Histochemistry of the porcine pilosebaceous unit

Acta histochem. (lena) 93, 256-263 (1992) Gustav Fischer Verlag lena' Stuttgart· New York

A~la Idsl~

Departments of Dermatology!) and Plastic Surgery2), and Centre of Animal Research 3 ), Friedrich-Schiller-University, lena, Germany

Histochemistry of the porcine pilosebaceous unit By UWE WOLLINA'), UWE BERGER2), CHRISTINE STOLLE l ), HEIKO STOLLE l ), HARALD SCHUBERT3) and CHRISTINA HIPLER l ) With 2 Figures (Received February 22, 1991; revised lune 28, 1991; accepted November 28, 1991) Key words: Porcine skin, Pilosebaceous unit, Human skin, Lectin histochemistry, Immunohistochemistry

Abstract The present study describes lectin and immunoreactivilY in the pilosebaceous unit of porcine skin. Complex carbohydrates of mucin and biantennary Man/Gluc types were distributed among hair follicle epithelia (hair root sheaths, cuticula, shaft, and shaft matrix). Sebaceous glands expressed biantennary Man/Gluc carbohydrates and GalNAc residues. The expression of simple-type and epidermis-like keratins was confmned by immunohistochemistry with monoclonal antibodies. Filaggrin-positive cells were found in the keratinizing zone of HENLE'S layer in anagen follicles. The innermost layer of the outer hair root sheath was stained with antibodies against the epidermal growth factor-receptor, keratin 10 and Ki67 antigen. The differences to humans were remarkably small.

1. Introduction The domestic pig has become an accepted model of medical research. In particular, porcine skin resembles human skin in some important qualities making it reliable for experimental dermatology, surgery, toxicology, pharmacology, and radiology (ARCHAMBEAU and BENNETT 1984; MARACRIAN and CALHOUN 1966; MONTAGNA and YUN 1964; MEYER et al. 1986; TSUKISE and MEYER 1983; WOLLINA et al. 1991 a, b). In contrast to human skin, pigs show a trio clustering of hair follicles. Hair growth in pigs but not in humans discloses a remarkable annual cyclic activity (WATSON and MOORE 1990). In most domestic mammals, sebaceous glands are paired. In pigs they are larger than in densely haired species (MARCARIAN and CALHOUN 1966; MONTAGNA and YUN 1964; MEYER et al. 1978; MEYER and NEURAND 1987). Therefore, it seems questionable, that porcine skin provides a reliable model in biomedical hair research. Since histochemical data on the porcine hair follicle and sebaceous gland are extremely sparse, the present study was initiated in order to elucidate the differences of porcine and human skin.

2. Material and methods 2.1. Tissue samples were obtained from the hairy skin ofthe trunk of 6 domestic pigs of both sexes (age 10 to 15 month; Institute of Animal Contagion Research, Academy of Veterinary Sciences, lena) and of 15 minipigs [MINILEWEj (age 2 to 2.5 a; 12 female and 3 castrated males; Centre of Animal Research, University of lena) under anesthesia. Skin specimens were snap frozen in liquid nitrogen and cut at 5 ~m.

Histochemistry of the porcine pilosebaceous unit

257

Table I. Lectins used. Abbreviation

Source

Major sugar specificity*

Con A PSA UEA I WGA

Canava/ia ensiformis Pi"lIIJ1 sativum Ulex europaeus I Triticulll vulgare

ty-D-Man/ty-D-Gluc D-Man, D-Gluc ty-L-Fuc GlcNAc131 -> 4 sequences

* According to GALLAGHER (1984) and OSAWA and TSUJI (1987).

2.2. Leetins (Table I) were labelled with f1uorescein-isothiocyanate and applied on unfixed frozen sections in the direct fluorescence technique [(DIF); (WOLLINA et al. 1990a)] with a modification of JOHNSON and DE NOGUEIRA ARANJO (1981) to reduce fading. PSA was purchased from Sigma. PNA from SIFIN (BerlinWeiBensee/FRG), ConA from the Academy of Agricultural Sciences (Gatersleben/FRG), WGA from the Department of Biotechnology. University of Leipzig, and UEA from Boehringer. Mannheim/FRG. Table 2. Antibodies lCK. cytokeratin(s)]. Specificity

Characteristics

Source

Cam 5.2

CK 8, 18. 19

IgG2a (mouse)

Becton Dickinson

K 8.12

CK 13, 16

IgG I (mouse)

Bio-Yeda

RKSE 60

CK 10

IgG I (mouse)

Euro-Diagnostics

Anti-vimentin

vimentin

polyclonal (rabbit)

Euro-Diagnostics

Vim 9 (I)

vimentin type 9

IgG I (mouse)

Monosan

Anti-filaggrin

filaggrin, profilaggrin

IgG I (mouse)

Paesel & Lorei

ACAM

calmodulin

polyclonal (rabbit)

Dr. WENZ, JenalFRG

Anti-collagen

collagen type IV

IgG 1 (mouse)

Biogenesis

29.11

EGF-receptor

19G 1 (mouse)

Sigma

Antibody

Antikeralins

Allli-Vimentins

Other antibodies

2.3. Po/y- and monoclonal antibodies against tissue antigens are listed in Table 2. Secondary antibodies for peroxidase technique (POX) were purchased from Dako (Hamburg/FRG). For immunostaining we followed the protocol of STERNBERGER et al. (1970) using acetone-fixed tissue sections. POX was developed with 3-amino-9ethylcarbazole (EGA-Chemie. Steinheim). 2.4. Staining CO/1/ro/s included inhibition of lectins by monosaccharides (WOLLINA et al. 1990a) and substitution of the primary antibodies by buffer solution. Examination was done with either the lenalumar fluorescence microscope or the Histova/light microscope (Carl Zeiss lena, FRG).

3. Results The distribution of lectin binding sites and tissue antigens in porcine and human epithelia of the pilosebaceous unit is summarized in Table 3. Domestic pigs and minipigs provided identical histochemical results. 17

Acta histochem .. Bd. 93. t

258

U. WOLLINA et al.

Table 3. Comparison of histochemical reactivity in the pilosebaceous units of porcine (P) and human skin (H). Tissue

Markers

P

H

PSA, PNA, ConA ACAM (supraglandular) K 8.12 anti-filaggrin supraglandular zone keratinizing zone

+1++

++ + ++

+ ++ + + +

+ ++

Hair follicle outer hair root sheath

innermost layer of the outer root sheath inner hair root sheath

29.11 Ki67, RKSE 60 ConA ACAM Cam 5.2 (infraglandular) RKSE 60

cuticula

PSA, PNA

shaft and shaft matrix

RKSE 60 PNA PSAlUEA I ACAM Cam 5.2 PSA

dermal papilla

anti-vimentin, Vim9(l)

dermal hair follicle sheath

anti-collagen anti-vimentin, Vim9(l)

Mm. arrectores pili

WGA, PSA

+ +

++ ++ ++ +* +*

+ + ++ ++ ++ ++ +* +*

+I++§

+I++§

+* ++ ++ ++ + +

+* +* ++ ++ + + +

++ ++ + ++

++ ++ + +

Sebaceous gland peripheral sebocytes

ConA, WGA ACAM

central sebocytes

ConA, WGA

sebaceous duct cells

ACAM, K 8.12

* Single cells or mosaic-like pattern. §

The shaft was PSA-negative.

3. 1. Sebaceous glands Sebocytes: No distinct patterns were observed between porcine and human specimens. In general, the basal (immature) sebocytes showed a stronger reactivity with lectins as well as with antibodies. These cells expressed
Fig. 1. Immunostaining of anagen porcine hair follicles (POX, x 1250): a. RKSE 60 reaction with the innermost cell layer (arrow) of the outer root sheath. b. Filaggrin-positive cells (arrow) of the keratogenic zone. c. Cam 5.2-positive cells of outer root sheath, HENLE'S [a] and HUXLE'(S [b]layer. d. Collagen type IV is strongly expressed in the dermal papilla (arrow).

Histochemistry of the porcine pilosebaceous unit

25()

260

U. WOLLlNA et al.

Fig. 2. Lectin histochemistry of the supraglandular portion of porcine hair follicles (DIF, x 512): a. WGA; b. PSA. Note the intense fluorescence of the hair root sheath in (a) and the lack of hair shaft reactivity in (a) and (b).

Histochemistry of the porcine pilosebaceous unit

261

Sebaceous duct cells.' Most intense labelling was obtained with ACAM, whereas in human samples, there was only a moderate calmodulin-reactivity. The duct cells were additionally decorated by K8.12. Infundibular epithelial cells: The infundibular keratinocytes behaved as the supraglandular outer hair root sheath cells, with one exception: Both porcine and human cells were not immunoreactive with ACAM.

3.2. Hair follicle

Outer hair root sheath: Distinct patterns were observed in the supra- and infraglandular portions of this sheath. ACAM and anti-filaggrin labelling were restricted to the supraglandular portion, whereas RKSE 60 binding was seen in the innermost layer of the outer hair root sheath in the infraglandular portion (Fig. I a). Additional markers of the innermost layer of the outer root sheath were the antibodies against Ki67 and (fainty) epidermal growth factorreceptor (EGF-R). The antibody K8.l2 stained the infraglandular portion more intense than the supraglandular one. Labelling was evident in particular in the basal layer. PSA did not stain the supraglandular portion, but WGA gave a strong signal (Fig. 2). There was no significant difference between porcine and human tissue specimens. Inner hair root sheath: In pigs, ~-D-Man/~-D-Gluc residues were detectable (Fig. 2a). Human specimens showed an additional reactivity for calmodulin. In the keratogenic zone of HENLE'S layer filaggrin-positive cells were seen in anagen follicles (Fig. I b). HUXLEY'S layer gave a moderate but HENLE'S layer a strong reaction with Cam 5.2 (Fig. Ie). Cuticula: The cuticula was stained by PNA and PSA indicating the presence of mucin-type and complex biantennary carbohydrate residues containing Gal and Glue/Man, respectively. Additionally, these cells were reactive with RKSE60. Shaft and shaft matrix: RKSE 60 gave a positive signal on hair shafts. The matrix cells showed a more or less stronger staining with PSA and UEA I. Single intermingled cells were marked by PNA (in pigs and humans), RKSE 60 (only in pigs) or ACAM (only in humans). In some cases, cells of the matrix (suprabulbar region) were labelled with Cam 5.2. Dermal papilla and dermal hair follicle sheath: Both components of the connective tissue were reactive with antibodies against vimentin. The dermal sheath was positive for collagen type IV and PSA (Fig. 2 b). The intensity of staining with anti-collagen was stronger in porcine skin (Fig. I d) than in man. Mm. arrectors pilli: They were labelled with WGA and PSA without species specific differences.

4. Discussion The present study deals with histochemical findings on the porcine pilosebaceous unit and a brief reference to man. Sebaceous glands of pig and human skin showed comparable staining patterns with lectins (ConA and WGA) demonstrating the expression of complex carbohydrates with a:-D-Man, a:-D-Gluc and GalNAc (cf. WOLLINA et al. 1990a). Our data are in agreement with recent reports on different mammals (TSUKISE and MEYER 1983; MEYER and TSUKISE 1989a), suggesting only minor differences in glycoconjugate expression between pig, monkey, and human. Mucin-type Gal containing carbohydrates were detected in epithelia of the hair folllide: outer hair root sheath, cuticula, and shaft matrix. Biantennary sugar chains containing Man/Glue were present in the outer and inner hair root sheath and the cuticula. However, there have been minor differences in expression of these complex carbohydrates as demonstrated by variability in ConA and PSA reactivity (cf. OSAWA and TSUJI 1987; GALLAGHER 1984). TSUKISE and MEYER (1983) reported comparable staining patterns in fixed porcine hairy skin. Otherwise, they found a more pronounced UEA I expression, which was not evident in the present study. The reason for this are the different methods for tissue processing and staining (BELL and SKERROW 1984). The Mm. arrectores pili expressed GalNAc and biantennary Man/Glue containing sugar chains (cf. WOLLINA et al. 1990a; TSUKISE and MEYER 1983).

262

U. WOLLINA et al.

Immunhistochemical investigations in the pig are difficult because tissue specific monoclonal antibodies are not yet commercially available (GROVES and TUCKER 1989). In a recent study we successfully applied antihuman poly- and monoclonal antibodies to investigate epidermal differentiation in porcine skin (WOLLINA et al. 1991 a, b; see also: MEYER et al. 1986). In the present paper our previous results were substantiated and extended to the specialized pilosebaceous epithelia. Immunoreactive calmodulin, thought to playa crucial role in skin permeability control (WOLLINA et al. 1991), was found in the other hair root sheath of pigs. In humans, calmodulin was also present in the inner hair root sheath and scattered matrix cells (WOLLINA et al. 1991 b). In sebaceous glands, both the basal (immature) sebocytes and the duct epithelium were stained by ACAM. Monoclonal antibodies against keratins used herein did not stain sebaceous glands of porcine skin. A possible explanation may be the low level of expression (GROVES and TUCKER 1989). In the hair follicle, on the other hand, a similar distribution of keratins was observed in porcine and human tissue samples. The bulbar region showed scattered cells positive for simple-type keratins recognized by Cam 5.2. Additionally, cells of the inner hair root sheath strongly expressed simple-type keratins. Recently, HElD et aI. (1988) demonstrated cytokeratins 8, 18, 19 in peribulbar epithelial cells of human and bovine hair follicles. Keratin 10 was seen in the innermost layer of the outer hair root sheath, cuticula, and scattered matrix cells of porcine skin (Fig. I c) like in human (and bovine) hair follicles (HElD et al. 1988). The innermost layer of the outer hair root sheath was distinctly labelled by anti-EGF-R and anti-Ki67, suggesting a special function of this cell layer (MIYAUCHI et al. 1990). Filaggrin was expressed in the upper layers of the outer hair root sheath of porcine and human skin (cf. IMCKE et al. 1988; KUROKAWA et al. 1988). The dermal hair follicle sheath was stained by PSA and antibodies against vimentin and collagen type IV. The immunoreactivity was more intense in the pig, which may be related to the grouping of porcine hairs (MARCARIAN and CALHOUN 1966; ARCHAMBEAU and BENNETT 1984). Surprisingly, there have been only minor differences in staining patterns between porcine and human pilosebaceous epithelia. Thus, histochemical phenotyping supports a closer relationship of pilosebaceous differentiation in mammals than estimated previously.

Acknowledgements The study was supported by a grant from the Paul-Gerson-Unna-Stiftung (Gottingen). The skillfull technical assistance of Miss SABINE FELDRAPPE is highly appreciated.

References ARCHAMBEAU, J. 0., and BENNETT, G. W.: Quantification of morphologic, cytologic, and kinetic parameters of unirradiated swine skin: A histologic model. Radiat. Res. 98, 254-273 (1984). BELL, C. M., and SKERROW, C. J.: Factors affecting the binding of lectins to normal human skin. Brit. J. Dermatol. 111,517-526 (1984). GALLAGHER, J. T.: Carbohydrate-binding properties of lectins: A possible approach to lectin nomenclature and classification. Biosci. Rep. 4, 621-632 (1984). GROVES, D. J., and TUCKER, E. M.: The production and application of non-rodent monoclonal antibodies in veterinary science. Veter. Immunol. Immunopathol. 23, 1-14 (1989). HElD, H. W., MOLL, I., and FRANKE, W. W.: Patterns of expression oftrichocytic and epithelial cytokeratins in mammalian tissues. I. Human and bovine hair follicles. Differentiation 37,137-157 (1988). IMCKE, E., GOLLNICK, H., und ORFANOS, C. E.: Vorkommen und Verteilung von Zytokeratinen und Filaggrin in humanen Anagenfollikeln. Hautarzt 39, 680-683 (1988). JOHNSON, G. G., and DE NOGUEIRA ARANJO, G. M.: A simple method of reducing the fading of immunofluorescence during microscopy. J. Immunol. Meth. 4, 349-350 (1981). KUROKAWA, I., MAYER-DA-SILVA, A., GOLLNICK, H., and ORFANOS, C. E.: Monoclonal antibody labelling for cytokeratins and filaggrin in the human pilosebaceous unit of normal, seborrheic and acne skin. J. Invest. Dermatol. 91, 566-571 (1988). MARCARIAN, H. Q., and CALHOUN, M. L.: Microscopic anatomy of the integument of adult swine. Amer. J. Veter. Res. 27, 765-772 (1966).

Histochemistry of the porcine pilosebaceous unit

263

MEYER, W., GORGEN, S., and SCHLESINGER, C.: Structural and histochemical aspects of epidermis development of fetal porcine skin. Amer. J. Anal. 176,207-219 (1986). and NEURAND, K.: A comparative scanning electron microscopic view of the integument of domestic mammals. Scanning Microscopy I, 169-180 (1987). SCHWARZ, R., and NEURAND, K.: The skin of domestic mammals as a model for the human skin, with special reference to the domestic pig. CUIT. Probl. Dermatol. 7, 39-52 (1978). and TSUKISE, A.: Histochemistry of complex carbohydrates in the scrotal skin of the monkey Macaca cyclopis (Swinehoe). Z. Saugetierkde. 54, 9-21 (l989a). - Histochemistry of glycoconjugates of the bovine muzzle, with special reference to glandular structures. Acta Anal. 136,226-234 (l989b). MIYAUCHI, S., HASHIMOTO, K., and MIKI, Y.: The innermost cell layer of the outer root sheath is positive with Ki-67. J. Invest. Dermatol. 95, 393-396 (1990). MONTAGNA, W., and YUN, J. S.: The skin of the pig (Sus scrofa). J. Invesl. Dermatol. 43, 11-21 (1964). aSAWA, T., and TSUJI, T.: Fractionation and structural assessment of oligosaccharides and glycopeptides by use of immobilized lectins. Ann. Rev. Biochem. 56, 21-42 (1987). STERNBERGER, L. A., HARDY jr., P. H., CUCULlS, J. J., and MEYER, H. G.: The unlabelled antibody enzyme method of immunohistochemistry. J. Histochem. Cytochem. 18,315-333 (1970). TSUKISE, A., and MEYER, W.: Histochemistry of complex carbohydrates in the hairy skin of the domestic pig. Histochem. J. 15,845-860 (1983). WATSON, S. A. 1., and MOORE, G. P. M.: Postnatal development of the hair cycle in the domestic pig. J. Anal. 170, 1-9 (1990). WOLLlNA, U., BERGER, U., and MAHRLE, G.: Immunohistochemistry of porcine skin. Acta histochem. 90, 87-91 (199Ia). SCHAARSCHMIDT, H.-H., HIPLER, C., and KNOPF, B.: Distribution of glycoconjugates in human skin appendages. Acta histochem. 87, 87-93 (l990a). - KNOPF, B., und FELDRAPPE, S.: lmmunhistologische Untersuchungen zur Verteilung proliferativer Kompartimente in den Anhangsgebilden humaner adulter Haul. Z. mikrosk.-anal. Forsch. 104, 485-496 (I 990b). WEYERS, A., and MAHRLE, G.: Localization of calmodulin in epidermis and skin glands. A comparative immunohistological investigation in different vertebrate species. Acta histochem. 90, 135-140 (1991 b). Author's address: Dr. sc. med. UWE WOLLlNA, Universitats-Hautklinik Jena, Abtlg. Experimentelle Dermatologie, Erfurter Str. 35, 0-0-6900 Jena, Germany.