Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Original Contribution Journal of Cosmetic Dermatology, 12, 179--186 Immunohistochemical sweat gland profiles €l,1 Ge rald E Pie rard, MD, PhD,1,2 Philippe Delvenne, MD, PhD,1 Pascale Quatresooz, MD, PhD,1,3 Fanchon Noe rard-Franchimont, MD, PhD1 Philippe Humbert, MD, PhD,2 & Claudine Pie Department of Dermatopathology, Liege University Hospital, Liege, Belgium Department of Dermatology, Saint-Jacques Hospital, University of Franche-Comte, Besancßon, France 3 Department of Histology, Liege University, Liege, Belgium 1 2 Summary Background Human sweat glands are heterogeneous in their structures and functions. Accordingly, eccrine, apocrine, and apoeccrine glands are distinguished. Aims Some immunohistochemical markers are expected to distinguish the sweat gland types in their secretory and excretory parts. Methods This study used two sets of antibodies. The first panel was composed of antibodies directed to well-defined sweat gland structures. The molecular targets included the low-molecular-weight cytokeratins CAM 5.2, the S100-B protein, the epithelial membrane antigen (EMA), the carcinoembryonic antigen (CEA), and the lectin Ulex europaeus agglutinin-1 (UEA-1). A second exploratory panel of antibodies targeted syndecan-1 (CD138), NKI-C3 (CD63), and CD68. They were used to disclose some undescribed antigen expressions in human sweat glands. Results The first set of antibodies confirmed previous findings. The immunoreactivities of the three sweat gland types were similar in the excretory ducts. By contrast, they were distinguished in the deeper coiled secretory portions of the glands. Conclusion Clues supporting their distinction and probably their functional activity were obtained by immunohistochemistry using the S100-B protein, CEA and CD63 antibodies. The immunoreactivity to the S100-B protein, CEA and CD63 possibly help identifying apoeccrine sweat glands or a peculiar functional activity of eccrine sweat glands. Keywords: sweat gland, Ulex europaeus agglutinin-1, CD63, CD68, CD138 Background In humans, two main types of sweat glands are distinguished according to their different secretory patterns of production. Histomorphology is typically used for such distinction.1 The eccrine sweat glands (ESG) also called atrichial sweat glands,2 secrete high amounts of an aqueous liquid following a merocrine mechanism. Thermoregulation is their main function during exposure to warm environment and during body hyperthermia related to fever and physical exercise. Other ESG stimuli Correspondence: G E Pierard, Department of Dermatopathology, Liege University Hospital, CHU Sart Tilman, BE-4000 Liege, Belgium. E-mail: [email protected] Accepted for publication October 27, 2012 © 2013 Wiley Periodicals, Inc. include psychologic stress and gustatory reflexes. The ESG level of activity is variable in time and different from one gland to another.3 They open directly onto the skin surface through individual acrosyringia.4 The apocrine sweat glands (ASG) also called epitrichial sweat glands2 secrete low amounts of a lipid-rich liquid, and they join up into the hair canal instead of the skin surface.5 The apocrine secretion is apparently derived from pinching off of the apical cytoplasm of the secretory cells. A positivity for the gross cystic disease fluid protein-15 appears typical for the ASG.6 A third group of sweat glands of the apoeccrine type was reported predominating in axillary skin. This finding remains a controversial topic.7–10 They were described sharing characteristic features of both ESG and ASG, and they possibly open onto the skin surface. The distinction between these apoeccrine glands 179 Sweat gland immunohistochemistry . F No€el et al. and both the ESG and ASG was mainly assessed on gross morphology.7 Skin of the forearms and the back exclusively contains ESG. By contrast, it was assumed that axillary skin contains approximately equal numbers of ESG, ASG, and apoeccrine glands.11 Although the morphological microanatomy remains the basis of sweat gland classification, molecular phenotyping, and functional characteristics should be combined when possible to the criteria.11,12 They could help the interpretation of the effects of dermocosmetic formulations on the sweat glands. In some conditions in humans, it is uncertain whether some glands are ESG or ASG depending on the selected criteria. Various immunohistochemical markers including cytokeratin (CK) phenotyping, as well as the epithelial membrane antigen (EMA), the carcinoembryonic antigen (CEA), and others were introduced for this purpose.6,13–16 In addition, the identification of glycoconjugates was used to charaterize sweat glands in health and disease.17–22 These findings are commonly extrapolated to identify the differentiation of sweat gland neoplasms.23,24 The present study was focused on the distinction between sweat gland types using conventional and undescribed immunohistochemical reactivities. We used a panel of antibodies directed to the low-molecularweight CAM 5.2 CK, the S100-B protein, EMA, CEA, the lectin Ulex europaeus agglutinin-1 (UEA-1), syndecan-1 (CD138), NKI-C3 (CD63), and CD68. The monoclonal CAM 5.2 CK antibody strongly reacts with CK 8 and at a lower extent with CK 7. No reactivity has been disclosed with CK 18. The polyclonal S100-B protein antibody detects one of the 19 Ca (2+)-binding proteins of the S100 family. The monoclonal anti-human EMA antibody22,24,25 and the polyclonal anti-human CEA antibody24,26 typically label sweat glands. UEA-1 is a lectin specifically binding to a-L fucosyl moieties. The antiUEA-1 antibody reveals the lectin-binding sites corresponding to oligosaccharides with terminal a-fucose, some of which are present in sweat glands.20,24,27 The monoclonal anti-human CD138 detects the transmembrane syndecan-1 proteoglycan.28–31 The monoclonal antibody to CD63 detects various proteins whose molecular weights range 25–100 kD. They are probably part of some lyzosomal antigens located in cytoplasmic vacuoles.32 No immunoreactivity of sweat glands has been reported so far to CD63. CD68 is another antigen present on lyzosomes, particularly in phagocytic cells. Material and methods This study was approved by the University Hospital Ethics Committee. It was conducted with the understanding 180 and consent of the volunteers. A total of 60 Caucasian adults aged 20–48 years were enrolled. At the time of biopsy, they were relaxed without any perceptible sweating, and they rested in an indoor environment controlled at 21 °C and 48–53% relative humidity. Deep punch biopsies, 6 mm in diameter, were performed on the mid inner forearm, the axillary vault, and the mid-lateral part of the back. Microscopic sections (6 lm thick) were cut from the formalin-fixed paraffin-embedded punch biopsies. Sections were dewaxed in xylene and rehydrated through grades of alcohol to PBS. They were subsequently processed for immunohistochemistry using a panel of antibodies (Table 1) and the avidin-biotin peroxidase method. After a 1-h incubation time with any of the primary antibodies, slides were washed in Tris-buffered saline (TBS) and incubated for 30 min with the secondary antibody (biotinylated swine anti-rabbit, 1:300, Dakopatts). Slides were rinsed in TBS and covered by the EnVision (Dakopatts, Glostrup, Denmark) polymerbased revelation system. After TBS washing, Fast Red (Dakopatts) was used as chromogen substrate. The last steps consisted of counterstaining with Mayer’s hemalum before mounting. Negative immunohistochemical controls were performed by omitting or substituting the primary and the secondary antibodies of the laboratory procedure. Results Data are summarized in Table 2. Low-molecular-weight CAM 5.2 cytokeratins The secretory coil and at a lesser extent the ductal cells of the ESG were identified by the CAM 5.2 antibody (Fig. 1a). All segments of the ASG including secretory and ductal cells were decorated by this antibody. Table 1 Panel of antibodies Antigen/antibody Dilution Source Carcinoembryonic antigen CD68 Cytokeratins CAM 5.2 1:200 1:200 1:250 Epithelial membrane antigen CD63 1:100 1:200 S100 protein CD138 Ulex europaeus agglutinin-1 1:2000 1:100 1:2000 Dako, Glostrup, Denmark Dako, Glostrup, Denmark Becton Dickinson, San Jose, CA, USA Dako, Glostrup, Denmark MP Biomedicals, Solon, OH, USA Dako, Glostrup, Denmark Dako, Glostrup, Denmark Vector laboratories, Burlingame, CA, USA © 2013 Wiley Periodicals, Inc. Sweat gland immunohistochemistry . F No€el et al. Table 2 Immunolocalization of sweat gland components Location CEA Acrosyringium Luminal + Dermal duct Luminal + Cytoplasm + Eccrine secretory coil Clear cell + Dark cell + Apocrine secretory coil Luminal 0 Cytoplasm 0 CD68 CAM 5.2 0 0 0 0 0 EMA + 0 0 0 0 + + 0 0 0 + + CD63 S100-B 0 0 CD138 UEA-1 + 0 0 0 + 0 0 0 + 0 0 0 + +/ 0 0 + + +: homogeneously positive, +/ : variable, : focally positive, 0: negative. S100-B protein The S100-B protein was found in the cytoplasm of secretory cells in the ESG (Fig. 1b). A weaker immunostaining was present in a thin luminal lining of the excretory duct. The ASG showed no immunoreactivity to the S100-B protein. In the secretory segments of ESG, clear cells exhibited strong cytoplasmic UEA-1 staining (Fig. 2a). The luminal part of sweat ducts showed apical membranous staining (Fig. 2b). The inner part of the acrosyringia was strongly labeled (Fig. 2c). In addition, sweat was strongly positive for UEA-1 (Fig. 2d). ASG globally appeared intensely labeled. The acinar cells of ASG revealed focal apical membrane UEA-1 immunostaining. Epithelial membrane antigen Epithelial membrane antigen (EMA) was present in several segments of the ESG and ASG. EMA was located in the luminal coating of the eccrine secretory segment and in the dark cells as well (Fig. 1c). It was present in the coiled intraepidermal portion of the acrosyringium. Eccrine sweat was intensely positive for the EMA (Fig. 1d). The ASG contained EMA in different segments including the glomerular structure and the excretory duct. Carcinoembryonic antigen Carcinoembryonic antigen (CEA) was present at all levels of the ESG, particularly lining the lumen of secretory and ductal portions (Fig. 1e). The ductal labeling appeared stronger than in the secretory epithelium. The luminal part of the acrosyringium was intensely positive (Fig. 1f). The ASG showed a heterogeneous CEA labeling on the luminal coating of the duct (Fig. 1g). Ulex europaeus agglutinin-1 In the secretory coils of the various sweat glands, the different cell types appeared to contain distinctive a-fucose glycoconjugates in both their plasma membranes and cytoplasm. © 2013 Wiley Periodicals, Inc. CD138 CD138 appeared abundant in the outer portion of the acrosyringia in an intercellular pattern similar to the epidermis (Fig. 3a). This antigen was present at a weaker extent or absent in the luminar portion. A variable number of cells of the secretory segment were strongly positive (Fig. 3b, c) as well as sweat (Fig. 3d). In the ASG, a thin CD138+ luminal lining was disclosed in the excretory duct, while no immunoreactivity was present in the secretory coil (Fig. 3e). CD63 In ESG, CD63 immunostaining was present in the apical part of secretory cells (Fig. 4a). The luminal lining was decorated in an irregular pattern in the acini and ducts. A thin luminal lining and a faint cytoplasmic positivity were disclosed in ASG (Fig. 4b,c,d). CD68 No CD68 immunoreactivity was disclosed in the sweat glands. Discussion The current concept about axillary sweat glands distinguishes ESG, producing abundant clear, nonodorous 181 Sweat gland immunohistochemistry . F No€el et al. (a) (b) (c) (d) (f) (e) (g) Figure 1 Protein immunoreactivity in sweat glands. (a) CAM 5.2 cytokeratins, (b) S100-B protein, (c, d) EMA, (e, f, g) CEA. sweat and ASG excreting small amounts of turbid, odorous milky sweat. A third type of sweat glands corresponding to the apoeccrine type was described in the literature.7 So far, they are not clearly and specifically distinguished using immunohistochemistry, and publications about this topic are scanty and controversial.8,9 Each ESG consists of a coiled portion in continuation with a straight intradermal duct ending as a spiral intraepidermal acrosyringium. The secretory portion represents about two-thirds of the coiled structure. It 182 consists of a single layer with clear and dark cells. The majority of the cells are large and clear. They contain glycogen and they produce the eccrine sweat. Changes in their tiny granules occur after episodes of intense sweating. Fine canaliculi collecting the secreted sweat are present between adjacent clear cells. The smaller granular dark cells are less numerous. They tend to be pyramidal in shape with a relatively narrow contact with the peripheral basement membrane. The intraepidermal acrosyringium is lined by cells resembling those to the straight part of the intradermal duct. © 2013 Wiley Periodicals, Inc. Sweat gland immunohistochemistry (a) (b) (c) (d) . F No€el et al. Figure 2 Glycoconjugate immunoreactivity in sweat glands. (a, b, c, d) UEA-1. (a) (b) (c) (d) (e) Figure 3 Proteoglycan immunoreactivity in sweat glands. (a, b, c, d, e) CD138. Each ASG consists of a coiled secretory portion connected to a duct. The glandular portion is formed by a single layer of cuboidal or columnar cells. The secretory cells contain granules and vacuoles, some of which may be pigmented. The apocrine duct closely resembles © 2013 Wiley Periodicals, Inc. the eccrine duct. It consists of a double or triple layer of rather similar cuboidal cells, those of the inner layer exhibiting a faint luminal fringe. The peripheral basal layer contains numerous mitochondria and microvilli. The apocrine sweat is scanty and sticky in consistency. 183 Sweat gland immunohistochemistry . F No€el et al. (a) (b) (c) (d) Figure 4 Lysosomal antigen immunoreactivity in sweat glands. (a, b, c, d) CD63. It is considered that the different portions of the ESG contain a cytoskeleton made of several distinct CK.33,34 They include CK 7, CK 8, CK 14, CK 18, and CK 19, as well as EMA and CEA.22 It was reported that the expression intensity of CK 7, CK 18, and CD 19 was stronger than that of CK 8 and CK 14.22 A controversy exists about the presence of CK 10.22,34 The present findings using the CAM 5.2 antibody are in line with the current concepts. A positivity was found in both the ESG and ASG, and in the presumed apoeccrine glands. By contrast, the S100-B immunoreactivity appeared restricted to the cytoplasm of eccrine secretory coils.35 Any S100-B immunoreactivity in larger sweat glands in axillary skin could be tentatively used to reveal apoeccrine glands. Epithelial membrane antigen (EMA) is present in the luminal coating of ESG. EMA immunoreactivity was found in ASG as well.24 The anti-EMA antibody is thus unlikely helpful in the identification of apoeccrine sweat glands. Carcinoembryonic antigen (CEA) is a glycoprotein found in diverse organs. In the skin, it is exclusively present in sweat glands. The CEA gene family belongs to the immunoglobulin superfamily. CEA has been presented as an immunologic marker of ESG.22,24 It possibly plays a role in the innate immune defense.26 CEA possibly binds and traps micro-organisms at the cell surface. Epidermal growth factor receptors are expressed as well.22,36 The secretory portion of ASG do 184 not exhibit CEA immunoreactivity. Any positivity in large acini could therefore represent a clue for apoeccrine differentiation. Lectins represent a group of proteins and glycoproteins distinct from enzymes and antibodies. UEA-1 binds to a-fucose moiety with a high specificity through complementary sugar-binding sites. Lectins were used to demonstrate the presence of complex carbohydrate moieties at the cell surface or in cytoplasmic organelles in skin appendages. However, little is known about the precise composition and distribution of glycoconjugates in the normal skin and skin appendages. Only few reports deal with the application of lectins to the determination of sweat gland differentiation. UEA-1 was shown to be present on the plasma membranes of the eccrine dark and clear cells, as well as in apocrine cells.27 Similarly, granules of the ESG and lysosomal granules in apoecrine cells were reported to contain binding sites for the UEA-1 lectin.27 Our findings are in line with previous works and suggest that UEA-1 immunoreactivity does not help identifying apoeccrine glands. CD138 antigen corresponding to syndecan is a cell surface proteoglycan, playing a prominent role in tissue remodeling and homeostasis. It binds to growth factors and interstitial matrix molecules. It modulates the effect of the primary ligand–receptor interaction at the cell membrane by increasing the affinity of cell– ligand interactions. Additionally, it influences the © 2013 Wiley Periodicals, Inc. Sweat gland immunohistochemistry strength of cell–cell and cell–matrix interactions. Under physiologic conditions, CD138 expression is restricted to the epidermis, the outer root sheath of the anagen hair follicle, and the sweat gland epithelium. The present study confirms the presence of CD138 in sweat glands. In addition, we stress the fact that the luminal part of the acrosyringium is not immunoreactive in normal conditions. Cells showing diffuse CD63 positivity indicate an unusual structure of lysosomes. CD63 was originally offered as a melanoma antigen, but it was subsequently shown to be directed against a probable lyzosomal antigen. The present study highlights the CD63 immunoreactivity in sweat glands. We did not find any specific report about that sweat gland immunoreactivity in the literature. The apical pole of the eccrine secretory coils was labeled. In apocrine secretory cells, the cytoplasmic immunoreactivity was faint to absent, but a thin luminal lining was clearly evidenced. Such different patterns of immunoreactivity possibly help identifying apoeccrine sweat glands. Of note contrasting with the CD63 immunoreactivity, CD68, another lysosomial marker, was not disclosed in both ESG and ASG. In conclusion, a multipronged immunohistochemical approach is helpful in the study of human sweat glands. Although the excretory ducts appear phenotypically similar in the distinct sweat glands, the deep secretory coils show distinctive differentiation patterns. The present findings suggest that the immunoreactivity to the S100-B protein, CEA, and CD63 may help identifying some apoeccrine sweat glands. Acknowledgments This work was supported by a grant from the “Fonds d’Investissement de la Recherche Scientifique” of the University Hospital of Liege. No other sources of funding were used to assist in the preparation of this manuscript. The authors have no conflict of interest that are directly relevant to the content of this review. The authors wish to thank Mrs. Jennifer EspinosaPerez, Mr. Pierrick Malengreaux, and Mr. Fabian Mattiuz for their skillful technical assistance. We appreciate the excellent secretarial assistance of Mrs. Ida Leclercq and Marie Pugliese. References 1 Saga K. Histochemical and immunohistochemical markers for human eccrine and apocrine sweat glands: an aid for histopathologic differentiation of sweat gland tumors. J Invest Dermatol Symp Proc 2001; 6: 49–53. © 2013 Wiley Periodicals, Inc. . F No€el et al. 2 Rees-Jones AM, Jenkinson DM. The effect of aluminium chlorhydrate on sweat gland activity in cattle. J Invest Dermatol 1978; 70: 134–7. 3 Xhauflaire-Uhoda E, Mayeux G, Quatresooz P et al. Facing up to the imperceptible perspiration. Modulation influences by diabetic neuropathy, physical exercise and antiperspirant. Skin Res Technol 2011; 17: 487–93. 4 Mayeux G, Xhauflaire-Uhoda E, Pierard GE. Patterns of aluminium hydroxychloride deposition onto the skin. Skin Res Technol 2012; 18: 64–9. 5 Stoeckelhuber M, Schubert C, Kesting MR et al. Human axillary apocrine glands: proteins involved in the apocrine secretory mechanism. Histol Histopathol 2011; 26: 177–84. 6 Mazoujian G, Pinkus GS, Davis S et al. Immunohistochemistry of a gross cystic disease fluid protein (BCDFP15) of the breast. A marker of apocrine epithelium and breast carcinomas with apocrine features. Am J Pathol 1983; 110: 105–12. 7 Sato K, Leidal R, Sato F. Morphology and development of an apoeccrine sweat gland in human axillae. Am J Physiol 1987; 252: R166–80. 8 Beer GM, Baum€ uller S, Zech N et al. Immunohistochemical differentiation and localization analysis of sweat glands in the adult human axilla. Plast Reconstr Surg 2006; 117: 2043–9. 9 Bovell DL, Corbett AD, Holmes S et al. The absence of apoeccrine glands in the human axilla has disease pathogenetic implications, including axillary hyperhidrosis. Br J Dermatol 2007; 156: 1278–86. 10 Bovell DL, MacDonald A, Meyer BA et al. The secretory clear cell of the eccrine sweat gland as the probable source of excess sweat production in hyperhidrosis. Exp Dermatol 2011; 20: 107–20. 11 Pierard GE, Elsner P, Marks R et al. EEMCO guidance for the efficacy assessment of antiperspirants and deodorants. Skin Pharmacol Appl Skin Physiol 2003; 16: 324–42. 12 No€el F, Pierard-Franchimont C, Pierard GE et al. Sweaty skin, background and assessments. Int J Dermatol 2012; 51: 647–55. 13 Cotton DWK. Immunohistochemical staining of normal sweat glands. Br J Dermatol 1986; 114: 441–5. 14 Wollina U. Human eccrine sweat gland. Expression of neuroglandular antigens and coexpression of intermediate filaments. Histol Histopathol 1991; 6: 191–8. 15 Wilke K, Keil FJ, Wittern KP et al. Immunolabelling is essential for the differentiation of human axillary apoeccrine glands. J Invest Dermatol 2004; 123: A93. 16 Wilke K, Wepf R, Keil FJ et al. Are sweat glands an alternate penetration pathway? Understanding the morphological complexity of the axillary sweat gland apparatus. Skin Pharmacol Physiol. 2006; 19: 38–49. 17 Hazen-Martin DJ, Sens DA, Spicer SS. Glycoconjugates in sweat glands and other structures of skin from normal and cystic fibrosis subjects. Am J Dermatopathol 1986; 8: 478–91. 185 Sweat gland immunohistochemistry . F No€el et al. 18 Wollina U, Schaarschmidt HH, Hipler C et al. Distribution of glycoconjugates in human skin appendages. Acta Histochem 1989; 87: 87–93. 19 Tsubura A, Fujita Y, Sasaki M et al. Lectin-binding profiles for normal skin appendages and their tumors. J Cutan Pathol 1992; 19: 483–9. 20 Illana M, Prada A, Verastegui C et al. Study of the distribution of glycosidic residues in eccrine sweat glands, with special reference to the content of sialic acid. Eur J Histochem 1997; 41: 41–6. 21 Sames K, Moll I, van Damme EJ et al. Lectin binding pattern and proteoglycan distribution in human eccrine sweat glands. Histochem J 1999; 31: 739–46. 22 Li HH, Zhou G, Fu XB et al. Antigen expression of human eccrine sweat glands. J Cutan Pathol 2009; 36: 318–24. 23 Saga K. Structure and function of human sweat glands studied with histochemistry and cytochemistry. Prog Histochem Cytochem 2002; 37: 323–86. 24 Metze D, Luger TA. Ultrastructural localization of carcinoembryonic antigen (CEA) glycoproteins and epithelial membrane antigen (EMA) in normal and neoplastic sweat glands. J Cutan Pathol 1996; 23: 518–29. 25 Murakami M, Ohtake T, Horibe Y et al. Acrosyringium is the main site of the vesicle/pustule formation in palmoplantar pustulosis. J Invest Dermatol 2010; 130: 2010–6. 26 Hammarstrom S. The carcinoembryonic antigen (CEA) family: structures, suggested functions and expression in normal and malignant tissues. Semin Cancer Biol 1999; 9: 67–81. 27 Schaumburg-Lever G, Metzler G, Tronnier M. Ultrastructural localization of lectin-binding sites in normal human eccrine and apocrine glands. J Dermatol Sci 1991; 2: 55–61. 186 28 Chandler WM, Bosenberg MW. Autoimmune acrosyringitis with ductal cysts: reclassification of case of eruptive syringoma. J Cutan Pathol 2009; 36: 1312–5. 29 Jung SY, Kim JM, Kang HK et al. A biologically active sequence of the laminin alpha 2 large globular 1 domain promotes cell adhesion through syndecan-1 by inducing phosphorylation and membrane localization of protein kinase C delta. J Biol Chem 2009; 284: 31764–75. 30 Richardson GD, Fantauzzo KA, Bazzi H et al. Dynamic expression of syndecan-1 during hair follicle morphogenesis. Gene Expr Patterns 2009; 9: 454–60. 31 Stepp MA, Pal-Ghosh S, Tadvalkar GLoss of syndecan-1 is associated with malignant conversion in skin carcinogenesis. Mol Carcinog 2010; 49: 363–73. 32 Sellheyer K, Smoller BR. Dermatofibroma: upregulation of syndecan-1 expression in mesenchymal tissue. Am J Dermatopathol 2003; 25: 392–8. 33 Metzler G, Schaumburg-Lever G, Liebig K. Ultrastructural localization of keratin and alpha-L-fucose in human eccrine sweat glands. Arch Dermatol Res 1990; 282: 12–6. 34 Demirkesen C, Hoede N, Moll R. Epithelial markers and differentiation in adnexal neoplasms of the skin: an immunohistochemical study including individual cytokeratins. J Cutan Pathol 1995; 22: 518–35. 35 Haimoto H, Hosoda S, Kato K. Differential distribution of immunoreactive S100-alpha and S100-beta proteins in normal nonnervous human tissues. J Tech Methods Pathol 1987; 57: 489. 36 Pierard-Franchimont C, Colige A, Arrese Estrada J et al. Immunohistochemical expression of epidermal growth factor receptors in nuclei of a subpopulation of keratinocytes and sweat gland cells. Dermatologica 1991; 183: 7–9. © 2013 Wiley Periodicals, Inc.