Cytokine regulation of major histocompatibility complex antigen expression by human oral and skin keratinocytes

Archs oral Biol. Vol. 41, No. 6, pp. 533-538, 1996 Copyright © 1996ElsevierScienceLtd. All rights reserved Printed in Great Britain PII: S0003-9969(96)00026-X 0003-9969/96$15.00+ 0.00

Pergamon

CYTOKINE REGULATION OF MAJOR HISTOCOMPATIBILITY COMPLEX ANTIGEN EXPRESSION BY H U M A N ORAL A N D SKIN KERATINOCYTES J. LI, ~ P. M. F A R T H I N G 2 and M. H. THORNHILL"* 'Clinical Academic Group of Oral Medicine and Dental Diagnostic Science, University Dental Hospital of Manchester, Higher Cambridge Street, Manchester M15 6FH, U.K. and 2Department of Oral Pathology, The London Hospital Medical College, Turner Street, London El 2AD, U.K. (Accepted 13 March 1996)

Summary--The expression, and cytokine modulation, of major histocompatibility complex (MHC) class 1 and class II molecules on oral and skin keratinocytes were compared in cell culture. Both cell types expressed class I, but not class II, constitutively. However, stimulation with interferon-?, but not interleukin-lc~, and -1/3, tumour necrosis factor-~ or lymphotoxin, induced increased expression of class 1 and de-novo expression of HLA-DR on both cell types. Oral keratinocytes differed from skin keratinocytes in that they exhibited greater sensitivity to interferon-./ stimulation and higher stimulated expression of both class I and HLA-DR. In addition, interferon-./ stimulated oral, but not skin, keratinocytes to express HLA-DP and -DQ. These observations suggest that, like skin keratinocytes, under certain conditions, oral keratinocytes may be able to act as antigen-presenting cells. This may be important in the initiation and progression of some immune-mediated mucocutaneous diseases. Moreover, differences in MHC expression may help to explain differences in the presentation of these diseases on the skin and oral mucosa. Copyright © 1996 Elsevier Science Ltd. Key words: keratinocyte, oral, skin, major histocompatibility complex, class I, class II, Interferon-¥.

INTRODUCTION

oral and skin lichen planus (Jontell et al., 1986; Hedberg and Hunter, 1987; Savage et al., 1987; The human MHC is encoded as a complex multiFarthing and Cruchley, 1989), psoriasis (Terui et gene family on chromosome 6 (Giles and Capra, al., 1987), allergic and irritant contact dermatitis 1985). HLA class I (HLA-A, -B, -C) alloantigens (Smolle, 1 9 8 5 ) and graft-versus-host disease are expressed on nearly all cells. Class II (HLA-DR, (Lampert et al., 1981). Indeed, Aub6ck et al. (1986) -DP, -DQ) alloantigens, on the other hand, are norhave reported 52 different skin disorders in which mally expressed only by dedicated antigen-presentHLA-DR expression is induced on keratinocytes. It ing cells. They play an important part in the is believed that the induction of MHC class II molgeneration and regulation of immune responses due ecules on keratinocytes endows them with the abilto their ability to bind and present foreign antigenic ity to present antigen (Czernielewski, 1985; peptides to CD4 ÷ T-lymphocytes (Benacerraf and Nickoloff et al., 1986; Ishii, 1987; van Lambalgen Germain, 1978; Unanue and Allen, 1987) and Wauben-Penris, 1989; Nickoloff et al., 1993a) Keratinocytes in normal skin and oral mucosa and enhances cellular immunity in the epidermis or express class I constitutively but do not express epithelium (Suitters and Lampert, 1982). These class II (Suitters and Lampert, 1982; Savage et al., properties may be important factors in the aetiology 1987; Messadi et al., 1988). However, HLA class II of mucocutaneous diseases. Moreover, differences in expression has been observed on keratinocytes in a MHC expression by skin and oral keratinocytes variety of diseases characterized by lymphocytic may help to determine the different presentation of infiltration of the epithelium or epidermis, such as diseases at these two sites. Several in-vitro studies have shown that IFN- 7 can induce the expression of HLA-DR but not -DP *To whom correspondence should be addressed at: Clinical Academic Group of Oral Medicine and Dental or -DQ on normal human skin keratinocytes in culDiagnostic Science, University Dental Hospital of ture (Basham et al., 1984, Basham et al., 1985; Manchester, Higher Cambridge Street, Manchester Barker et al., 1988; Griffiths et al., 1989b; Barker et M15 6FH, U.K. Tel.: 0161-275-6640; Fax: 0161-275- al., 1990). In addition, IFN-7 induced HLA-DR 6840. and -DP, but not -DQ, expression on the KB oral Abbreviations: IFN-./, interferon-'/; IL-I, interleukin-l; MHC, major histocompatibility complex; TNF, tumour squamous carcinoma cell line (Mutlu et al., 1991) and class II expression on normal keratinocytes and necrosis factor. 533

534

J. Li et al.

oral squamous carcinoma cell lines derived from Sprague-Dawley rats (Crane et al., 1988). However, little is known about the expression of MHC class I and class II molecules by normal human oral keratinocytes in culture or the conditions necessary for this expression. Our aim now was to compare the expression and cytokine modulation of MHC class I and class II molecules on normal human oral and skin keratinocytes in culture.

MATERIALS AND METHODS

Antibodies

IgG1 mouse antihuman monoclonal antibodies against H L A class I and H L A - D R were produced using the BB7.5 (Brodsky and Parham, 1982) and L243 (Lampson and Levy, 1980) hybridoma cell lines, respectively, purchased from the American Type Culture Collection (Rockville, MD, U.S.A.) and used as neat hybridoma supernatants. Mouse antihuman monoclonal antibody B7/21 directed against H L A - D P and mouse antihuman monoclonal antibody SK10 directed against HLA-DQ were obtained from Becton Dickinson (Mountain View, CA, U.S.A.) and used at 50x dilution in culture medium (Watson et al., 1983). Biotinylated rabbit antimouse Ig was obtained from Dako (High Wycombe, U.K.). Provisional ELISA studies confirmed that each antibody was being used at greater than four times the saturating dilution on optimally stimulated cells. Cytokines

Recombinant human (rh) IFN-7 with a specific activity of 2 × 104 IU/~tg was obtained from Biogen SA (Geneva, Switzerland); Rh TNF-~ was a gift from Dr B. A. Beutler, University of Texas Health Sciences Centre, Dallas, Texas. IL-I~ and -1[~ and lymphotoxin (also called TNF~I3 ) were obtained from the National Institute for Biological Standards and Control (Potters Bar, U.K.). All cytokines were used at the following concentrations: 1, 10, 100, 250 and 1000 U/ml. Cell culture

Keratinocytes were prepared from samples of normal human skin or oral mucosa using a modification of the method described by Rheinwald and Green (1975). Skin keratinoeytes were obtained from three females and one male (age range 32-56 yr) using excess skin removed during breast reduction or apronectomy procedures. Oral keratinocytes were prepared from normal oral mucosa removed during frenectomy, gingival surgery or from surgical-flap margins from one male and three females (age range 28-72 yr). In brief, samples of oral mucosa and skin were treated in 0.25% trypsin in Tris-saline (ICRF, London, U.K.) at 37°C for 2

h. The epithelium was then removed and the keratinocytes dispersed with forceps to obtain a single-cell suspension. The cells were placed on to feeder layers of mitomycin C-treated (Sigma, Poole, U.K.) (10 tag/ml, 2 h, 37°C) mouse 3T3 fibroblasts and cultured at 37°C in a humidified atmosphere containing 5% CO2 in keratinocyte growth medium consisting of three parts Dulbecco's modified Eagle's medium (Gibco, Paisley, U.K.) and one part of Hams FI2 medium (Sigma) supplemented with 10% foetal bovine serum (Globepharm, Esher, U.K.), 10 ng/ml epidermal cell growth factor, 0.4 lag/ml hydrocortisone, 5 ~tg/ml transferrin, 5 ~tg/ml insulin, 1.8 × 10 - 4 M adenine, 1 × 10 - 10 M cholera toxin, 50 U/ml penicillin, 50 rng/ml streptomycin (all from Sigma) and 2.5 ~tg/ml fungizone (Gibco). All the experiments were performed on third- to fifth-passage keratinocytes. Cell-based E L I S A for H L A class I and class H expression

Keratinocytes were seeded at 4 × l04 cells/well into 96-well flat-bottomed plates (Corning, New York, U.S.A.) in 0.2 ml of keratinocyte growth medium. After overnight incubation, the medium was replaced with 0.2 ml of Dulbecco's modified Eagle's medium plus 10% foetal bovine serum and the cells cultured for various times with or without cytokines. The cultures were then washed with Dulbecco's modified Eagle's medium and incubated with 50 lal/well of mouse antihuman class I or HLA-DR, -DP or -DQ monoclonal antibodies at 37°C for 30 min. The keratinocyte monolayers were washed and fixed at 4°C for 8 min with periodatelysine-paraformaldehyde fixative consisting of 2% paraformaldehyde (BDH, Poole, U.K.), 0.075% Llysine HCI (Sigma) and 2.1 mg/ml sodium periodate (BDH) in an aqueous solution. The fixed monolayers were washed and incubated for at least 1 h with blocking solution containing 1% bovine serum albumin (Sigma), 100 mM glycine (Sigma) and 0.05% sodium azide (Sigma) in Hank's balanced salt solution before performing the ELISA. The ELISA was done at room temperature. After aspiration of blocking solution, cells were washed three times with 1% bovine serum albumin in phosphate buffered saline and incubated with 50 ~tl/well of biotinylated rabbit antimouse Ig for 30 min. After a further three washes, the cells were incubated for 30 min with 50 lal/well of avidin-biotinhorseradish peroxidase complex (Dako). After three more washes, 150 lal of the enzyme substrate solution, containing 0.05 mg% O-phenylenediamine (Sigma) and 1 p.l/ml of 30% H202 (BDH) in phosphate-citrate buffer, pH 5.0, were added to each well and the colour development stopped by adding 50 lal/well of 1 M sulphuric acid. The optical density of each well was read at 492 nm in a Titertek ELISA plate reader.

535

Regulation of keratinocyte MHC expression

Statistical analysis Student's t-test was used for the statistical analysis. All experiments were repeated at least three times, and each data-point represents the mean + SD of triplicate wells.

RESULTS

H L A - D P and -DQ expression

HLA class I expression HLA class I was expressed constitutively on both oral and skin keratinocytes. Stimulation with IFN-? increased class I expression on both cell types. Optimal expression of class I occurred after stimulation with 100 U/ml IFN- 7 on oral keratinocytes and 25-1000 U/ml on skin keratinocytes (Fig. 1). The kinetics of class I expression following IFN-7 stimulation were similar on oral and skin keratinocytes, with peak expression occurring after 72 h, the latest time-point studied (Fig. 2). IFN- 7 induced a significantly greater increase in class I expression on oral keratinocytes than it did on skin keratinocytes under the same conditions (p < 0.05) (Figs 1 and 2). Stimulation with IL-I~, IL-II3, T N F - a or lymphotoxin had no effect on class I expression by oral or skin keratinocytes at any of the concentrations tested (data not shown).

H L A - D R expression Constitutive HLA-DR expression was not detected on oral or skin keratinocytes. However, stimulation with IFN-y induced HLA-DR expression on both cell types, with optimal induction occurring with 250-1000 U/ml IFN-7 (Fig. 1). The kinetics of HLA-DR expression following IFN-7 stimulation were similar on oral and skin keratinocytes with peak expression occurring after 72 h, the latest time-point studied (Fig. 2).

0.5

"• t",l

The level of HLA-DR expression induced on oral keratinocytes by IFN-y stimulation was significantly greater than the level induced on skin keratinocytes under the same conditions (p < 0.05) (Figs 1 and 2). Stimulation with IL-I~, IL-113, TNF-c~ or lymphotoxin had no effect on HLA-DR expression by oral or skin keratinocytes at any of the concentrations tested (data not shown).

Oral

There was no constitutive expression of HLA-DP or -DQ on skin or oral keratinocytes. Moreover, as previously demonstrated (Basham et al., 1984; Kerr et al., 1990), IFN-y was unable to induce HLA-DP or -DQ expression on skin keratinocytes. In contrast, IFN- 7 induced significant levels of HLA-DP and -DQ expression on oral keratinocytes (Fig. 3). Stimulation with IL-lm IL-113, TNF-~ or lymphotoxin had no effect on HLA-DP or HLA-DQ expression by oral or skin keratinocytes at any of the concentrations tested (data not shown).

DISCUSSION We demonstrate that, in culture, normal human oral keratinocytes, like skin keratinocytes, do not express MHC class II antigens constitutively, but do express class I. This correlates with the observed expression of MHC antigens in biopsies of normal skin and oral mucosa (Suitters and Lampert, 1982; Messadi et al., 1988). Stimulation of keratinocytes with IFN-7, but none of the other cytokines tested, resulted in a large increase in expression of class I. Potentially, this could enhance the susceptibility of virally infected keratinocytes to cytotoxic T-cell lysis and increase the immunogenicity of malignant cells expressing altered self antigens. 0.5-

0.4

0.4-

0.3

0.3-

Skin

CLASS 1 ---o--- DR

/ -

0.2

C~ 0 o.1 ......

-0.1

0.1-

.K

"~'¢~'

o

.0 ~

o

I

I

1

lO IFN-~/u/ml

I

lOO

I

lOOO

. . . . . . ~, . . . . . . ~ . . . . . . .

-0.1

-g.'

0

I

/

1

10

g i

100

i

1000

IFN-~/u/ml

Fig. 1. HLA class I and -DR expression on keratinocytes stimulated with IFN-y. Oral and skin keratinocytes were stimulated with different concentrations of IFN-7 for 48 h and the expression of class I and HLA-DR was measured using a cell-based ELISA. The experiments shown are representatitive of three similar experiments, and each point represents the mean + SD of triplicate wells. OD, optical density.

536

J. Li et al. 0,6-

O~ v

0.a-

Oral

Skin

0.5-

0.5-

0.4-

0.4//

0.3-

0.3- .

~

~

0.2-

0.2-

© II

0.1-

..0

----o--- CLASS1

,

---o---

-0.1

I

I

20

40

0.1O,

DR

"0.1

60

80

Time (h)

0

I

I

I

I

20

40

60

80

Time (h)

Fig. 2. Kinetics of HLA class I and -DR expression on keratinocytes stimulated by IFN-y. Oral and skin keratinocytes were stimulated with 1000 U/ml of IFN-y for different periods of time (0, 4, 8, 24, 48 and 72 h) and the expression of HLA class I and -DR measured using a cell-based ELISA. The experiments shown are representatitive of three similar experiments, and each point represents the mean + SD of triplicate wells. OD, optical density. MHC class II molecules are not normally expressed on skin or oral keratinocytes but are observed in a variety of mucocutaneous disorders characterized by the presence of lymphocytes (Lampert et al., 1981; Smolle, 1985; Aub6ck et al., 1986; Jontell et al., 1986; Hedberg and Hunter, 1987; Terui et al., 1987; Farthing and Cruchley, 1989). Production of IFN-y by these lymphocytes is likely to be responsible for the induction of MHC class II on the overlying keratinocytes and may enable them to present antigen to lymphocytes (Czernielewski, 1985; Nickoloff et al., 1986, 1993a). Indeed, it has been shown that IFN-y-treated skin keratinocytes are able to present bacterial antigens 0.14 0.12 0.10 0.08 c'q

0.06-

,•/"/•'""

v

O

0.040.02 -

---~--'

0; -0.02

r~¢

0

0----

I

I

1

10

DpA DQ

I

I

100

1000

IFN-~/u/ml Fig. 3. IFN-y-induced HLA-DP and -DQ expression on oral keratinocytes. Oral keratinocytes were stimulated with different concentrations of IFN-y for 48 h and the expression of HLA-DP and -DQ measured using a cellbased ELISA. The experiments shown are representatitive of three similar experiments, and each point represents the mean ___SD of triplicate wells. OD, optical density.

to T-cells and activate them in culture (Nickoloff et al., 1993a, b). The ability to present extrinsic antigens or altered self antigens to T-cells could make keratinocytes active initiators of immune responses or the targets of cell-mediated damage. These properties may be important in the aetiology of immunoinflammatory conditions such as lichen planus, psoriasis, allergic and irritant contact dermatitis/stomatitis and graftversus-host disease. Although we found similarities between skin and oral keratinocytes, there were also important differences. Firstly, oral keratinocytes exhibited greater sensitivity to IFN-y stimulation than skin keratinocytes: 1 U/ml IFN-y stimulated the same level of class I expression on oral keratinocytes as 1000 U/ ml on skin keratinocytes and 100 U/ml IFN-y induced a higher level of HLA-DR expression on oral keratinocytes than 1000 U/ml on skin keratinocytes. Moreover, maximal expression of class I and HLA-DR was significantly greater on oral than skin keratinocytes. In vivo, this may result in oral keratinocytes being more immunoreactive than skin keratinocytes. Secondly, oral keratinocytes were induced to express HLA-DP and -DQ by IFN-y stimulation, whereas, in common with other researchers (Basham et al., 1984; Messadi et al., 1988; Griffiths et al., 1989a; Kerr et al., 1990), we found no evidence of HLA-DP or -DQ expression on skin keratinocytes. Little is known about the distribution and functional significance of HLA-DP and -DQ expression. In vivo, they are not detected on normal keratinocytes; however, infrequent HLA-DQ expression has been reported in cutaneous T-cell lymphoma, lichen planus (Volc-Platzer et al., 1987), Borrelia-induced skin disease (Tjernlund et al., 1986), oral candidosis (Jontell et al., 1986) and allergic and irritant contact

Regulation of keratinocyte MHC expression dermatitis ( G a w k r o d g e r et al., 1987). Less is k n o w n a b o u t H L A - D P expression. H L A - D P a n d - D Q are believed to function in a similar fashion to H L A - D R . However, some lymphocyte clones m a y only recognize antigen in the context o f H L A - D P or -DQ. F o r example, some C a n d i d a albicans antigens a p p e a r to be H L A - D Q restricted ( G o n w a et al., 1983). The ability o f oral keratinocytes to express H L A - D P a n d - D Q m a y therefore enable t h e m to interact with certain clones o f lymphocytes t h a t are u n a b l e to interact with skin keratinocytes. Thus, oral keratinocytes are able to express M H C class II molecules following I F N - 7 stimulation and, like skin keratinocytes, this m a y e n d o w t h e m with the ability to present antigen a n d initiate i m m u n e responses. However, they also exhibit differences from skin keratinocytes t h a t m a y m a k e t h e m more i m m u n o g e n i c a n d different in their i m m u n e responsiveness.

acknowledge the support provided by The South Essex Medical Education and Research Trust, and the Research Advisory Committee of the Special Trustees of The Royal London Hospital. We would also like to acknowledge the technical help from Mr S. Cannon of the Department of Oral Pathology, The London Hospital Medical College. Acknowledgements--We

REFERENCES

Aubrck J., Romani N., Grubauer G. and Fritsch P. (1986) HLA-DR expression on keratinocytes is a common feature of diseased skin. Br. J. Dermatol. 114, 465-472. Barker J. N., Sarma V., Mitra R. S., Dixit V. M. and Nickoloff B. J. (1990) Marked synergism between tumor necrosis factor-alpha and interferon-gamma in regulation of keratinocyte-derived adhesion molecules and chemotactic factors. J. elin. Invest. 85, 605-608. Barker J. N. W. N., Navsaria H. A., Leigh I. M. and Macdonald D. M. (1988) Gamma-interferon induced human keratinocyte HLA-DR synthesis: the role of dermal activated T lymphocytes. Br. J. Dermatol. 119, 567572. Basham T. Y., Nickoloff B. J., Merigan T. C. and Morhenn V. B. (1984) Recombinant gamma interferon induces HLA-DR expression on cultured human keratinocytes. J. invest. Dermatol. 83, 88-90. Basham T. Y., Nickoloff B. J., Merigan T. C. and Morhenn V. B. (1985) Recombinant gamma interferon differentially regulates class II antigen expression and biosynthesis on cultured normal human keratinocytes. J. Interferon Res. 5, 23-27. Benacerraf B. and Germain R. N. (1978) The immune response genes of of the major histocompatibility complex. lmmunol. Rev. 38, 70-119. Brodsky F. M. and Parham P. (1982) Monomorphic ant° HLA-A, B, C monoclonal antibodies detecting molecular subunits and combinatorial determinants. J. lmmunol. 128, 129-135. Crane I. J., Rice S. Q., Luker J., de Gay L., Scully C. and Prime S. S. (1988) The expression of MHC antigens on cultured oral keratinocytes and relationship to malignancy. Br. J. exp. Pathol. 69, 749-758. Czernielewski J. M. (1985) Mixed skin cell-lymphocyte culture reaction (MSLR) as a model for the study of

537

lymphoepidermal interactions. Br. J. Dermatol. 113 Suppl. 28, 17-23. Farthing P. M. and Cruchley A. T. (1989) Expression of MHC class II antigens (HLA DR, DP and DQ) by keratinocytes in oral lichen planus. J. oral pathol. Med. 18, 305-309. Gawkrodger D. J., Carr M. M., McVittie E., Guy K. and Hunter J. A. (1987) Keratinocyte expression of MHC class II antigens in allergic sensitization and challenge reactions and in irritant contact dermatitis. J. invest. Dermatol. 88, 11-16. Giles R. C. and Capra J. D. (1985) Structure, function, and genetics of human class II molecules. Adv. Immunol. 37, 1-71. Gonwa T. A., Picker L. J., Raft H. V., Goyert S. M., Silver J. and Stobo J. D. (1983) Antigen-presenting capabilities of human monocytes correlates with their expression of HLA-DS, an Ia determinant distinct from HLA-DR. J. Immunol. 130, 706-711. Griffiths C. E., Voorhees J. J. and Nickoloff B. J. (1989) Characterization of intercellular adhesion molecule-I and HLA- DR expression in normal and inflamed skin: modulation by recombinant gamma interferon and tumor necrosis factor. J. Am. Acad. Dermatol. 20, 617629. Griffiths C. E. M., Voorhees J. J. and Nickoloff B. J. (1989) Gamma interferon induces different keratinocyte cellular patterns of expression of HLA-DR and DQ and intercellular adhesion molecule-1 (ICAM-1) antigers. Br. J. Dermatol. 120, 1-7. Hedberg N. M. and Hunter N. (1987) The expression of HLA-DR on keratinocytes in oral lichen planus. J. oral Pathol. 16, 31-35. Ishii T. (1987) Immunohistochemical demonstration of T cell subsets and accessory cells in oral lichen planus. J. oral Pathol. 16, 356-361. Jontell M., Scheynius A., Ohman S. C. and Magnusson B. (1986) Expression of Class II transplantation antigens by epithelial cells in oral candidosis, oral lichen planus and gingivitis. J. oral Pathol. 15, 484-488. Kerr L. A. P., Navsaria H. A., Barker J. N. W. N., Sakkas L. I., Leigh I. M., MacDonald D. M. and Welsh K. I. (1990) Interferon-gamma activates co-ordinate transcription of HLA-DR, DQ, and DP genes in cultured keratinocytes and requires de novo protein synthesis. J. invest. Dermatol. 95, 653 656. Lampert 1. A., Suitters A. J. and Chisholm P. M. (1981) Expression of Ia antigen on epidermal keratinocytes in graft-versus-host disease. Nature 293, 149-150. Lampson L. A. and Levy R. (1980) Two populations of Ia-like molecules on a human B cell line. J. Immunol. 125, 293-299. Messadi D. V., Pober J. S. and Murphy G. F. (1988) Effects of recombinant gamma-interferon on HLA-DR and DQ expression by skin cells in short-term organ culture. Lab. Invest. 58, 61 67. Mutlu S., Scully C. and Prime S. S. (1991) Effect of IFNgamma on the expression of MHC class I and class II antigens in a human malignant oral epithelial cell line. J. oral. pathol. Med. 20, 218-221. Nickoloff B. J., Basham T. Y., Merigan T. C., Torseth J. W. and Morhenn V. B. (1986) Human keratinocytelymphocyte reactions in vitro. J. invest. Dermatol. 87, 11-18. Nickoloff B. J., Mitra R. S., Green J., Zheng X. G., Shimizu Y., Thompson G. and Turka L. A. (1993) Accessory cell function of keratinocytes for superantigens. Dependence on lymphocyte function-associated antigen-l/intercellular adhesion molecule-1 interaction. J. Immunol. 150, 2148-2159. Nickoloff B. J., Mitra R. S., Lee K., Turka L. A., Green J., Thompson C. and Shimizu Y. (1993) Discordant expression of CD28 ligands, BB-I, and B7 on keratino-

538

J. Li et al.

cytes in vitro and psoriatic cells in vivo. Am. J. Pathol. 142, 1029-1040. Rheinwald J. G. and Green H. (1975) Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 6, 331344. Savage N. W., Walsh U J. and Seymour G. H. (1987) Expression of class I and class II major histocompatibility complex antigens on oral mucosal epithelium. J. oral Pathol. 16, 153-157. Smolle J. (1985) HLA-DR-antigen bearing keratinocytes in various dermatologic disorders. Acta Dermat. Venereol. 65, 9-13. Suitters A. J. and Lampert I. A. (1982) Expression of Ia antigen on epidermal keratinocytes is a consequence of cellular immunity. Br. J. exp. Path. 63, 207-213. Terui T., Aiba S., Kato T., Tanaka T. and Tagami H. (1987) HLA-DR antigen expression on keratinocytes in

highly inflamed parts of psoriatic lesions. Br. J. Dermat. 116, 87-93. Tjernlund U., Scheynius A., Asbrink E. and Hovmark A. (1986) Expression of HLA-DQ antigens on keratinocytes in Borrelia spirochete-induced skin lesions. Scand. J. Immunol. 23, 383-388. Unanue E. R. and Allen P. M. (1987) The basis for the immunoregulatory role of macrophages and other accessory cells. Science 236, 551-557. van Lambalgen R. and Wauben-Penris P. J. J. (1989) Concepts in immunodermatology. Acta Dermat. Venereol. 69, 56-60. Volc-Platzer B., Groh V. and Wolff K. (1987) Differential expression of class II alloantigens by keratinocytes in disease. J. invest. Dermatol. 89, 64-68. Watson A., DeMars R., Trowbridge I. S. and Bach F. H. (1983) Detection of a novel human class II HLA antigen. Nature 3114, 358-361.