was incubated for 1–5 min in methanol, rinsed in phosphate-buffered saline, followed by incubation in 0.5% hematoxylin or 0.1% Toluidine Blue, as described (Hardman et al., 1998Hardman M.J. Sisi P. Banbury D.N. Byrne C. Patterned acquisition of barrier function during development.Development. 1998; 125: 1541-1552Crossref PubMed Google Scholar). After extensive destaining dye penetration reveals barrier status. White skin has barrier, stained skin lacks barrier. In some instances, where the skin invaginates due to dermatoglyphic patterning, artifactual white streaks appear (e.g., Figure 1). The new assay has been validated previously by comparison with standard methods for barrier assay, e.g., TEWL assay (Hardman et al., 1998Hardman M.J. Sisi P. Banbury D.N. Byrne C. Patterned acquisition of barrier function during development.Development. 1998; 125: 1541-1552Crossref PubMed Google Scholar). For histologic analysis samples were excised from skin prior to permeability staining. The permeability status of the samples was determined by staining the surrounding skin after sample removal, i.e., samples for histologic analysis were never subjected to permeability assay. Disaggregating periderm is visualized on post-barrier animals after staining as above. Periderm stains transiently with applied dyes and can be visualized against epidermis which is resistant to dye penetration due to barrier activity. Samples for light and electron microscopy were fixed in half strength Karnovsky's fixative and osmium tetroxide using standard techniques. Samples were visualized with a Philips 400 transmission electron microscope at 80 kV. We recently demonstrated barrier formation in rodents and rabbit (Hardman et al., 1998Hardman M.J. Sisi P. Banbury D.N. Byrne C. Patterned acquisition of barrier function during development.Development. 1998; 125: 1541-1552Crossref PubMed Google Scholar). Barrier forms at distinct epidermal sites, called initiation sites (asterisks, Figure 1a–c), then spreads around the fetus as moving fronts that converge ventrally and dorsally. Although the three species showed patterned permeability change (presence of initiation sites and moving fronts) the detail differed. Comparison of the similarly shaped rodent fetuses showed that mice and rats used identical initiation sites; however, they activate sites in a different temporal order. The larger rabbit fetus used additional initiation sites but shape and size differences complicate comparison with rodent. Hence, it was proposed that patterning was probably a generally mode of barrier acquisition in mammals. We sought to verify this proposal by examining barrier formation in an evolutionarily distant mammalian species. Here we demonstrate barrier formation in a marsupial possum, M. domestica (Figure 1d). Monodelphis domestica is born after 13.5 d gestation and, like all marsupials, completes late stages of development, including skin development, ex utero (Armstrong and Ferguson, 1995Armstrong J. Ferguson M. Ontogeny of the skin and the transition from scar-free to scarring phenotype during wound healing in the pouch young of a marsupial, Mondelphis domestica.Dev Biol. 1995; 169: 242-260Crossref PubMed Scopus (146) Google Scholar, and references within). We show patterned barrier formation on the day of birth (Figure 1d). Significantly, barrier forms via moving fronts as in other mammals. Demonstration of pattern in a marsupial lends considerable weight to our proposal that skin matures in mammals by a common mechanism. Therefore, we looked for patterned change in the human infant. Ethical considerations preclude whole-body analysis of barrier formation in the late gestation human infant. Therefore, skin samples were collected during post mortem examination from separate sites (abdominal midline, dorsal head, neck, chest, and lateral torso skin) on infants of 17–30 wk EGA and tested for barrier activity using the new dye permeability assays. Infants who had survived for more than a few hours post-birth were excluded from the study as previous researchers have demonstrated accelerated skin and barrier development upon air contact (Harpin and Rutter, 1983Harpin V.A. Rutter N. Barrier properties of the newborn infant's skin.J Pediatr. 1983; 102: 419-425Abstract Full Text PDF PubMed Scopus (229) Google Scholar;Evans and Rutter, 1986Evans N.J. Rutter N. Development of the epidermis in the newborn.Biol Neonate. 1986; 49: 74-80Crossref PubMed Scopus (198) Google Scholar;Hanley et al., 1997Hanley K. Jiang Y. Elias P.M. Feingold K.R. Williams M.L. Acceleration of barrier ontogenesis in vivo through air exposure.Pediatr Res. 1997; 441: 293-299Crossref Scopus (43) Google Scholar;Kalia et al., 1998Kalia Y.N. Nonato L.B. Lund C.H. Guy R.H. Development of skin barrier function in premature infants.J Invest Dermatol. 1998; 111: 320-326Crossref PubMed Scopus (197) Google Scholar;Williams and Feingold, 1998Williams M.L. Feingold K.R. Barrier function of neonatal skin.J Pediatr. 1998; 133: 467-468Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar). Barrier differed in a site-dependent manner. Detailed results are presented for abdominal midline and dorsal head (scalp) sites as they represent extremes of developmental rates from areas sampled. Results are summarized in Table 1.Table 1Barrier formation and keratinization are temporally linked during human fetal developmentGestational ageBarrierKeratinizationAbdomenHeadAbdomenHead17----19-±-±21-+-+22±+-+ +23±+±+ +24++++ +28+++ ++ + +30+++ ++ + + Open table in a new tab Epidermis from abdominal midline at 17 wk gestational age tested negatively by permeability assay (Figure 2a). By 18/19 wk, however, follicular epidermis had clear barrier activity that increased in prominence by 21 wk (white streaks, arrows; Figure 2b,c). Sectioning revealed that the white streaks correspond to keratinization confined to the hair canal (Figure 2n,o; arrow). This has been previously reported in section analysis of fetal skin at 22 wk EGA (Holbrook and Odland, 1980Holbrook K.A. Odland G.F. Regional development of the human epidermis in the first trimester embryo and the second trimester fetus (ages related to the time of amniocentesis and fetal biopsy).J Invest Dermatol. 1980; 74: 161-168Crossref PubMed Scopus (55) Google Scholar). Additionally, the follicular plug expelled to the epidermal surface (Hardy, 1992Hardy M. The secret life of the hair follicle.Trends Genet. 1992; 8: 54-61Abstract Full Text PDF Scopus (777) Google Scholar) provides a physical barrier (Figure 2p,q; arrow) that contributes to the white streak. Between 22 and 24 wk abdominal epidermis undergoes a transition from negative to positive barrier activity. At 22 wk barrier is provided by follicles but is also present in the vicinity of some of the follicles (Figure 2d, asterisks). By 23 wk barrier has moved out from some of the hair follicles appearing to join in places (Figure 2e), suggesting that follicular epidermis can act as initiation sites in human. This differs markedly from the primary mode of barrier initiation in other mammals (Figure 1). Several skin samples, however, show sharp changes in interfollicular barrier that resemble the animal moving fronts (e.g., Figure 2e, arrows, also see below). Therefore, it is not clear from the analysis of small skin samples whether barrier forms in abdominal skin by movement from follicles or as a result of a moving front superimposed over the follicular barrier. By 24 wk barrier formation appears complete (Figure 2f). Infants of 25 wk or greater always test positively for barrier at this site (Figure 2g). From 17 to 19 wk head epidermis is heterogeneous, with interfollicular skin yet to form barrier and follicular skin testing positively for barrier (Figure 2h–j) as in abdominal skin. Head skin, however, matures much more rapidly than abdominal skin (cf. 19 and 21 wk samples – Figure 2j,k with b, c). Detection of barrier this early in development was unexpected. Emanation of barrier from follicles, noted in abdominal skin at 22 wk, is much more prominent in head epidermis at 19 wk (Figure 2j, asterisks). This may be due to higher scalp follicle density. Epidermis from the dorsal side of the head at 21 wk or later tests uniformly positive for barrier (Figure 2k–m). Accumulating vernix caseosa (Agorastos et al., 1988Agorastos T. Hollweg A.T. Grussendorf E.I. Papaloucas A. Features of vernix caseosa cells.Am J Perinatol. 1988; 5: 253-259Crossref PubMed Scopus (24) Google Scholar and references within) on the epidermal surface appears as a purple remnant increasing in prominence in later samples (Figure 2k–m). Our finding that head epidermis forms barrier before abdominal epidermis correlates precisely with a previous report of accelerated epidermal keratinization in head (e.g.,Holbrook and Odland, 1980Holbrook K.A. Odland G.F. Regional development of the human epidermis in the first trimester embryo and the second trimester fetus (ages related to the time of amniocentesis and fetal biopsy).J Invest Dermatol. 1980; 74: 161-168Crossref PubMed Scopus (55) Google Scholar). Human barrier formation, however, has never been reported this early in development. Analysis of barrier formation using new permeability/dye exclusion assays demonstrates the following. (i) Interfollicular barrier first forms at approximately 20–21 wk on head epidermis and 23–24 wk on abdominal epidermis. This places initiation of barrier formation earlier than suspected. Our results, however, correlate well with previous reports describing timing of keratinization in skin (e.g.,Holbrook and Odland, 1980Holbrook K.A. Odland G.F. Regional development of the human epidermis in the first trimester embryo and the second trimester fetus (ages related to the time of amniocentesis and fetal biopsy).J Invest Dermatol. 1980; 74: 161-168Crossref PubMed Scopus (55) Google Scholar) and a report of change in fetal skin water permeability between 18 and 22 wk (Parmley and Seeds, 1970Parmley T.H. Seeds A.E. Fetal skin permeability to isotopic water (THO) in early pregnancy.Am J Obstet Gynecol. 1970; 108: 128-131PubMed Scopus (49) Google Scholar). (ii) Barrier formation is distinctly regional in the human infant. (iii) Human barrier could develop by two mechanisms. Barrier arises first from follicular epidermis but could complete either by convergence from follicular sites or front movement, as in other mammals. Although examination of small skin samples demonstrated that barrier first forms in follicular epidermis it was unclear whether full barrier is achieved by convergence from follicular sites or whether fronts cross the epidermis, as in other mammals. Several samples contained fronts (Figure 3a,b) that superimpose barrier over the islands of barrier radiating from follicles (Figure 3a, torso skin), or appear independent of follicles (Figure 3b, neck skin). At high magnification the follicle-independent front is identical to the animal fronts (cf. Figure 3b,c), indicating that barrier can form via fronts in human. The barrier fronts appear at different body locations at different gestational ages suggesting that they are moving across the epidermal sheets, as in other mammals. Although whole-body analysis is not permissible it was possible to obtain a limited number of circumferential torso samples (Figure 3d,e). Analysis of these samples demonstrates the regional nature of barrier formation in the human. Barrier forms first at ventral–lateral sites on torso (arrows, Figure 3e). The regional nature of barrier formation is confirmed by analysis of skin strips encompassing ear to ear dorsal scalp samples (Figure 3f). Patterned barrier formation is apparent across the scalp with barrier initiating at lateral locations (near the ears) and fronts (Figure 3g,h) appearing to converge dorsally. The circumferential torso samples were used to analyse the two proposed methods of barrier formation. Follicles across the human dorsoventral torso region are of approximately equivalent maturity, as demonstrated by presence of hairs of approximately equal length (Figure 3i). Follicular epidermis has barrier activity uniformly around the body (Figure 3d). Later a region similar to an animal initiation site appears on the ventral side of the infant (Figure 3e). Interfollicular epidermis acquires barrier activity only within this region. Therefore, we conclude that in humans barrier arises from follicles, and also via larger initiation sites and moving fronts as in other mammals. We were particularly intrigued by the apparent difference between barrier formation in human and other species. Barrier is propagated across the body primarily via moving fronts in rodents, rabbit, and the marsupial possum, Monodelphis. Humans are the only species so far examined where barrier initiates at follicles. Closer examination of the animal models, however, shows that animal follicles, at certain stages of development, can also act as initiation regions. Developing vibrissae follicles in rodents clearly act as barrier initiation sites (Figure 3j) whereas the body or pelage follicles do not function as initiation sites (Figure 3j). A primary difference between vibrissae and pelage follicles in rodents is the degree of development relative to interfollicular differentiation. Vibrissae follicles develop precociously, appearing 2–3 d earlier than pelage follicles (Van Exan and Hardy, 1980Van Exan R.J. Hardy M.H. A spatial relationship between innervation and the early differentiation of vibrissa follicles in the embryonic mouse.J Anat. 1980; 131: 643-656PubMed Google Scholar). In addition, vibrissae follicles demonstrate a hierarchy of maturation with the most mature follicles displaying most prominent barrier formation (arrows show decreasing maturity, Figure 3j). Other skin appendages, such as nails, can also initiate barrier formation prior to subsequent propagation by moving fronts (Figure 3k). Therefore, in species such as humans where follicle development is advanced relative to interfollicular differentiation, the follicles can act as initiators. It is possible that in regions such as the head/scalp where follicular density is high and/or follicular maturation is accelerated, the movement of barrier from follicles may be the primary mode of barrier formation. The close temporal correlation between barrier initiation and previous reports of fetal keratinization (Holbrook and Odland, 1980Holbrook K.A. Odland G.F. Regional development of the human epidermis in the first trimester embryo and the second trimester fetus (ages related to the time of amniocentesis and fetal biopsy).J Invest Dermatol. 1980; 74: 161-168Crossref PubMed Scopus (55) Google Scholar) were gratifying as we have previously demonstrated that initiation of barrier formation correlates with conversion of the outer, stratum corneum cells precursor cell layer to an electron-dense morphology, an early step in keratinization (Hardman et al. 1988). In addition, demonstration of barrier initiation as early as 20–24 wk EGA means that barrier is forming near the reported time of periderm release (Holbrook and Odland, 1975Holbrook K.A. Odland G.F. The fine structure of developing human epidermis: light, scanning and transmission electron microscopy of the periderm.J Invest Dermatol. 1975; 65: 16-38Crossref PubMed Scopus (139) Google Scholar). This suggests a very close temporal relationship between the processes. We examined human fetal samples at the ultrastructural level to define the relationship between barrier formation, keratinization and periderm release. Again, detailed results are presented for abdominal midline and dorsal head sites (Table 1). Changes in keratinization mirror the permeability changes between 18 and 30 wk. At the cellular level pre-barrier abdominal epidermis (19 wk EGA) is thin, lacks interfollicular cornification (Figure 4a) and has intact blebbed periderm with clear microvilli protruding from the outer surface (Figure 4b). Periderm decreases in thickness up to 23 wk (Figure 4c–e), and the microvilli noticeably reduce in number (Figure 4f). At 23 wk the periderm is regionally detaching (Figure 4e), and absent by 28 wk (Figure 4g). Abdominal interfollicular skin undergoes keratinization between weeks 22 and 23. A single layer of electron-dense cornified cells appears at 23 wk (Figure 4e; arrow). By 28 wk up to seven cornified cells are present (Figure 4g) and by 30 wk the epidermis has a prominent stratum corneum (Figure 4h). At 17/19 wk head skin lacks interfollicular keratinization and the periderm appears to be partially degraded (Figure 4i). Follicular epidermis is cornified. By 20 wk skin from the head site is entirely covered by dense, flattened cells with prominent cornified envelopes (Figure 4j). The number of keratinized cells at the epidermal surface in head skin continues to increase (Figure 4k,l), reaching approximately three–five layers by 23 wk and eight layers by 30 wk (Figure 4
Referência(s)