t cell subsets and langerhans cells in lichen planus: in situ characterization using monoclonal...

7
British Journal of Dermatology (1981) 105, 617-622. Clinical and Laboratory Investigations T cell subsets and Langerhans cells in lichen planus: in situ characterization using monoclonal antibodies A.K.BHAN, T.J.HARRIST, G.F.MURPHY AND M.C.MIHM, JR Dermatopathology and Immunopathology Units, Departments of Pathology and Dermatology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, U.S.A. Accepted for publication 26 June 1981 SUMMARY Skin biopsies from four patients with lichen planus were studied using monoclonal antibodies directed against T lymphocytes. Anti-Ti and anti-T3 antibodies, which react with all peripheral T cells, stained most cells in the dermal infiltrates. The majority of infiltrating cells also stained with anti-T4 and anti-T4b antibodies, which react with helper/inducer cells, whereas a minority of cells stained with anti-T8 antibody, which reacts with cytotoxic/suppres- sor cells. Surface IgM was not identified on any infiltrating cells, providing evidence against B cell participation. Intraepidermal and dermal cells with long cytoplasmic extensions stained with anti-T6 antibody in all cases, defining them as Langerhans cells or their precursors. T6-positive cells were seen in greater number than in normal control epidermis and dermis. The results indicate that well-developed lesions of lichen planus are characterized by an influx of helper/inducer T lymphocytes and increased numbers of Langerhans cells. These observations support the contention that cellular immunity is important in the pathogenesis of this disorder. The pathogenesis of lichen planus is unknown. While the initial event in lesion development is obscure, considerable evidence implicates both humoral and cell-mediated immunity in the formation of lesions. Globular deposits (Civatte bodies) containing immunoglobulins, princi- pally IgM, complement and fibrin are present in the dermis in 85-100% of cases. Granular deposits of immunoglobulin and complement have also been observed at the dermo-epidermal junction and in foci of inflammatory cells (Harrist & Mihm, 1979). Because immunoglobulin and complement deposition in lichen planus is limited primarily to Civatte bodies, which are apoptotic keratinocytes (Weedon, 1974), it appears that humoral events are most likely to be secondary. The clinical and histological findings in certain cases of graft-vs-host disease are virtually identical to those of lichen planus (Hashimoto, 1976) implying a role for cellular immunity in Correspondence: Atul K.Bhan, M.D., Department of Pathology, Massachusens General Hospital, Boston, Maeiarh..setts 021 I4, U.S.A. 0007-0963/81/1200-0617802.00 © 1981 British Association of Dermatologists 617

Upload: independent

Post on 26-Nov-2023

0 views

Category:

Documents


0 download

TRANSCRIPT

British Journal of Dermatology (1981) 105, 617-622.

Clinical and Laboratory Investigations

T cell subsets and Langerhans cells in lichen planus:in situ characterization using monoclonal antibodies

A.K.BHAN, T.J.HARRIST, G.F.MURPHY AND M.C.MIHM, JRDermatopathology and Immunopathology Units, Departments of Pathology and Dermatology, Massachusetts General

Hospital and Harvard Medical School, Boston, MA, U.S.A.

Accepted for publication 26 June 1981

SUMMARY

Skin biopsies from four patients with lichen planus were studied using monoclonal antibodiesdirected against T lymphocytes. Anti-Ti and anti-T3 antibodies, which react with allperipheral T cells, stained most cells in the dermal infiltrates. The majority of infiltrating cellsalso stained with anti-T4 and anti-T4b antibodies, which react with helper/inducer cells,whereas a minority of cells stained with anti-T8 antibody, which reacts with cytotoxic/suppres-sor cells. Surface IgM was not identified on any infiltrating cells, providing evidence against Bcell participation. Intraepidermal and dermal cells with long cytoplasmic extensions stainedwith anti-T6 antibody in all cases, defining them as Langerhans cells or their precursors.T6-positive cells were seen in greater number than in normal control epidermis and dermis. Theresults indicate that well-developed lesions of lichen planus are characterized by an influx ofhelper/inducer T lymphocytes and increased numbers of Langerhans cells. These observationssupport the contention that cellular immunity is important in the pathogenesis of this disorder.

The pathogenesis of lichen planus is unknown. While the initial event in lesion development isobscure, considerable evidence implicates both humoral and cell-mediated immunity in theformation of lesions. Globular deposits (Civatte bodies) containing immunoglobulins, princi-pally IgM, complement and fibrin are present in the dermis in 85-100% of cases. Granulardeposits of immunoglobulin and complement have also been observed at the dermo-epidermaljunction and in foci of inflammatory cells (Harrist & Mihm, 1979). Because immunoglobulinand complement deposition in lichen planus is limited primarily to Civatte bodies, which areapoptotic keratinocytes (Weedon, 1974), it appears that humoral events are most likely to besecondary.

The clinical and histological findings in certain cases of graft-vs-host disease are virtuallyidentical to those of lichen planus (Hashimoto, 1976) implying a role for cellular immunity in

Correspondence: Atul K.Bhan, M.D., Department of Pathology, Massachusens General Hospital, Boston,Maeiarh..setts 021 I4, U.S.A.

0007-0963/81/1200-0617802.00 © 1981 British Association of Dermatologists

617

6i8 A.K.Bhanetal

lichen planus. Indeed, investigators, using a variety of techniques, have shovi'n the lichenoidinfiltrate in lichen planus to be composed primarily of T lymphocytes (Walker, 1976; Alario etai, 1978; Bjerke & Krogh, 1978; MacDonald et ai, 1978; Dockrell & Greenspan, 1979). Todefine further the T cell infiltrate and to characterize Langerhans cell participation in lichenplanus, we examined the cell populations in well-developed lesions, with monoclonal antibodiesdirected against T lymphocytes as well as Langerhans cell surface antigens (Reinherz &Schlossman, 1980; Murphy et ai, 1981; Fithian et ah, 1981).

MATERIALS AND METHODS

Processing of tissueA 4 mm punch biopsy specimen was taken from a single lesion of lichen planus in four patients.For controls, a single 4 mm punch biopsy of normal lateral upper ann skin of two healthy adultmales was performed. The tissue was immediately bisected, with one half placed in 10%buffered formalin for routine histological examination and the other half frozen in OCTCompound (Ames Co., Div. of Miles Laboratory, Inc., Elkhart, Ind.) and stored at - 7o''C.

Monoclonal antibodiesA series of seven monoclonal antibodies reactive with thymocyte and/or peripheral Tlymphocyte surface antigens was used (Reinherz et ai, I979a,b,c; Bhan et ai, 1980; Reinherz &Schlossman, 1980; Reinherz et ai, 1980; Poppema et ai, 1981). Anti-Ti antibody and anti-T3antibody react with ioo% of peripheral E-rosette positive T cells and approximately lo",, ofthymocytes. Anti-T6 antibody reacts with 70",, of thymocytes but not with peripheral T cells.This antibody has recently been shown to react also with Langerhans cells in the skin (Murphyet al,y 1981; Fitbian et ai, 1981). Anti-T4 antibody, which detects the helper/inducer T cellsubset, reacts with 60",, of peripheral T ceils, while anti-T8 antibody, which detectssuppressor/cytotoxic T cells, reacts with 20-30'̂ ',, of peripheral T cells. Another monoclonalantibody termed anti-T4b (clone SFCI-12T4DU, provided by Dr Ellis Reinherz, SidneyFarber Cancer Institute, Boston, Massachusetts), which shows reactivity similar to that ofanti-T4 antibody in both cell suspensions and tissue sections (Reinherz & Bhan, unpublishedobservations), was used as well.

In addition, a monoclonal antibody (anti-Ii) reactive with a non-polymorphic region ofhuman Ia-Iike antigens (Reinherz et ai, I979d) and a monoclonal antibody reactive with humanIgM (provided by Dr L. Nadler, Sidney Farber Cancer Institute, Boston, Massachusetts), wereused.

Immunoperoxidase studiesFour micron-thick frozen sections were air-dried, fixed in acetone for 5 min and stained by afour-step peroxidase-anti-peroxidase (PAP) method (Bhan et ai, 1980; Poppema et ai, 1981).Sections were first incubated with a i: 500 dilution of mouse ascites containing one of themonoclonal antibodies for 60 min at room temperature followed by 30 min incubations withrabbit anti-mouse Ig (Dakopatts A/S, Copenhagen, Denmark), swine anti-rabbit Ig, and lastlywith peroxidase-rabbit-antiperoxidase reagent (Dakopatts A/S, Copenhagen, Denmark).Between each incubation, sections were washed in three changes of phosphate-buffered saline,pH 74.

Staining was achieved by incubation of sections in an acetate buffer solution (pH 50) thatcontained 3-amino-9-ethylcarbazol (Aldrich Chemical Co., Inc., Milwaukee, Wis.), dimethyl-

T cell subsets and Langerhans cells 619formamide and hydrogen peroxide. The sections were washed in acetate buffer and mounted inElvanol (E.L. Dupont de Nemours & Co., Wilmington, Del.).

RESULTS

Light microscopic and direct immunofiuorescence examination of the tissue confirmed theclinical impression of lichen planus. Each of the four specimens revealed the characteristichistological changes oflichen planus and direct immunofluorcscencc revealed moderate to largenumbers of colloid bodies which fiuoresced after application of anti-human IgM, IgG, C3 andfibrin. Granular immune reactant deposition was not identified at the dermo-epidermaljunction.

The vast majority of infiltrating cells stained with anti-Ti, anti-T3, anti-T4 and anti-T4bantibodies, whereas only a minority of cells reacted with anti-T8 antibody (Figs i and 2). Thestaining achieved with anti-Ti, anti-T3 and anti-T4 antibodies was weak, whereas intensestaining was seen with anti-T4b (Fig. i). In one case, significant epidermal alteration associatedwith a slight cellular infiltrate and papillary dermal fibrosis suggested that the lesion was in aphase of resolution. T8 ^ cells formed an appreciable proportion of the T cell population.

There was no staining of cells with antibody to IgM, indicating an absence of IgM * B cells inthe infiltrate.

FIGURE. I. Frozen tissue section of a lesion of lichen planus. A large number of T 4 * cells are observed inthe papillary dermis, some in close coniaci with the epidermis (E) ( x 256).

620 A.K.Bhan et aL

•i

•S '

FIGURE 2. Scattered TS"^ cells arc present in the papillary dermis and adjacent to the epidermis (E) in alesion of lichen ptanus ( x 256).

FiGtJRe 3. A lesion of lichen planus shows a large number of dendritic cells (Langerhans cells) stained withanti-T6 antibody in the epidermis. Similar cells are present in the papillary dermis (arrows) ( x 236).

T cell subsets and Langerhans cells 621Throughout the epidermis were large dendritic cells which stained with anti-T6 monoclonal

antibody (Fig. 3). Morphologically similar, less numerous cells also stained with anti-Il (anti-la)antibody. The T6^ cells were considered to be Langerhans cells. In the dermal infiltrates,similar, although fewer, dendritic cells were also T6'* .̂ A small number of other mononuclearcells in the infiltrate and endothelial cells also stained with anti-la antibody.

In the two control specimens, only rare dermal T8 ' cells were identified. Within theepidermis, T6+ dendritic cells were observed; however, they were much fewer in number thanthose seen in the lesions of lichen planus.

DISCUSSION

A series of monoclonal antibodies were employed to study the nature of the cellular infiltrate inthe frozen tissue sections of lichen planus by the immunoperoxidase technique. A majority ofthe cells in the infiltrate reacted with anti-Ti, anti-T3, and anti-T4 antibodies. The presentstudy confirms the T cell nature of the infiltrate in lichen planus, as suggested by previousstudies employing E-rosetting techniques or heteroantisera against T cells (Alario et al., 1978;MacDonald et al., 1978). More importantly, most of the T cells in the infiltrate are of thehelper/inducer subset. In contrast, only a small fraction of cells in the infiltrate stained withanti-T8, which reacts with suppressor/cytotoxic T cells. In one presumably resolving lesion,T8-positive cells formed an appreciable proportion of the cellular infiltrate indicating thatdynamic alterations of the infiltrate occur during the course of these lesions.

The absence of plasma cells in the infiltrate and lack of staining for B-cells in the present studyindicate that the immunoglobulins deposited in lichen planus lesions are derived from plasmaproteins and are not locally produced.

As compared to normal skin, increased numbers of large dendritic cells in the epidermis andin the dermis stained with anti-T6 antibody and to a lesser extent, with anti-la antibody.Anti-T6 antibody reacts with a majority of cortical thymocytes but not with mature peripheralT cells (Reinherz & Schlossman, 1980; Bhan et al., 1980). In a recent study (Poppema et aL,1981), large cells staining with anti-T6 antibody were found in the paracortex of certain lymphnodes, including a case of dermatopathic lymphadenitis. In addition, T6'^ cells in the normalepidermis appear to be Langerhans cells, as shown by their dendritic nature and the presence ofla-antigens. More recently, employing immunoperoxidase electron microscopy, we havedemonstrated Birbeck granules in the cytoplasm of T6^ cells in the epidermis (Murphy etal.,1981), establishing T6 antigen as a new immunological marker for epidermal Langerhans cells.Additional evidence of the association of T6 antigen with Langerhans cells using double-labelling techniques has recently been reported (Fithian et al.i 1981). Dermal dendritic T6'''cells were also observed in lichen planus. These cells may represent either so-called'indeterminate' (Langerhans precursor) cells or dermal Langerhans cells.

Langerhans cells, which have some characteristics of mononuclear phagocytes, are theantigen-presenting cells in allergic contact reactions (Thorbecke, Silberberg-Sinakin & Flotte,1980). The presence of a largenumber of Langerhans cells associated with numerous T4^ cellsadjacent to the epidermis as well as the absence of B cells, strongly suggests that cell-mediatedimmunity is important in the pathogenesis of lichen planus. The development of lichenplanus-like lesions in some cases of chronic graft-vs-host disease would also support thiscontention (Saurat & Gluckman, 1977).

In lichen planus, the basal keratinocytes appear to be the primary site of immunologicalinjury. Alterations in basal keratinocyte membranes in lichen planus have been suggested by

622 A.K.Bhanetal.diminished lectin binding to these cells in contrast to supra-basal cells, which have normalbinding (Holmstrup & Dabelsteen, 1979). Thus, lichen planus lesions may represent a T cell(T4* helper/inducer cell) mediated response directed against altered basal keratinocytes.Furthermore, Langerhans cells may play an important role in the processing and presentation ofantigen to T4 ' helper/inducer cells in this disorder.

ACKNOWLEDGMENTS

We thank Mr Bruce Kaynor, Mrs Christine Howard and Ms Karen Sokal for their technicalassistance. This work was supported by U.S. Public Health Science Grant HL-18646.

REFERENCES

ALARIO, A., ORTONNE, J., SCHMITT, D . & THIVOLET, J. (1978) Lichen planus: study with anti-human T lymphocyteantigen (anii-HTLA) scrum on frozen tissue seciions. British Journal of Dermatology, 98, 601.

BHAN, A , K . , REINHERZ, E.L., POPPEMA, S.. MCCLITSKEY, R.T. & SCHLOSSMAN, S.F. (19S0) Location of T cells and

major histocompaiibility complex antigens in the human ihymus. Journal of Experimental Medicine, 152, 771.BJERKE, J.R. & KROGH, H.K. (1978) Identification of mononuclear cells in situ in skin lesions of lichen planus. Brihsh

Journal of Dermatology, 98, 605.

DOCKRELL, H . M . & GREENSPAN, J.S. (1979) Histochemical identification of T cells in oral lichen planus. Oral Surgery,

48,42.FITHIAN, E., RUNG, P., GOLDSTEIN, G . , RUBENFELD, M . , FENOGLIO, C . AEDELSON, R . (1981) Reactivity of Langerhans

cells with hybridoma antibody. Proceedings of the National Academy of Sciences, U.S.A., 78, 2544.HARBIST, T J . & MiHM, M . C , JR (1979) Cutaneous immunopathology—the use of direct and indirect immunofluores-

ccncc techniques in the diagnosis nf cutaneous disease. Human Pathology, 10, 625.HASHIMOTO, K . (1976) Apoptosis in lichen planus and several other dermaioses. Intra-epidermal cell death with

filamentous degeneration. Acta dermatirvenereologica iStockholm,, 56, 187.HOLMSTRUP, P. & DABELSTEEN, E . [ 1979) Changes in carbohydrate expression of lichen planus affected epithelial cell

membranes. Journal of Investigative Dermatology, 73, 364.MACDONALD, D.M., SCHMITT, D . , GERMAIN, D . & THIVOLET, J. (1978) Ultrastructural demonstration of T cells in

cutaneous tissue sections using specific anti-human T cell antiserum. Brihsh Journal of Dermatology, 99, 641.MURPHY, G.F., BHAN, A.K., SATO, S., M I H M , M . C , JR & HARRIST, T.J. (1981) A new immunological marker for

human Langerhans cells. New England Journal of Medicine, 34, 791.PoppHMA, S., BHAN. A.K., REINHERZ, E . L , MCCLUSKEY, R.T. & SCHLOSSMAN, S.F. (1981) Distribution of T cell

subsets in human lymph nodes. Journal of Experimental Medicine, 753, 30.REINHERZ, E.L., KtJNc;, P .C , GOLDSTEIN, G . & SCHLOSSMAN, S.F. (1979a) A monoclonal antibody with selective

reactivity with functionally mature thymocytes and all peripheral human T cc\\%,Jounuilof Imwumlogy, 123,1312.REINHERZ, E.L., KUNG, P . C , GOLDSTEIN, G , & SCKLOSSMAN, S.F. (1979b) Separation of functional subsets of human T

cells by e monoclonal antibody. Proceedings of the National Academy of Sciences, U.S.A., 76, 4061.REINHERZ, E.L., KL'NG, P . C , GOLDSTEIN, G . & SCHLOSSMAN, S.F. (1979c) Further characterization of the human

inducer T cell subset defined by monoclonal antibody. Journal of Immunology, 123, 2894.REINHERZ, E.L., KUNG, P . C , PESANDO, J.M., RITZ, J., GOLDSTEIN, G . & SCHLOSSMAN, S.F. (I979d) la determinants on

human T cell subsets defined by monoclonal antibody; activation stimuli required for expression. Journal ofExperimental Medicine, 150, 1472.

REINHERZ, E.L. & SCHLOSSMAN, S.F. (1980) The differentiation and function of human T lymphocytes. Cell, 19, 821.REINHERZ, E.L., KtJNG, P . C , GOLDSTEIN, G . , LEVEY, R.H. & SCHLOSSMAN. S.F. (1980) Discrete stages of human

intrathymic differentiation: analysis of normal thymocytes and leukemic lymphoblasts of T-cell lineage.Proceedings of the National Academy of Sciences, U.S.A., 77, 1588.

SAURAT, J . H . & GLUCKMAN, E . (1977) Lichen planus-like eruption following bone marrow transplantation: amanifestation of the graft-versus-host disease. Clmicai and Experimental Dermatology, 2, 335.

THORBECKE, G,J,, SiLBERBERG-SiNAKiN, I. & FLOTTE, T.J. C1980) Langerhans cells as macrophages in skin andlymphoid organs. Journal of tnvestigatiz'e Dermatology, 75, 32.

WALKER, D.M, (1976) Identification of subpopulations of lymphocytes and macrophages in the infiltrate of lichenplanus lesions of skin and oral mucosa. British Journal of Dermatology, 94, 529.

WEEDON, D . (1974) Civatte Bodies and apoptosis. British Journal of Dermatology, 91, 357.