0021-972x/92/7505-1368$03.00/0 Journal of Clinical Endocrinology and Metabolism Copynght (c) 1992 by The Endocrine Society
Vol. 15, No. 5 Prmted in U.S.A.
Demonstration and Localization of Growth Hormone Receptor in Human Skin and Skin Fibroblasts” S. R. OAKESf$, K. M. HAYNES, AND G. A. WERTHER
M. J. WATERS,
A. C. HERINGTONT,
Department of Endocrinology, Royal Children’s Hospital, and Prince Henry’s (A. C.H.), Prince Henry’s Hospital, Melbourne, and Department of Physiology University of Queensland, St. Lucia, Qld., Australia ABSTRACT
Institute of Medical Research and Pharmacology (M. J. W.),
resulted in granular cytoplasmic staining. Fibroblast poly A+ RNA was prepared from cultured skin fibroblasts, separated by denaturing agarose gel electrophoresis, blotted onto nitrocellulose, and hybridized to a 32P-labeled, 847 base pair (bp) hGH receptor complementary DNA (cDNA) clone. Human liver and nonpregnant rabbit liver total RNA were used as controls. Fibroblast poly A+ RNA contained a single hybridizing species of approximately 5.2 kilobase. Human liver total RNA also contained a single hybridizing species of 4.9 kilobase. We have demonstrated the presence of GH receptor protein in human skin and growth hormone receptor mRNA and protein in cultured human skin fibroblasts. These observations suggest that GH may indeed have a direct role in modulating keratinocyte and fibroblast function. (J Clin Endocrinol Metab 75: 13681373, 1992)
Clinical evidence suggests that skin is responsive to GH status in uiuo. We sought to demonstrate the presence of GH receptors in human skin and in cultured skin fibroblasts using the techniques of immunohistochemistry and northern blotting. Human foreskin was obtained at surgery for preparation of sections and primary fibroblast cultures. Skin sections and fibroblast monolayers were immunostained using a monoclonal antibody which recognizes the hGH receptor (MAb 263). Positive immunoperoxidase staining was seen in all epidermal layers except the stratum corneum, in dermal sweat and sebaceous glands, and in dermal fibroblasts. In cultured fibroblasts capping of surface GH receptor was observed after aqueous formaldehyde fixation, whereas fixation in Carnoy’s solution
A
LTERATIONS in skin texture and thickness occur in clinical states of GH deficiency and excess (1). Thus, in acromegaly the skin is thickened and coarsened primarily due to an increase in the amount of dermal collagen (2), whereas in GH deficiency the skin is thin and delicate in character. Studies in intact rats have also shown an effect of administered biosynthetic hGH on collagen content and mechanical strength of skin (3). In dogs, GH deficiency is associated with generalized alopecia (hair loss) and disorganized skin histology, with recovery of hair growth after GH therapy (4). These effects suggest that skin may be a target tissue for GH, either indirectly via circulating insulin-like growth factor I (IGF-I) or directly via interaction of GH with specific receptors on the surface of epidermal and dermal cells. Direct effects of GH on cultured human skin fibroblasts have been demonstrated, including increased IGF-I (5) and IGF binding protein production (6), and increased thymidine uptake (7). Direct effects of GH require the presence of specific GH receptors which, apart from one study (8) in which ‘%GH binding was extremely low, has not been reported on skin fibroblasts. We now report, using the techniques of immunohistochemistry and northern blotting the presence of GH receptor in sections of human skin and in cultured human skin fibroblasts.
MATERIALS
AND
METHODS
Tissue culture Skin tissue was obtained from normal children aged 3-15 yr undergoing circumcision. Primary cultures of fibroblasts were established from this tissue by culture in Dulbecco’s Modification of Eagle’s Medium supplemented with 10% fetal calf serum (DMEM\lO% FCS) and added antibiotics at 37 C in a 5% COJ95% air atmosphere. The cells were maintained by passage at a 1:4 split ratio.
Immunohistochemistry Human foreskins were obtained within 30 min of surgery, fixed in Carnoy’s solution (ethanol/chloroform/glacial acetic acid 6:3:1) for 4 h at room temperature and paraffin embedded overnight. 12-pm sections were then cut, slide mounted and dewaxed by immersion in xylene and graded alcohol solutions before immunochemical staining. Fibroblasts of passage 5-15 were grown in slide chambers (Labtek, Nunc, Naperville, IL) and washed in PBS before fixation at room temperature for 30 min in either 4% formaldehyde/PBS or Carnoy’s solution. The primary antibody used for immunohistochemistry was a mouse monoclonal antiGH receptor antibody, MAb 263 (9), used at a concentration of 18 Kg/ mL. This monoclonal antibody was raised against rat liver membrane GH receptor but also recognizes the hGH receptor as evidenced by immunoprecipitation of solubilized. ‘251-labeled IM-9 lymphocyte GH receptor (10) and human serum GH-binding protein (11). An unrelated anti-Brucella monoclonal antibody was used for control immunostaining at the same protein concentration. The Vectastain (Vector Laboratories Inc., Burlingame, CA) avidin-biotin-peroxidase kit and protocol were used for immunostaining. Slides were then hematoxylin-counterstained and mounted.
Received September 16, 1991. Address requests for reprints to: Dr. Simon Oakes, Royal Children’s Hospital, Flemington Road, Parkville, Victoria 3052, Australia. * These studies were supported in part by the National Health and Medical Research Council of Australia (NHMRC). t NHMRC Medical Postgraduate Research Scholar. $ Present address: Department of Biochemistry, Royal Children’s Hospital, Flemington Road, Parkville, Victoria, Australia 3052.
Northern
blotting
Normal human skin fibroblasts (passage 5) were grown to confluence in DMEM\lO% FCS in roller bottles, trypsinized, and harvested. Cells were then washed in PBS, resuspended in TENS buffer [Tris, 10 mM, pH 7.5; EDTA, 1 mM; sodium chloride (NaCl), 0.1 M; and sodium dodecyl 1368
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
GH RECEPTOR
IN HUMAN
sulfate (SDS) 0.5%] with Proteinase K (Boehringer Mannheim, Mannheim, Germany) at 0.15 mg/mL, homogenized in a Dounce homogenizer (10 strokes at room temperature) and incubated overnight at 37 C. Poly A+ RNA was prepared from this cell lysate by chromatography on oligo dT cellulose (Boehringer Mannheim) and the yield determined by absorbance at 260 nm. Twenty five micrograms of poly A+ RNA per lane was denatured by heating at 65 C for i5 min inAdeionized formamide (50%), formaldehvde (18%). and MOPS x20 (2.5%) (moroholinouroRanesulfonic acid,‘0.8 h, pH.7.0; sodium acetate; 0.2 &;‘and’EDTA, 6.02 M). Samples were then loaded onto a 0.8% agarose gel containing formaldehyde (18%) and MOPS x20 (2.5%) and run at 75 mA constant current. Female nonpregnant rabbit liver total RNA and male human liver total RNA were included as positive controls. An RNA ladder (Bethesda Research Laboratories, Inc., Gaithersburg, MD) was run concurrently on the same gel to allow estimation of molecular size. The gel was treated with sodium hydroxide, 50 mM/NaCl, 10 rnM for 30 min, and then neutralized in Tris/HCl, 0.1 M, pH 7.5 before being blotted overnight onto nitrocellulose (Schleicher and Schuel, Dassel, Germany). The blot was then baked at 80 C for 2 h in a vacuum oven. The filter was prehybridized overnight at 42 C with 100 pg/mL denatured herring sperm DNA (Boehringer Mannheim) in hybridization buffer [deionized formamide (50%), 2~ Denhardt’s solution (1X Denhardt’s is polyvinylpyrrolidone 0.02%, BSA 0.02%, Ficoll 0.02%, and EDTA, 0.1 mM); 0.1% SDS; 5X SSC (1X SSC is NaCl, 150 mM and trisodium citrate, 15 mM), and 4 mM EDTA]. The filter was probed with the 847 bp hGH receptor cDNA clone pghr 501.1 (12), a gift from Dr. W. I. Wood (Genentech Pty. Ltd., South San Francisco, CA). The cDNA probe was labeled with 3ZP-CTP to a specific activity of 5x10’ cpm/pg DNA using a nick translation kit (Boehringer Mannheim) and purified on a Sephadex G50 (Pharmacia, Uppsala, Sweden) column. Hybridization was at 42 C overnight in hybridization buffer. The filter was washed at room temperature in 2X SSC/O.l% SDS, then at 50 C in 0.1~ SSC/O.l% SDS for 30 min, and exposed to Kodak XAR-5 XOMAT photographic film with intensifying screens for 4 days at -70 C.
Results Human foreskin sections showed strongly positive immunoperoxidase staining for GH receptor antigen in both epidermis and dermis. Epidermal staining occurred in all cell layers except the stratum corneum (Fig. 1A). Dermal staining occurred in the epithelium of sweat glands (Fig. lc) and the acini of sebaceous glands (Fig. 1E) as well as in scattered connective tissue fibroblasts. Staining was granular in nature and evenly distributed within cells. Identical staining patterns were observed in specimens obtained from seven different children ranging in age from 3-15 yr. Fibroblasts derived from foreskin tissue by primary culture also showed positive staining for GH receptor antigen. The nature of the staining observed varied with the method of fixation. Cells fixed in Carnoy’s solution (ethanol/chloroform/acetic acid) showed evenly distributed granular staining (Fig. 2B) whereas cells fixed in aqueous formaldehyde showed staining which was highly localized to one region of the cell (Fig. 2A). No staining was observed with the control monoclonal antibody at the same protein concentration in skin sections (Fig. 1, B, D, F) or cultured fibroblasts fixed in either aqueous formalin (Fig. 2C) or Carnoy’s solution (Fig. 2D). Fibroblast staining with monoclonal antibody 263 for GH receptor antigen was unaffected by preincubation with GH at 10 pLg/ mL (not shown). Pure GH receptor was not available for immunodepletion studies. Northern blot hybridization of poly A+ RNA (25 pg/lane) from normal human skin fibroblasts with a 847 bp hGH receptor cDNA detected a single hybridizing mRNA species
SKIN
AND
1369
FIBROBLASTS
of approximately 5.2 kilobases in size (Fig. 3). Human liver total RNA also showed one major mRNA species of approximately 4.9 kilobases. Control rabbit liver total RNA showed the expected major mRNA species of 4.6 and 3.2 kilobases (13). Discussion This study reports the presence of GH receptors in human skin. Similar results have been reported in a parallel study using the same GH receptor antibody in rat skin (14). The current study extends these findings to cultured human skin fibroblasts, and presents evidence for expression of receptor protein and its messenger RNA (mRNA) in these cells. Receptor immunoreactivity was evenly distributed throughout the epidermal layers with the exception of the stratum corneum where it was absent. The absence of a maturational variation in receptor expression from the basal germinal keratinocytes to the mature cells of the spinous layer is significant in view of the postulated role of GH as a differentiation factor (15). The staining seen suggests that GH may have a role in the normal functioning of the fully differentiated keratinocyte. The difference in staining pattern with the different fixation methods may reflect detection of intracellular and cell surface GH receptor antigen, respectively. Ethanol based fixatives allow good penetration of antibody for detection of intracellular antigen (16), whereas mild aqueous or no prefixation is necessary for exclusive detection of cell surface antigen (17). The staining pattern observed with aqueous fixation is similar in appearance to the cap formation induced in lymphocytes by antibodies directed to surface membrane immunoglobulin (18) and membrane insulin receptors (19). We suggest that the observed staining pattern may represent GH receptor antigen aggregation induced by the double antibody detection system used. The apparent lateral mobility of the fibroblast membrane GH receptor is consistent with the aggregation of GH receptor observed after GH binding to IM-9 lymphocytes (20). Aqueous formaldehyde fixation is mild and reversible within l-2 h (21) and therefore would not be inconsistent with the suggested receptor aggregation. The intracellular staining for GH receptor detected with Carnoy’s fixation is not surprising given that the GH receptor is known to cycle rapidly between the plasma membrane and a cytoplasmic pool (22) and cell cytosols of several rabbit tissues contain GH-binding proteins (23). Fibroblast staining with monoclonal antibody 263 for GH receptor antigen was unaffected by preincubation with GH at 10 pg/mL (not shown) consistent with the recognition by this antibody of the GH-GHR complex (10) and indicating that its epitope is not limited to, or does not primarily involve, the GH-binding site. We have shown however, in coincubation experiments using cultured human chondrocytes, that radiolabeled GH binding is inhibited by this antibody (24), indicating proximity between the GH binding site and the epitope recognized by MAb 263. The presence of GH receptors on human fibroblasts is supported by the demonstration of specific GH receptor mRNA in these cells. The cDNA used in this study codes for the extracellular and cell membrane portions of the cloned
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
GH RECEPTOR
IN HUMAN
SKIN
AND
FIBROBLASTS
1371
FIG. 2. Immunochemical staining of cultured human foreskin fibroblast ;s with anti-GH receptor (A, B) and control (C, D) monoclonal antibodies after fixation in aqueous formaldehyde (A, C) and Carnoy’s (B, D) solutic ms. Positive staining for GH receptor antigen occurs on the cell membrane with aqueous formaldehyde fixation (arrowed) and intracellularly (C) al ‘ter Carnoy’s fixation. Fibroblasts derived from primary culture of human foreskin explants in DMEM with 10% FCS and added antibiotics. (Mag :nification, X80.)
hepatic human GH receptor and also detected a single mRNA in human liver. The larger sized mRNA in fibroblasts suggestsa tissue specific difference in mRNA processing.A low molecular size mRNA that might code for a soluble GH binding protein was not seen in fibroblasts or liver. The demonstration of GH receptor antigen in human skin
tissue is consistent with the clinical observation that skin is sensitive to GH status. Our data support the likelihood that GH has a direct effect in skin in addition to a possibleindirect effect mediated by circulating IGF-I. A direct action of GH may be mediated locally by IGF-I (7), as GH stimulates IGFI production in cultured skin fibroblasts (5) and IGF-I has
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
1372
OAKES ET AL.
3. Northern blot analysis of human skin fibroblast poly A+ RNA (25 pg/lane) (a), human liver total RNA (20 rg/lane) (b) and rabbit liver total RNA (20 fig/lane) (c), using a 847 bp hGH receptor cDNA clone. A single hybridizing species of approximately 5.2 kilobases was detected in fibroblasts. Human liver also showed a single mRNA species of approximately 4.9 kilobases. Rabbit liver showed two mRNA species of 4.6 and 3.2 kilobases. Position of RNA size markers indicated.
FIG.
been shown to stimulate several fibroblast functions including cell replication (25) and type 1 collagen synthesis (26, and S. Oakes unpublished). Possibleroles for GH in connective tissue might include involvement in the early stagesof cell differentiation as has been shown for mouse embryonic 3T3 preadipose (27) and myoblast (28) cell lines, or involvement in the specialized functions of tissue matrix synthesis. Localization of GH receptor antigen to dermal sweat gland epithelium and sebaceousgland acini is an interesting finding in view of the clinical observation of increased sweating and sebaceoussecretions in acromegaly (1) and impaired sweat production in GH-deficient patients (29). The reasons for this effect are not known but it is possible that GH is a trophic factor for sweat gland and sebaceousgland epithelium. A direct role in hair growth is suggestedby the reversal of hair loss after GH administration in GH-deficient dogs (4). The tissuesexamined in this study were genital skin samples from males aged 3-15 yr. The question of possible
JCE & M .1992 Vol75.No5
age-, sex- or tissue-related changes in skin GH receptor antigen expression was not specifically addressed. Apart from a parallel study in the rat (14), no other reports of GH receptor expression in skin have been published. Animal studies have shown speciesand tissue differences with respect to the effects of age and sex on GH receptor expression (reviewed in 22), making prediction of the effects of these factors in human skin difficult. GH receptor immunoreactivity is present in fetal rat skin (30) but whether this truly represents GH receptor or an antigenically related protein is not known. Genital skin undergoes growth at puberty under the influence of sex hormones and in this respect differs from nongenital skin, however no difference was discernable in the immunoreactivity seen in skin obtained from prepubertal compared to pubertal males. The existence of GH receptors on skin fibroblasts is to be expected since several fibroblast functions have been shown to be modulated by GH. Murphy et al. (8) have reported specific binding of GH to embryonic and postnatal human fibroblast cell lines (l-2% specific binding per lo6 cells). Using similar techniques in several adult skin fibroblast lines we have obtained lower levels of specific binding (0.140.57% per lo6 cells, mean 0.30%, n = 13, unpublished observation). Satisfactory and reproduceable data however could not be obtained. The apparent discrepancy between the difficulty in demonstrating GH receptors by binding techniques and the results of immunochemical studies presented above is most easily explained by the low abundance of GH receptors on fibroblasts and the intrinsically greater sensitivity of enzymatically amplified immunochemical detection compared with radiolabel binding techniques. A low abundance of GH receptor is supported by the fact that the large amount of 25 pg/lane of poly A+ mRNA was required to detect the presence of GH receptor mRNA by northern blotting. An alternative explanation for the discrepancy between specific GH binding and immunochemical detection is that the GH receptor antigen detected by immunostaining is nonfunctional. This is not likely given the: reported action of GH in fibroblasts. In summary, we have demonstrated by immunochemical means the presence of GH receptor antigen in human skin and cultured skin fibroblasts and by northern blotting the presenceof GH mRNA in cultured fibroblasts. Thesefindings suggestthat skin may be a target for direct GH action in vim. References 1. Daughaday WH. 1985 The anterior pituitary. In: Wilson JD, Foster DW, eds. William’s textbook of endocrinology. 7th ed. Philadelphia: WB Saunders; 598-602. 2. Gabrilove JL, Schwartz A, Churg J. 1962 Effect of hormones on the skin in endocrinologic diseases. J Clin Endocrinol Metab. 22: 688-92. 3. Jorgensen PH, Andreassen TT, Jorgensen KD. 1989 Growth hormone influences collagen depo&tio% and mechanical strength of intact rat skin. Acta Endocrinol (Couenh). 120:767-72. 4. Eigenmann JE. 1981 Diagnosis a;d treatment of dwarfism in a G&man shepherd dog. Journal Am Anim Hosp Assoc. 17:798-804. 5. Clemmons DR. Underwood LE. Van Wvk II. 1981 Hormonal control of immbnoreactive somatomedin iro&ction by cultured human fibroblasts. J Clin Invest 67:10-9. 6. Conover CA, Liu F, Powell D, Rosenfeld RG, Hintz RL. 1989 Insulin-like growth factor binding proteins from cultured human
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
GH RECEPTOR
7.
8.
9.
10.
11.
12.
13.
14.
15. 16.
17.
18.
IN
HUMAN
fibroblasts: characterisation and hormonal regulation. J Clin Invest. 83:852-9. Cook JJ, Haynes KM, Werther GA. 1988 Mitogenic effects of growth hormone in cultured human fibroblasts. J Clin Invest. 81:206-12. Murphy LJ, Vrhovsek E, Lazarus L 1983 Identification and characterisation of specific growth hormone receptors in cultured human fibroblasts. J Clin Endocrinol Metab. 57:1117-24. Barnard R, Bundesen PG, Rylatt DB, Waters MJ. 1985 Evidence from the use of monoclonal antibody probes for structural heterogeneity of the growth hormone receptor. Biochem J. 231:459-68. Asakawa K, Hedo JA, McElduff A, Rouiller DG, Waters MJ, Gorden P. 1986 The human growth hormone receptor of cultured lymphocytes. Biochem J. 238:379-86. Barnard R, Quirk P, Waters MJ. 1989 Characterisation of the growth hormone-binding protein of human serum using a panel of monoclonal antibodies. J-&docrinol. 123:327-32. - ^ Leune DW, Suencer SA, Cachianes G, et al. 1987 Growth hormone receptor and serum binding protein: purification, cloning and expression. Nature 330:537-43. Tiong TS, Freed KA, Herington AC. 1989 Identification and tissue distribution of messenger RNA for the growth hormone receptor in the rabbit. Biochem Biophys Res Commun. 158141-8. Lobie PE, Breipohl W, Lincoln DT, Garcia-Aragon J, Waters MJ. 1990 Localisation of the growth hormone/binding protein in skin. J Endocrinol. 126:467-72. Green H, Morikawa M, Nixon T. 1985 A dual effector theory of growth-hormone action. Differentiation. 29:195-8. Hartman AL, Nakane PK. 1983 Immuno-peroxidase methods in scanning electron microscopy. In: Bullock GR, Petrusz I’, eds. Techniques in immunocytochemistry. Vol. 2. London: Academic Press; 108-9. Roth J. 1983 Colloidal gold markers for cytochemistry. In: Bullock GR, Petrusz P, eds. Techniques in immunocytochemistry. Vol. 2. London: Academic Press; 239-41. Depetris S, Raff MC. 1973 Normal distribution, patching and capping of lymphocyte surface immunoglobulin studied by electron
SKIN
AND
FIBROBLASTS
1373
microscopy. Nature. 241:257-9. 19. Schlessinger J, Van Obberghen E, Kahn CR. 1980 Insulin and antibodies against insulin receptor cap on the membrane of cultured human lymphocytes. Nature. 286:729-31. 20. Eshet R, Peleg S, Laron Z. 1984 Direct visualization of binding, aggregation and internalisation of human growth hormone in cultured human lymphocytes. Acta Endocrinol (Copenh). 107:9-15. 21. Tokuyasu KT, Singer SJ. 1976 Improved procedures for immunoferritin labelline of ultrathin frozen sections. 1 Cell Biol. 71:894-906. 22. Roupas P, He>on AC. 1989 Cellular mechanisms in the processing of growth hormone and its receptor. Mol Cell Endocrinol. 61:1-12. 23. Herington AC, Ymer S, Roupas P, Stevenson J. 1986 Growth hormo\e-binding proteins in high-speed cytosols of multiple tissues of the rabbit. Biochim Bioohvs Acta. 881:236-40. 24. Werther GA, Haynes K-M,‘Barnard R, Waters M. 1990 Visual demonstration and localisation of growth hormone receptors on human developing growth plate cartilage. J Clin Endocrinol Metab. 70:1725-31. 25. Clemmons DR, Van Wyk JJ. 1981 Somatomedin-c and plateletderived growth factor stimulate human fibroblast replication. J Cell Physiol. 106:361-7. 26. Goldstein HR, Poliks CF, Pilch PF, Smith BD, Fine A. 1989 Stimulation of collagen formation by insulin and insulin-like growth factor I in cultures of human lung fibroblasts. Endocrinology. 124:964-70. 27. Morikawa M, Nixon T, Green H. 1982 Growth hormone and the adipose conversion of 3T3 cells. Cell 29:783-9. 28. Nixon BT, Green H. 1984 Growth hormone promotes the differentiation of myoblasts and preadipocytes generated by azacytidine treatment of lOT1/2 cells. Proc Nat1 Acad Sci USA. 81:3429-32. 29 Pedersen SA, Welling K, Michaelsen KF, Jorgensen JO, Christiansen JS, Skakkebaek NE. 1989 Reduced sweating in adults with growth hormone deficiency. Lancet. 2:681-2. 30 Garcia-Araeon 1. Lobie PE, Gobius KS, Waters MI. Prenatal expression of the growth hormone receptor/binding in the rat: a role for growth hormone in embryonic and fetal development? [Abstract]. Proc of the Endocrine Society of Australia. 1990;33:60.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
Vol. 15, No. 5 Prmted in U.S.A.
Demonstration and Localization of Growth Hormone Receptor in Human Skin and Skin Fibroblasts” S. R. OAKESf$, K. M. HAYNES, AND G. A. WERTHER
M. J. WATERS,
A. C. HERINGTONT,
Department of Endocrinology, Royal Children’s Hospital, and Prince Henry’s (A. C.H.), Prince Henry’s Hospital, Melbourne, and Department of Physiology University of Queensland, St. Lucia, Qld., Australia ABSTRACT
Institute of Medical Research and Pharmacology (M. J. W.),
resulted in granular cytoplasmic staining. Fibroblast poly A+ RNA was prepared from cultured skin fibroblasts, separated by denaturing agarose gel electrophoresis, blotted onto nitrocellulose, and hybridized to a 32P-labeled, 847 base pair (bp) hGH receptor complementary DNA (cDNA) clone. Human liver and nonpregnant rabbit liver total RNA were used as controls. Fibroblast poly A+ RNA contained a single hybridizing species of approximately 5.2 kilobase. Human liver total RNA also contained a single hybridizing species of 4.9 kilobase. We have demonstrated the presence of GH receptor protein in human skin and growth hormone receptor mRNA and protein in cultured human skin fibroblasts. These observations suggest that GH may indeed have a direct role in modulating keratinocyte and fibroblast function. (J Clin Endocrinol Metab 75: 13681373, 1992)
Clinical evidence suggests that skin is responsive to GH status in uiuo. We sought to demonstrate the presence of GH receptors in human skin and in cultured skin fibroblasts using the techniques of immunohistochemistry and northern blotting. Human foreskin was obtained at surgery for preparation of sections and primary fibroblast cultures. Skin sections and fibroblast monolayers were immunostained using a monoclonal antibody which recognizes the hGH receptor (MAb 263). Positive immunoperoxidase staining was seen in all epidermal layers except the stratum corneum, in dermal sweat and sebaceous glands, and in dermal fibroblasts. In cultured fibroblasts capping of surface GH receptor was observed after aqueous formaldehyde fixation, whereas fixation in Carnoy’s solution
A
LTERATIONS in skin texture and thickness occur in clinical states of GH deficiency and excess (1). Thus, in acromegaly the skin is thickened and coarsened primarily due to an increase in the amount of dermal collagen (2), whereas in GH deficiency the skin is thin and delicate in character. Studies in intact rats have also shown an effect of administered biosynthetic hGH on collagen content and mechanical strength of skin (3). In dogs, GH deficiency is associated with generalized alopecia (hair loss) and disorganized skin histology, with recovery of hair growth after GH therapy (4). These effects suggest that skin may be a target tissue for GH, either indirectly via circulating insulin-like growth factor I (IGF-I) or directly via interaction of GH with specific receptors on the surface of epidermal and dermal cells. Direct effects of GH on cultured human skin fibroblasts have been demonstrated, including increased IGF-I (5) and IGF binding protein production (6), and increased thymidine uptake (7). Direct effects of GH require the presence of specific GH receptors which, apart from one study (8) in which ‘%GH binding was extremely low, has not been reported on skin fibroblasts. We now report, using the techniques of immunohistochemistry and northern blotting the presence of GH receptor in sections of human skin and in cultured human skin fibroblasts.
MATERIALS
AND
METHODS
Tissue culture Skin tissue was obtained from normal children aged 3-15 yr undergoing circumcision. Primary cultures of fibroblasts were established from this tissue by culture in Dulbecco’s Modification of Eagle’s Medium supplemented with 10% fetal calf serum (DMEM\lO% FCS) and added antibiotics at 37 C in a 5% COJ95% air atmosphere. The cells were maintained by passage at a 1:4 split ratio.
Immunohistochemistry Human foreskins were obtained within 30 min of surgery, fixed in Carnoy’s solution (ethanol/chloroform/glacial acetic acid 6:3:1) for 4 h at room temperature and paraffin embedded overnight. 12-pm sections were then cut, slide mounted and dewaxed by immersion in xylene and graded alcohol solutions before immunochemical staining. Fibroblasts of passage 5-15 were grown in slide chambers (Labtek, Nunc, Naperville, IL) and washed in PBS before fixation at room temperature for 30 min in either 4% formaldehyde/PBS or Carnoy’s solution. The primary antibody used for immunohistochemistry was a mouse monoclonal antiGH receptor antibody, MAb 263 (9), used at a concentration of 18 Kg/ mL. This monoclonal antibody was raised against rat liver membrane GH receptor but also recognizes the hGH receptor as evidenced by immunoprecipitation of solubilized. ‘251-labeled IM-9 lymphocyte GH receptor (10) and human serum GH-binding protein (11). An unrelated anti-Brucella monoclonal antibody was used for control immunostaining at the same protein concentration. The Vectastain (Vector Laboratories Inc., Burlingame, CA) avidin-biotin-peroxidase kit and protocol were used for immunostaining. Slides were then hematoxylin-counterstained and mounted.
Received September 16, 1991. Address requests for reprints to: Dr. Simon Oakes, Royal Children’s Hospital, Flemington Road, Parkville, Victoria 3052, Australia. * These studies were supported in part by the National Health and Medical Research Council of Australia (NHMRC). t NHMRC Medical Postgraduate Research Scholar. $ Present address: Department of Biochemistry, Royal Children’s Hospital, Flemington Road, Parkville, Victoria, Australia 3052.
Northern
blotting
Normal human skin fibroblasts (passage 5) were grown to confluence in DMEM\lO% FCS in roller bottles, trypsinized, and harvested. Cells were then washed in PBS, resuspended in TENS buffer [Tris, 10 mM, pH 7.5; EDTA, 1 mM; sodium chloride (NaCl), 0.1 M; and sodium dodecyl 1368
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
GH RECEPTOR
IN HUMAN
sulfate (SDS) 0.5%] with Proteinase K (Boehringer Mannheim, Mannheim, Germany) at 0.15 mg/mL, homogenized in a Dounce homogenizer (10 strokes at room temperature) and incubated overnight at 37 C. Poly A+ RNA was prepared from this cell lysate by chromatography on oligo dT cellulose (Boehringer Mannheim) and the yield determined by absorbance at 260 nm. Twenty five micrograms of poly A+ RNA per lane was denatured by heating at 65 C for i5 min inAdeionized formamide (50%), formaldehvde (18%). and MOPS x20 (2.5%) (moroholinouroRanesulfonic acid,‘0.8 h, pH.7.0; sodium acetate; 0.2 &;‘and’EDTA, 6.02 M). Samples were then loaded onto a 0.8% agarose gel containing formaldehyde (18%) and MOPS x20 (2.5%) and run at 75 mA constant current. Female nonpregnant rabbit liver total RNA and male human liver total RNA were included as positive controls. An RNA ladder (Bethesda Research Laboratories, Inc., Gaithersburg, MD) was run concurrently on the same gel to allow estimation of molecular size. The gel was treated with sodium hydroxide, 50 mM/NaCl, 10 rnM for 30 min, and then neutralized in Tris/HCl, 0.1 M, pH 7.5 before being blotted overnight onto nitrocellulose (Schleicher and Schuel, Dassel, Germany). The blot was then baked at 80 C for 2 h in a vacuum oven. The filter was prehybridized overnight at 42 C with 100 pg/mL denatured herring sperm DNA (Boehringer Mannheim) in hybridization buffer [deionized formamide (50%), 2~ Denhardt’s solution (1X Denhardt’s is polyvinylpyrrolidone 0.02%, BSA 0.02%, Ficoll 0.02%, and EDTA, 0.1 mM); 0.1% SDS; 5X SSC (1X SSC is NaCl, 150 mM and trisodium citrate, 15 mM), and 4 mM EDTA]. The filter was probed with the 847 bp hGH receptor cDNA clone pghr 501.1 (12), a gift from Dr. W. I. Wood (Genentech Pty. Ltd., South San Francisco, CA). The cDNA probe was labeled with 3ZP-CTP to a specific activity of 5x10’ cpm/pg DNA using a nick translation kit (Boehringer Mannheim) and purified on a Sephadex G50 (Pharmacia, Uppsala, Sweden) column. Hybridization was at 42 C overnight in hybridization buffer. The filter was washed at room temperature in 2X SSC/O.l% SDS, then at 50 C in 0.1~ SSC/O.l% SDS for 30 min, and exposed to Kodak XAR-5 XOMAT photographic film with intensifying screens for 4 days at -70 C.
Results Human foreskin sections showed strongly positive immunoperoxidase staining for GH receptor antigen in both epidermis and dermis. Epidermal staining occurred in all cell layers except the stratum corneum (Fig. 1A). Dermal staining occurred in the epithelium of sweat glands (Fig. lc) and the acini of sebaceous glands (Fig. 1E) as well as in scattered connective tissue fibroblasts. Staining was granular in nature and evenly distributed within cells. Identical staining patterns were observed in specimens obtained from seven different children ranging in age from 3-15 yr. Fibroblasts derived from foreskin tissue by primary culture also showed positive staining for GH receptor antigen. The nature of the staining observed varied with the method of fixation. Cells fixed in Carnoy’s solution (ethanol/chloroform/acetic acid) showed evenly distributed granular staining (Fig. 2B) whereas cells fixed in aqueous formaldehyde showed staining which was highly localized to one region of the cell (Fig. 2A). No staining was observed with the control monoclonal antibody at the same protein concentration in skin sections (Fig. 1, B, D, F) or cultured fibroblasts fixed in either aqueous formalin (Fig. 2C) or Carnoy’s solution (Fig. 2D). Fibroblast staining with monoclonal antibody 263 for GH receptor antigen was unaffected by preincubation with GH at 10 pLg/ mL (not shown). Pure GH receptor was not available for immunodepletion studies. Northern blot hybridization of poly A+ RNA (25 pg/lane) from normal human skin fibroblasts with a 847 bp hGH receptor cDNA detected a single hybridizing mRNA species
SKIN
AND
1369
FIBROBLASTS
of approximately 5.2 kilobases in size (Fig. 3). Human liver total RNA also showed one major mRNA species of approximately 4.9 kilobases. Control rabbit liver total RNA showed the expected major mRNA species of 4.6 and 3.2 kilobases (13). Discussion This study reports the presence of GH receptors in human skin. Similar results have been reported in a parallel study using the same GH receptor antibody in rat skin (14). The current study extends these findings to cultured human skin fibroblasts, and presents evidence for expression of receptor protein and its messenger RNA (mRNA) in these cells. Receptor immunoreactivity was evenly distributed throughout the epidermal layers with the exception of the stratum corneum where it was absent. The absence of a maturational variation in receptor expression from the basal germinal keratinocytes to the mature cells of the spinous layer is significant in view of the postulated role of GH as a differentiation factor (15). The staining seen suggests that GH may have a role in the normal functioning of the fully differentiated keratinocyte. The difference in staining pattern with the different fixation methods may reflect detection of intracellular and cell surface GH receptor antigen, respectively. Ethanol based fixatives allow good penetration of antibody for detection of intracellular antigen (16), whereas mild aqueous or no prefixation is necessary for exclusive detection of cell surface antigen (17). The staining pattern observed with aqueous fixation is similar in appearance to the cap formation induced in lymphocytes by antibodies directed to surface membrane immunoglobulin (18) and membrane insulin receptors (19). We suggest that the observed staining pattern may represent GH receptor antigen aggregation induced by the double antibody detection system used. The apparent lateral mobility of the fibroblast membrane GH receptor is consistent with the aggregation of GH receptor observed after GH binding to IM-9 lymphocytes (20). Aqueous formaldehyde fixation is mild and reversible within l-2 h (21) and therefore would not be inconsistent with the suggested receptor aggregation. The intracellular staining for GH receptor detected with Carnoy’s fixation is not surprising given that the GH receptor is known to cycle rapidly between the plasma membrane and a cytoplasmic pool (22) and cell cytosols of several rabbit tissues contain GH-binding proteins (23). Fibroblast staining with monoclonal antibody 263 for GH receptor antigen was unaffected by preincubation with GH at 10 pg/mL (not shown) consistent with the recognition by this antibody of the GH-GHR complex (10) and indicating that its epitope is not limited to, or does not primarily involve, the GH-binding site. We have shown however, in coincubation experiments using cultured human chondrocytes, that radiolabeled GH binding is inhibited by this antibody (24), indicating proximity between the GH binding site and the epitope recognized by MAb 263. The presence of GH receptors on human fibroblasts is supported by the demonstration of specific GH receptor mRNA in these cells. The cDNA used in this study codes for the extracellular and cell membrane portions of the cloned
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
GH RECEPTOR
IN HUMAN
SKIN
AND
FIBROBLASTS
1371
FIG. 2. Immunochemical staining of cultured human foreskin fibroblast ;s with anti-GH receptor (A, B) and control (C, D) monoclonal antibodies after fixation in aqueous formaldehyde (A, C) and Carnoy’s (B, D) solutic ms. Positive staining for GH receptor antigen occurs on the cell membrane with aqueous formaldehyde fixation (arrowed) and intracellularly (C) al ‘ter Carnoy’s fixation. Fibroblasts derived from primary culture of human foreskin explants in DMEM with 10% FCS and added antibiotics. (Mag :nification, X80.)
hepatic human GH receptor and also detected a single mRNA in human liver. The larger sized mRNA in fibroblasts suggestsa tissue specific difference in mRNA processing.A low molecular size mRNA that might code for a soluble GH binding protein was not seen in fibroblasts or liver. The demonstration of GH receptor antigen in human skin
tissue is consistent with the clinical observation that skin is sensitive to GH status. Our data support the likelihood that GH has a direct effect in skin in addition to a possibleindirect effect mediated by circulating IGF-I. A direct action of GH may be mediated locally by IGF-I (7), as GH stimulates IGFI production in cultured skin fibroblasts (5) and IGF-I has
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
1372
OAKES ET AL.
3. Northern blot analysis of human skin fibroblast poly A+ RNA (25 pg/lane) (a), human liver total RNA (20 rg/lane) (b) and rabbit liver total RNA (20 fig/lane) (c), using a 847 bp hGH receptor cDNA clone. A single hybridizing species of approximately 5.2 kilobases was detected in fibroblasts. Human liver also showed a single mRNA species of approximately 4.9 kilobases. Rabbit liver showed two mRNA species of 4.6 and 3.2 kilobases. Position of RNA size markers indicated.
FIG.
been shown to stimulate several fibroblast functions including cell replication (25) and type 1 collagen synthesis (26, and S. Oakes unpublished). Possibleroles for GH in connective tissue might include involvement in the early stagesof cell differentiation as has been shown for mouse embryonic 3T3 preadipose (27) and myoblast (28) cell lines, or involvement in the specialized functions of tissue matrix synthesis. Localization of GH receptor antigen to dermal sweat gland epithelium and sebaceousgland acini is an interesting finding in view of the clinical observation of increased sweating and sebaceoussecretions in acromegaly (1) and impaired sweat production in GH-deficient patients (29). The reasons for this effect are not known but it is possible that GH is a trophic factor for sweat gland and sebaceousgland epithelium. A direct role in hair growth is suggestedby the reversal of hair loss after GH administration in GH-deficient dogs (4). The tissuesexamined in this study were genital skin samples from males aged 3-15 yr. The question of possible
JCE & M .1992 Vol75.No5
age-, sex- or tissue-related changes in skin GH receptor antigen expression was not specifically addressed. Apart from a parallel study in the rat (14), no other reports of GH receptor expression in skin have been published. Animal studies have shown speciesand tissue differences with respect to the effects of age and sex on GH receptor expression (reviewed in 22), making prediction of the effects of these factors in human skin difficult. GH receptor immunoreactivity is present in fetal rat skin (30) but whether this truly represents GH receptor or an antigenically related protein is not known. Genital skin undergoes growth at puberty under the influence of sex hormones and in this respect differs from nongenital skin, however no difference was discernable in the immunoreactivity seen in skin obtained from prepubertal compared to pubertal males. The existence of GH receptors on skin fibroblasts is to be expected since several fibroblast functions have been shown to be modulated by GH. Murphy et al. (8) have reported specific binding of GH to embryonic and postnatal human fibroblast cell lines (l-2% specific binding per lo6 cells). Using similar techniques in several adult skin fibroblast lines we have obtained lower levels of specific binding (0.140.57% per lo6 cells, mean 0.30%, n = 13, unpublished observation). Satisfactory and reproduceable data however could not be obtained. The apparent discrepancy between the difficulty in demonstrating GH receptors by binding techniques and the results of immunochemical studies presented above is most easily explained by the low abundance of GH receptors on fibroblasts and the intrinsically greater sensitivity of enzymatically amplified immunochemical detection compared with radiolabel binding techniques. A low abundance of GH receptor is supported by the fact that the large amount of 25 pg/lane of poly A+ mRNA was required to detect the presence of GH receptor mRNA by northern blotting. An alternative explanation for the discrepancy between specific GH binding and immunochemical detection is that the GH receptor antigen detected by immunostaining is nonfunctional. This is not likely given the: reported action of GH in fibroblasts. In summary, we have demonstrated by immunochemical means the presence of GH receptor antigen in human skin and cultured skin fibroblasts and by northern blotting the presenceof GH mRNA in cultured fibroblasts. Thesefindings suggestthat skin may be a target for direct GH action in vim. References 1. Daughaday WH. 1985 The anterior pituitary. In: Wilson JD, Foster DW, eds. William’s textbook of endocrinology. 7th ed. Philadelphia: WB Saunders; 598-602. 2. Gabrilove JL, Schwartz A, Churg J. 1962 Effect of hormones on the skin in endocrinologic diseases. J Clin Endocrinol Metab. 22: 688-92. 3. Jorgensen PH, Andreassen TT, Jorgensen KD. 1989 Growth hormone influences collagen depo&tio% and mechanical strength of intact rat skin. Acta Endocrinol (Couenh). 120:767-72. 4. Eigenmann JE. 1981 Diagnosis a;d treatment of dwarfism in a G&man shepherd dog. Journal Am Anim Hosp Assoc. 17:798-804. 5. Clemmons DR. Underwood LE. Van Wvk II. 1981 Hormonal control of immbnoreactive somatomedin iro&ction by cultured human fibroblasts. J Clin Invest 67:10-9. 6. Conover CA, Liu F, Powell D, Rosenfeld RG, Hintz RL. 1989 Insulin-like growth factor binding proteins from cultured human
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
GH RECEPTOR
7.
8.
9.
10.
11.
12.
13.
14.
15. 16.
17.
18.
IN
HUMAN
fibroblasts: characterisation and hormonal regulation. J Clin Invest. 83:852-9. Cook JJ, Haynes KM, Werther GA. 1988 Mitogenic effects of growth hormone in cultured human fibroblasts. J Clin Invest. 81:206-12. Murphy LJ, Vrhovsek E, Lazarus L 1983 Identification and characterisation of specific growth hormone receptors in cultured human fibroblasts. J Clin Endocrinol Metab. 57:1117-24. Barnard R, Bundesen PG, Rylatt DB, Waters MJ. 1985 Evidence from the use of monoclonal antibody probes for structural heterogeneity of the growth hormone receptor. Biochem J. 231:459-68. Asakawa K, Hedo JA, McElduff A, Rouiller DG, Waters MJ, Gorden P. 1986 The human growth hormone receptor of cultured lymphocytes. Biochem J. 238:379-86. Barnard R, Quirk P, Waters MJ. 1989 Characterisation of the growth hormone-binding protein of human serum using a panel of monoclonal antibodies. J-&docrinol. 123:327-32. - ^ Leune DW, Suencer SA, Cachianes G, et al. 1987 Growth hormone receptor and serum binding protein: purification, cloning and expression. Nature 330:537-43. Tiong TS, Freed KA, Herington AC. 1989 Identification and tissue distribution of messenger RNA for the growth hormone receptor in the rabbit. Biochem Biophys Res Commun. 158141-8. Lobie PE, Breipohl W, Lincoln DT, Garcia-Aragon J, Waters MJ. 1990 Localisation of the growth hormone/binding protein in skin. J Endocrinol. 126:467-72. Green H, Morikawa M, Nixon T. 1985 A dual effector theory of growth-hormone action. Differentiation. 29:195-8. Hartman AL, Nakane PK. 1983 Immuno-peroxidase methods in scanning electron microscopy. In: Bullock GR, Petrusz I’, eds. Techniques in immunocytochemistry. Vol. 2. London: Academic Press; 108-9. Roth J. 1983 Colloidal gold markers for cytochemistry. In: Bullock GR, Petrusz P, eds. Techniques in immunocytochemistry. Vol. 2. London: Academic Press; 239-41. Depetris S, Raff MC. 1973 Normal distribution, patching and capping of lymphocyte surface immunoglobulin studied by electron
SKIN
AND
FIBROBLASTS
1373
microscopy. Nature. 241:257-9. 19. Schlessinger J, Van Obberghen E, Kahn CR. 1980 Insulin and antibodies against insulin receptor cap on the membrane of cultured human lymphocytes. Nature. 286:729-31. 20. Eshet R, Peleg S, Laron Z. 1984 Direct visualization of binding, aggregation and internalisation of human growth hormone in cultured human lymphocytes. Acta Endocrinol (Copenh). 107:9-15. 21. Tokuyasu KT, Singer SJ. 1976 Improved procedures for immunoferritin labelline of ultrathin frozen sections. 1 Cell Biol. 71:894-906. 22. Roupas P, He>on AC. 1989 Cellular mechanisms in the processing of growth hormone and its receptor. Mol Cell Endocrinol. 61:1-12. 23. Herington AC, Ymer S, Roupas P, Stevenson J. 1986 Growth hormo\e-binding proteins in high-speed cytosols of multiple tissues of the rabbit. Biochim Bioohvs Acta. 881:236-40. 24. Werther GA, Haynes K-M,‘Barnard R, Waters M. 1990 Visual demonstration and localisation of growth hormone receptors on human developing growth plate cartilage. J Clin Endocrinol Metab. 70:1725-31. 25. Clemmons DR, Van Wyk JJ. 1981 Somatomedin-c and plateletderived growth factor stimulate human fibroblast replication. J Cell Physiol. 106:361-7. 26. Goldstein HR, Poliks CF, Pilch PF, Smith BD, Fine A. 1989 Stimulation of collagen formation by insulin and insulin-like growth factor I in cultures of human lung fibroblasts. Endocrinology. 124:964-70. 27. Morikawa M, Nixon T, Green H. 1982 Growth hormone and the adipose conversion of 3T3 cells. Cell 29:783-9. 28. Nixon BT, Green H. 1984 Growth hormone promotes the differentiation of myoblasts and preadipocytes generated by azacytidine treatment of lOT1/2 cells. Proc Nat1 Acad Sci USA. 81:3429-32. 29 Pedersen SA, Welling K, Michaelsen KF, Jorgensen JO, Christiansen JS, Skakkebaek NE. 1989 Reduced sweating in adults with growth hormone deficiency. Lancet. 2:681-2. 30 Garcia-Araeon 1. Lobie PE, Gobius KS, Waters MI. Prenatal expression of the growth hormone receptor/binding in the rat: a role for growth hormone in embryonic and fetal development? [Abstract]. Proc of the Endocrine Society of Australia. 1990;33:60.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 10:28 For personal use only. No other uses without permission. . All rights reserved.
