Distribution of Interferon-Gamma Receptors in Normal and Psoriatic Skin

Path. Res. Pract. 191, 530-534 (1995)

Distribution of Interferon-Gamma Receptors in Normal and Psoriatic Skin >:. J. J. van den Oord, M. De Ley1 and C. De Wolf-Peeters Department of Pathology, Laboratory of Histo- and Cytochemistry, and 1Department of Chemistry, Laboratory of Biochemistry, Katholieke Universiteit Leuven, Leuven, Belgium

SUMMARY Recent data suggest that imbalances in production and secretion of cytokines, in particular interferon-gamma (IFN-y), may be crucial in the pathogenesis of psoriasis. In order to exert its role on target cells, IFN-y has to interact with a specific cell membrane receptor termed the IFN-y-receptor (IFN-yR). We studied the distribution of IFN-yRs in frozen skin biopsies from 25 psoriatics and 5 normal controls with two unrelated monoclonal antibodies, and compared its distribution with that of the IFNy-inducible HLADR- and ICAM-1 antigens. In normal skin, IFN-yRs were restricted to the basal cell layer; weak staining was found on scattered mononuclear cells in the papillary dermis. In 13/25 active psoriatic lesions, additional suprabasal immunoreactive foci, and in 5/25 cases, diffuse immunoreactivity of the entire epidermis were seen. No striking topographical similarities between the site and number of IFN-yR+, HLADR+ and ICAM-1+ keratinocyte foci were observed, suggesting that cytokines other than IFN-y induce HLADR-antigens on psoriatic keratinocytes in vivo. The restricted distribution of IFN-yR on the germinative cell layer in normal skin confirms the role played by IFN-y in the normal growth regulation of the epidermis. The de novo suprabasal expression of IFN-y R in psoriasis argues against the current hypothesis that IFN-yR are down-regulated due to a local excess of IFN-y or transforming growth factor alpha (TGF-r:t.). Whether IFN-yRs in psoriatic skin are functionally normal and involved in signal transmission, remains to be studied.

Introduction Recent progress in immunodermatology has suggested that imbalances in cytokine production and secretion by inflammatory cells and keratinocytes playa central role in the pathogenesis and/or maintenance of the psoriatic lesions. In particular interferon-gamma (IFN-y) has been considered to be of crucial importance

* Dedicated to the memory of Prof. Dr. Pierre Dustin (1914-1993). 0344-0338/95/0191-0530$3.50/0

in the pathogenesis of psoriasis. This soluble cytokine is produced by activated T-Iymphocytes and would be involved in the recruitment of lymphocytes 32, in the induction of various activation and adhesion molecules on epidermal keratinocytes, e.g. HLA-DR and intercellular adhesion molecule-1 (IeAM-1) 14, as well as in the abnormal keratinocyte proliferation characteristic of psoriasis 6 • IFN-y has been detected in psoriatic epidermis by immunohistochemistry25 and its mRNA by polymerase chain reaction? Moreover, intradermal administration of IFN-y to healthy subjects induces psoriatic changes 5 and subcutaneous injection of © 1995 by Gustav Fischer Verlag, Stuttgart

Interferon-Gamma Receptors in Normal and Psoriatic Skin· 531

recombinant IFN-y in clinically symptomless skin of psoriatic patients can evoke punctate psoriatic lesions at the injection site 15 • Finally, local or systemic therapy with cyclosporin A clears psoriasis in man through inhibition of IFN-y transcription in T-cells 17,23. In order to exert its role on target cells, IFN-y has to interact with a specific cell membrane receptor, termed the IFN-y-receptor (IFN-yR). This receptor consists of a 90-kDa glycoprotein that binds IFN-y with high affinity, and is found on both lymphoid and non-lymphoid cells39 • It has been shown by several groups, however, that in order to constitute a fully functional IFN-yR, at least one additional component coded on human chromosome 21, is required 24 • We have studied the distribution of the IFN-yR in active psoriatic lesions using two different, previously characterized monoclonal antibodies, and have compared its distribution with that of the IFN-y inducible antigens HLA-DR and ICAM-1. Results In histologically normal skin, both mabs GIR-208 and R1G10 revealed identical staining patterns. An intense, cytoplasmic and membranous immunoreactivity was observed in the basal cell layer of the epidermis; no nuclear staining was observed (Fig. 1a). This immunoreactivity extended into the basal epithelium of the infundibular and isthmic parts of the hair follicle, whereas the outer hair root sheath below the isthmus remained unreactive. Positivity was also observed in the cytoplasm of the eccrine ducts whereas the cells of the secretory coil were negative. No immunoreactivity was observed on supra basal keratinocytes, stratum corneum, acrosyringium or Langerhans' cells. In the papillary dermis, weak staining was observed with mab GIR-208 on scattered perivascular mononuclear cells which occasionally exhibited dendritic features. In normal skin biopsies, HLA-DR-expression in the epidermis was restricted to the acrosyringium and scattered dendritic cells. No ICAM-1 staining was observed on keratinocytes of normal skin. In 22 out of 25 cases of psoriasis, HLA-DR-expression by keratinocytes was focal and involved the basal cell layer as well as variable numbers of suprabasal cellclusters. In one case of erythrodermic psoriasis, the epidermis showed diffuse HLA-DR staining. In two

Fig. 1. Expression ofIFN-yR in normal (la), chronic plaquetype psoriasis (lb) and erythrodermic psoriasis (lc). - la: In normal skin, immunoreactivity is confined to the basal cell layer. Ib: In the active psoriatic lesion, de-novo expression of IFN-yR is found in supra basal epidermal cell layers (lb) and additional scattered immunoreactive keratinocytes are found throughout the epidermis in erythrodermic psoriasis (lc). - Three-step indirect immunoperoxidase method on acetone-fixed frozen sections using mabs RIGI0 (a, c) and GIR-208 (b), counter stained with Harris' haematoxylin, x 110.

cases, no HLA-DR-expression was observed in the epidermis. In 16 cases, the epidermis showed multifocal, basal and supra basal staining for ICAM-1. Comparison of serial, immunostained sections revealed that ICAM-1 immunoreactivity coincided poorly with HLA-DR-po-

.

."

,

. ,. 1a

,"

,. ,/

J

,

,""

\

..

532 .

J. J. van den Oord et al.

sitivity; ICAM-1 + foci usually outnumbered HLADR+ foci but staining intensity was usually less than that for HLA-DR antigens. In all cases of psoriasis, the basal cell layer of the epidermis showed immunoreactivity for R1G10 and GIR208 but the intensity of staining varied from case to case. In addition, variably sized supra basal foci of epidermal immunoreactivity were observed in 13 cases (Fig. 1b). Comparison of serial sections did not reveal striking topographical similarities between HLA-DR+ or ICAM-1+ foci on the one hand, and R1G10+ or GIR108+ foci on the other. In 5 cases, including 2 cases presented with erythroderma, diffuse staining of the spinous layers occurred; the granular and horny layers remained negative. Within the diffusely stained epidermis, scattered keratinocytes showed a more intense labelling of their cytoplasm (Fig. 1c). Four out of these 5 cases showed many foci of HLA-DR+ keratinocytes, and 3 out of 5 showed a multifocal honeycomb pattern for ICAM-I. Discussion Using in-situ immunohistochemistry, we have shown the presence and distribution of IFN-yR in normal and psoriatic human skin. For this purpose, two unrelated mabs were used. GIR208 is a well characterized and extensively used mab that blocks the binding of radiolabelled IFN-y to human placental membranes and that inhibits IFN-y dependent antiviral activity in fibroblasts. It reacts with the 90 kD receptor as well as with the 55 kD break-down product that is closely linked to the receptor's ligand binding site 35 . On the other hand, R1G10 is a recently produced and characterized mab that inhibits the biological activity of IFN-y but does not recognize the IFN-y-binding site on the receptor; this antibody interferes with the signal transduction and appears to be directed against a transmembranous molecule, associated with the receptor12. Auxiliary molecules, associated with the ligand binding chain of the IFN-yR, are required for receptor signalling22 . There is recent evidence that mab R1G10 detects such an accessory molecule, involved in signal transduction 13. Although the two mabs used in this study, therefore, recognize different epitopes, both revealed an identical staining pattern in normal and psoriatic skin. In normal skin, immunoreactivity for both mabs was restricted to the basal cell layer where it stained the cytoplasm as well as the surface of basal keratinocytes. This cellular distribution is in line with a previous immuno-electronmicroscopical study that showed immunoreactivity for R1G10 on the cell membrane as well as in the rough endoplasmic reticulum and perinuclear vesicles of hepatocytes in areas of inflammation16 . It proves the actual synthesis of the receptor by the cell, and suggests the existence of an intracellular pool of IFN_yRll,21,26 that may serve in the immediate replacement of receptors following internalisation.

The restricted immunoreactivity for IFN-yR on basal cells correlates well with the distribution of the IFN-yinducible protein IP-10 that is found only in the basal cell layer of normal skin 19 . Moreover, recent studies employing affinity-purified rabbit polyclonal antibodies directed against the extracellular domain of the receptor have shown restricted immunoreactivity in the basal cell layer of oral cavity, larynx and esophagus 39 . Since the basal cell layer comprises the germinative compartment of the epidermis, our data may suggest that IFN-y itself (or one of the molecules induced by it), plays a role in normal growth regulation of the epidermis. Intradermal injection of IFN-y indeed upregulates keratinocyte cell growth in normal human skin6. Our observation that the most intense immunoreactivity was restricted to basal cells is also in line with the finding that basal keratinocytes have a greater susceptibility to IFN-y produced by activated dermal T-cells, than suprabasal keratinocytes5. Whereas the distribution of IFN-yR was restricted to the basal cell layer in normal skin, de novo expression of IFN-yR on suprabasal keratinocytes was observed in 18 out of 25 cases of psoriasis. This finding correlates well with the upregulated expression of the IFN-y-inducible protein IP-10 that appears in the supra basal layers in psoriatic skin 19 and with the finding of IFN-ymRNA in psoriatic but not in normal epidermis using the polymerase chain reaction? Moreover, our findings reflect the general observation that the proliferative compartment in active lesions of psoriasis is enlarged2, and that suprabasal keratinocytes acquire certain basal cell determinants 3?, 41, 42. Our results are in sharp contrast to the data of Scheynius et al. who recently conducted a similar study using a cocktail of mabs A6C5 and 4D7A1234. Both antibodies compete with receptor binding of human IFN-y and were used in an indirect immunofluorescence technique. They observed IFN-yR to be expressed in normal skin by keratinocytes throughout the epidermal cell layers except stratum corneum, whereas in psoriatic plaques, the lower part of the epidermis showed the same reactivity as normal skin, but no or minimal staining was found in the upper lesional epidermis. The down-regulation of IFN-yR in the upper lesional epidermis, observed by Scheynius et aI.3\ differs considerably from the up-regulation of IFN-yR in supra basal epidermal cell layers, observed in the present study. We have no definite explanation for these discrepancies although the application of different antibodies, the use of freshly cut frozen sections by us and the different sensitivity of the staining techniques may account for some of the divergent results. Apart from inducing epidermal hyperproliferation, IFN-y also induces a number of cell-surface molecules in cultured keratinocytes, including MHC class II molecules 8 and ICAM _1 32 . The expression of MH C class II or HLA-DR-antigens on keratinocytes is a well known phenomenon in a variety of inflammatory dermatoses which are accompanied by an infiltrate rich in activated T-cells 1, 3, 36. In psoriasis, however, variable results have

Interferon-Gamma Receptors in Normal and Psoriatic Skin· 533

been noted. Several investigators failed to find HLADR-expression on keratinocytes 1, 9, 10 whereas others found focal positivity in up to half of the patients 3, 18, 29, 38. In the present study, focal HLA-DRstaining was observed in 22 out of 25 cases, and diffuse staining of the entire epidermis was observed in only one case. Comparison with the distribution of IFNyR on serial sections however did not reveal any topographic correlation between foci immunoreactive for HLA-DR and those expressing the IFN-yR. Since keratinocytes express high affinity IFN-yR in vitr0 31, and since cultured psoriatic keratinocytes show a diminished response to IFN-y with regard to induction of HLA-DR-expression\ our data may suggest that cytokines other than IFN-y are involved in the induction of HLA-DR-antigens in vivo. The only part of the epidermis in normal human skin that constitutively expresses HLA-DR-antigens is the acrosyringium20, 27, 30. The underlying mechanism for this phenotype is as yet unknown, but our data suggest that it is not related to an increased susceptibility to the inductive effect of IFN-y through an enhanced number ofIFN-yR. Within the dermis, eccrine ducts were found to be immunoreactive for both mabs. Eccrine ducts are believed to represent a rich source of cytokines and growth factors since they have been found to exhibit immunoreactivity for interleukin (IL)-l r:t, and ~33, basic fibroblast growth factor, platelet derived growth factor and transforming growth factor r:t,16. Some of these cytokines may actually be produced by ductal epithelium, and be involved in autocrine loops since eccrine ducts display receptors for a great variety of cytokines, e.g. IL-1, IL-6 and IL-10 (personal unpublished data). Whether the immunoreactivity for IFN-yR reflects the true presence of the receptor and thus the capacity to respond to IFN-y, remains to be elucidated. In conclusion, we have shown that in the majority of psoriatic skin biopsies, IFN-yR are upregulated in suprabasal keratinocytes. These data speak against the current hypothesis that IFN-yR in psoriatic skin are down-regulated due to the local presence of increased levels of IFN _y9 or transforming growth factor-r:t,28. Whether the IFN-yR are functionally normal and involved in signal transmission, remains to be studied. Materials and Methods Skin biopsy specimens, taken from 25 patients for confirmation of the clinical diagnosis of psoriasis, formed the basis of this study. Twenty patients suffered from chronic plaquetype psoriasis (one with the clinical and histological picture of pustular psoriasis) and 4 patients presented erythroderma. One patient presented with guttate psoriasis. The majority of these skin biopsies included both lesional and adjacent normal, uninvolved skin. In addition, 5 biopsies without histological changes taken from non-psoriatic individuals were used as healthy controls. All biopsies were received fresh, and divided into two parts; one part was snap frozen in liquid nitrogen-cooled isopentane and used for immunohistochemistry, and the other part was fixed in B5 or Bouin's solution and embedded in paraffin for routine light microscopy.

For immunohistochemistry, serial frozen sections were cut and allowed to dry at room temperature. After fixation in acetone, sections were rehydrated and stained with a threestep indirect immunoperoxidase method using the following mouse monoclonal antibodies (mabs): R1G10 (diluted 1:500), an IgM antibody that blocks specifically human but not mouse IFN-y activity and does not react with human IFN-y itself12 nor with the extracellular 90kDa human IFNyR; GIR-208, an IgGl antibody directed to the IFN-yR35, used in a dilution of 1: 100 and obtained through the Cytokine Receptors Section of the 5th International Wor~shop and Conference on Human Leukocyte Differentiation Antigens. The same antibody is distributed by Genzyme (Boston, MA). MHC class II products or HLA-DR antigens were detected by mab L243 (Becton-Dickinson, Hialeah, FL; dilution 1:10); ICAM-1 was detected with mab 84H10 (Serotec, Oxford, UK; dilution 1:25). Secondary and tertiary antibodies consisted of peroxidase-conjugated rabbit anti-mouse and swine anti-rabbit immunoglobulins, both obtained from DAKO (Copenhagen, Denmark) and diluted 1:50 and 1:100 respectively in phosphate buffered saline (PBS), pH 7.2, containing 10% normal human AB+ serum to reduce unwanted background staining. The reaction-product was developed using aminoethylcarbazole and H 20 2. The sections were slightly counterstained with Harris' haematoxylin and mounted with glycerin jelly. Controls included the use of non-immune mouse ascites (Cappel Labs, Cochranville, FL) as well as irrelevant mcabs of similar immunoglobulin subclass specificity. These controls only yielded cells with endogenous peroxidase activity, but were otherwise negative.

References 1 Aiba, S, Tagami H (1984) HLA-DR antigen expression on the keratinocyte surface in dermatoses characterized by lymphocytic exocytosis (e.g. pityriasis rosea). Br J Dermatol 111: 285-294 2 Ando M, Kawashima T, Kobayashi H, Ohkawara A (1990) Immunohistological detection of proliferating cells in normal and psoriatic epidermis using Ki-67 monoclonal antibody. J Dermatol Sci 1: 441-446 3 Aubock J, Romani N, Grubauer G, Fritsch P (1986) HLA-DR expression on keratinocytes is a common feature of diseased skin. Br J Dermatoll14: 465-472 4 Baker BS, Powles AV, Valdimarsson H, Fry L (1988) An altered response by psoriatic keratinocytes to gamma interferon. Scand J Immunol 28: 735 -740 5 Barker JNWN, Allen MH, MacDonald DM (1990) Alterations induced in normal human skin by in-vivo interferon-gamma. Br J Dermatol 122: 451-458 6 Barker JNWN, Goodlad JR, Ross EL, Yu CC, Groves RW, MacDonald DM (1993) Increased epidermal cell proliferation in normal human skin in vivo following local administration of interferon-yo Am J Pathol142: 1091-1097 7 Barker JNWN, Karabin GD, Stoof TJ, Sarma VJ, Dixit VM, Nickoloff BJ (1991) Detection of interferon-gamma mRNA in psoriatic epidermis by polymerase chain reaction. J Dermatol Sci 2: 106-111 8 Basham TY, Nickoloff BJ, Merigan TC, Morhenn VB (1984) Recombinant gamma interferon induces HLA-DR expression on cultured human keratinocytes. J Invest Dermatol 83: 88-90 9 Bjerke JR, Matre R (1983) Demonstration of la-like antigens on T lymphocytes in lesions of psoriasis, lichen planus

534 . ].]. van den Oord et al. and discoid lupus erythematosus. Acta Derm Venereol (Stockh) 63: 103-107 10 Bos ]D, Hulsebosch H], Krieg SR, Bakker PM, Cormane RH (1983) Immunocompetent cells in psoriasis. Arch Dermatol Res 275: 181-189 11 Celada A (1988) The interferon gamma receptor. Lymphokine Res 7: 61- 73 12 Depla E, De Ley M (1990) Monoclonal antibodies against the human interferon-y receptor(s). Mol Immunol 27: 745-750 13 Depla E (1991) Immunologische karakterisering van het humaan interferon-y receptor-systeem via monoklonale en anti-idiotypische antilichamen. Academic Thesis, Katholieke Universiteit Leuven, Leuven 14 Dustin ML, Singer KH, Tuck DT, Springer TA (1988) Adhesion of T-Iymphoblasts to epidermal keratinocytes is regulated by interferon and is mediated by intercellular adhesion molecule 1 (ICAM-1).] Exp Med 167: 1323-1340 15 Fierlbeck G, Rasner G, Muler C (1990) Psoriasis induced at injection site by recombinant interferon-gamma. Arch Dermatol126: 351-355 16 Fleming MG, Howe SF, Candel AG (1988) Immunohistochemical localization of cytokines in nevi. Am ] Dermatopathol 1992, 14: 496-503 17 Fry L, Psoriasis. Br] Dermatol199: 445-461 18 Gottlieb AB, Lifshitz B, Fu SM, Staiano-Coico L, Wang CY, Carter DM (1986) Expression of HLA-DR molecules by keratinocytes and presence of Langerhans cells in the dermal infiltrate of active psoriatic plaques. ] Exp Med 164: 10131028 . 19 Gottlieb AB, Luster AD, Posnett DN, Carter DM (1988) Detection of a interferon-induced protein IP-10 in psoriatic pla~ues.] Exp Med 168: 941-948 2 Harrist T], Ruiter D], Mihm MC, Bhan AK (1983) Distribution of major histocompatibility antigens in normal skin. Br] Dermatol109: 623-628 21 Hershey GK, Schreiber RD (1989) Biosynthetic analysis of the human interferon-gamma receptor. Identification of Nlinked glycoxylation intermediates.] BioI Chern 264: 1198111988 22 Hibino Y, Kumar CS, Mariano TM (1992) Chimeric interferon-gamma receptors demonstrate that an accessory factor required for activity interacts with the extracellular domain.] Bioi Chern 267: 3741-3749 23 Ho V, Griffiths CEM, Gupta AK, McCuaig CC; Nickoloff B], Cooper KD, Voorhees 11 (1990) Intralesional cyclosporin A in the treatment of psoriasis: A clinical, immunologic and pharmacokinetic study.] Am Acad Dermatol22: 94 -1 00 24 lung V, Rashidbaigi A, Jones C, Tischfield ]A, Shows TB, Pestka S (1987) Human chromosomes 6 and 21 are required for sensitivity to human interferon gamma. Proc Natl Acad Sci USA 84: 4151-4155 25 Kaneko F, Suzuki M, Takiguchi Y, hoh N, Minagawa T (1990) Immunohistopathologic studies in the development of psoriatic lesion influenced by gamma-interferon and the producing cells. ] Dermatol Sci 1: 425 -434 26 Mao C, Merlin G, Ballotti R, Metzler M, Aguet M (1990) Rapid increase of the human IFN-gamma receptor phophorylation in response to human IFN-gamma and phor-

bol myristate acetate. Involvement of different serinelthreonine kinases. ] Immunol 145: 4257 -4264 27 McGregor ]M, Barker ]NWN, Allen MH, MacDonald DM (1991) Antigenic profile of human acrosyringium. Br] Dermatol125: 413-418 28 Mitra RS, NickoloffB] (1992) Epidermal growth factor and transforming growth factor-alpha decrease gamma interferon receptors and induction of intercellular adhesion molecule-I (ICAM-1) on cultured keratinocytes. ] Cell Physiol 150: 264-268 29 Morhenn VB, Abel EA, Mahrle G (1982) Expression of HLA-DR antigen in skin from patients with psoriasis.] Invest Dermatol 78: 165 -169 30 Murphy GF, Shephard RS, Harrist T], Bronstein BR, Bhan AK (1983) Ultrastructural documentation of HLADR antigen reactivity in normal human acrosyringial epithelium. ] Invest Dermatol 81: 181-183 31 Nickoloff B] (1987) Binding of 1251-gamma interferon to cultured keratinocytes. ] Invest Dermatol 89: 513-516 32 Nickoloff B] (1988) Role of gamma interferon in cutaneous trafficking of lymphocytes with emphasis on molecular and cellular events. Arch Dermatol124: 1835 -1843 33 Reitamo S, Anttilla HSI, Didierjean L, Saurat J-H (1990) Immunohistochemical identification of interleukin 1 \l and ~ in human eccrine sweat-gland apparatus. Br] Dermatoll22: 315-322 34 Scheynius A, Fransson], Johansson C, Hammar H, Baker B, Fry L, Valdimarsson H (1992) Expression of interferongamma receptors in normal and psoriatic skin. ] Invest Dermatol 98: 255-258 35 Sheehan KFC, Calderon], Schreiber RD (1988) Generation and characterization of monoclonal antibodies specific for the human IFN-g-reseptor.] Immunol140: 4231-4237 36 Smolle] (1985) HLA-DR-antigen bearing keratinocytes in various derma to logic disorders. Acta Derm Venereol (Stockh) 65: 9-15 37 Tank B, van Dijk G, van der Sluis ], van der Kamp AWM, Schonk D, Stolz E, Versnel M, van Joost T (1986) Detection of basal-cell determinant(s) in human psoriatic skin: a monoclonal antibody study. Arch Dermatol Res 279: 68-70 38 Terui T, Aiba S, Tanaka T, Tagami H (1987) HLA-DR antigen expression on keratinocytes in highly inflamed parts of fssoriatic lesions. Br] Dermatol116: 87-93 9 Valente G, Ozmen L, Novelli F, Geuna M, Palestro G, Forni G, Garotta G (1992) Distribution of interferon-y receptor in human tissues. Eur] Immunol22: 2403-2412 40 Volpes R, van den Oord 11, De Vos R, Depla E, De Ley M, Desmet V] (1991) Expression of interferon-y receptor in normal and pathological human liver tissue. ] Hepatol 12: 195-202 41 Watanabe S, Wagatsuma K, Ichikawa E, Takahashi H (1991) Abnormal distribution of epidermal protein antigens in gsoriatic epidermis. ] Dermatol 18: 143 -151 2 Weiss RA, Guillet GYA, Freedberg 1M, Farmer ER, Small EA, Weiss MM, Sun TT (1983) The use of monoclonal antibody to keratin in human epidermal disease. Alterations in immunohistochemical staining pattern. ] Invest Dermatol 81: 224-230

Received January 20, 1995 . Accepted April 10, 1995

Key words: Cytokine - Cytokine-receptor - Immunohistochemistry Dr. J. J. van den Oord, MD, PhD, Dept. Pathology II, Laboratory of Histo-and Cytochemistry, University Hospital St. Rafael, K. U. L., Minderbroedersstraat 12, B-3000 Leuven, Belgium