Lichen planus: Immunologic and morphologic identification of the submucosal infiltrate

Lichen planus: Immunologic and morphologic identification of the submucosaf infiltrate Joseph A. Regezi, D.D.S., M.S., * Michael J. Deegan, M.D., ** and James R. Hayward, D.D.S., M.S.,*** Ann Arbor, Mich. DEPARTMENT

OF ORAL

PATHOLOGY,

SCHOOL

SCHOOL

OF DENTISTRY,

PATHOLOGY,

SCHOOL

OF MEDICINE;

AND

UNIVERSITY

OF DENTISTRY;

DEPARTMENT

DEPARTMENT

OF ORAL

OF

SURGERY,

OF MICHIGAN

The purpose of this investigation was to specifically identify T cells, B cells, and histiocytes in the infiltrate typically seen in lichen planus. In frozen tissue sections, AET-treated sheep erythrocytes formed immunologic rosettes with the lymphocytes in the infiltrate, designating them as T cells. Rosette assays with reagent erythrocytes, IgGEA, IgMEA, IgMEAC, and E resulted in nonadherence, indicating a lack of B cells and macrophages, and indirectly implicating them as T cells. Scanning electron microscopy of the cellular infiltrate, in situ, showed that the cells had smooth, nonvillous surfaces. These observations were consistent with a T cell origin and were considered supportive of the immunologic data. The results of this investigation support the hypothesis that lichen planus is a disease mediated by thymus-dependent lymphocytes.

A lthough lichen planus has been recognized

for more than 100 years, little is known of its etiology and pathogenesis. Morphologic studies have been helpful but have provided a limited amount of information on the nature of the disease. The generally accepted light microscopic features of liquefaction degeneration of basal cells with accompanying lymphoytic infiltrate and hyperkeratosis have led most investigators to believe that the basic defect was epithelial in origin. Electron microscopic studies which have focused primarily on epithelial cell alterations have supported this contention. Fragmentation of basement membrane, degeneration of basal cells, and destruction of basal-cell desmosomes and hemidesmosomes have been consistently observed and emphasized the importance of basal epithelial cells in the pathogenesis of lichen planus.le3 Studies have recently shifted from the examination of epithelial changes to the exam*Associate Professor, Department of Oral Pathology. **Assistant Professor, Department of Pathology. (Present Pennsylvania School of Medicine, Philadelphia, Pa. ***Professor and Chairman, Department of Oral Surgery.

44

address,

Department

0030-4220/78/0146-tlO44$00.90/0

of Pathology,

0

University

1978 The C. V. Mosby

of

Co.

Volume 46 Number I

Lichen planus

45

Fig. 1. Light micrograph of frozen section from biopsy of Patient 3. The section was treated with IgGEA and shows adherence of only a few red cells (arrows) among the cellular infiltrate. This reaction was graded as O/+ on a scale of 0 to 4+. (Hematoxylin and eosin stain. Magnification, X 200.)

ination of submucosalchanges.4The role of the mononuclear infiltrate characteristically seen in lichen planus has received more attention since the development of new immunologic methods for the detection of lymphocyte subpopulations.The general lack of plasma cells has suggestedthat the cells in the infiltrate of lichen planus are thymusdependent lymphocytes (T cells) rather than bone marrow-dependent lymphocytes (B cells). Identification of an infiltrate consisting primarily of T cells could be an important step toward an understandingof the pathogenesisof lichen planus. Immunologic studies have shown that macrophages, T cells, and B cells can be specifically identified in frozen tissue sections by their reaction (rosette formation) with specially treated sheep erythrocytes. 5 Erythrocytes sensitized with subagglutinating amountsof immunoglobulin G (IgGEA) adhereto macrophages(Fc receptor) but not to T cells or B cells. Erythrocytes sensitized with immunoglobulin M and complement (IgMEAC) adhere to B cells and macrophages(Qeceptor) but not to T cells. Unsensitized erythrocytes adhere to T cells but not to B cells or macrophages.However, this latter reaction was, until recently, possible in lymphocyte suspensionsonly. Therefore, in previous studies on tissue sections, T cells were identified indirectly by exclusion of B cells and macrophages. Recent developments in immunologic methods have made it possibleto identify T cells in tissuesectionsby the useof anti-T cell serum6or by rosette formation with sheep erythrocytes treated with 2-aminoethylisothiouronium bromide (AET).’ Walked studied immunologic surface markers of lymphocytes in frozen sectionsof skin and mucous membranelesionsof lichen planus patients and showedthat only a few of the cells in the infiltrate were B cells or macrophages.By the processof elimination, the

46

Regezi,

Deegan,

and Hayward

Fig. 2. Light micrographs

of E-AET-treated frozen sections. Left: submucosal infiltrate from Patient 3 showing positive red cell adherence test (hollow circles are AET-treated red cells). Right: positive control showing red cell adherence to thymocytes in frozen thymus tissue. (Both sections, Hematoxylin and eosin stain, Magnification, X 250.)

Table 1. Resultsof rosette assays

*Erosive

Case No.

1nGEA

1*

+

0

2 3 4* 5 6* I 8*

0 o/+ +++ 0 0 0 0

0 0 0 0 0 0 0

form

of lichen

1aMEAC

E-AET +++

+++ +++ ++ ++ +++ i-k/-+++ -

planus.

large majority of cells were assumedto be T cells. Lymphocytes extracted from skin biopsiesof four patients with lichen planuswere shown to be T cells becausethey formed spontaneousrosettes with sheeperythrocytes in suspension.’Parallel studiesof mucous membranelesionshave not been done, nor have T cells been identified directly in tissue sections. An additional, although less reliable way to identify mononuclear cells has been through differences in surface morphology. Scanning electron microscopic studieshave shown that under appropriate conditions, T cells, B cells, and macrophagesexhibit distinctive surface architecture. Polliak and associateslohave demonstratedthat B cells have villous projections, whereasT cells have smooth surfaces. Since their original publica-

Volume Number

46 1

Lichen planus

Fig. 3. Scanning electron abundant smooth-surfaced

micrograph of submucosal cells covering connective

47

connective tissue after epitklium was stripped away. Note tissue ridges. (Magnification: x300; inset, x 1,200.)

tion, other investigators have questioned the validity of their results, suggesting that surface morpology is a function of environmental influences induced in preparation of the lymphocyte suspensions. l1 Others have also demonstrated intermediate forms between villous and smooth cells. l2 Observations on normal lymph nodes and spleen indicate that macrophages can be identified and T and B cells can be differentiated from each other when scanning electron microscopy (SEM) is coupled with immunologic studies.13z l4 Investigators using SEM to study lymphoreticular neoplasms have been able to determine the cell of origin of many lymphomas and leukemias.15-18 In some cases SEM was not considered to be a useful adjunct in cell identification because of overlap of surface characteristics.ig, *O The purpose of this investigation is to seek support for the hypothesis that lichen planus is a disease mediated by thymus-dependent lymphocytes through the identification of the specific cell types making up the mononuclear infiltrate. This will be accomplished by the recognition of immunologic surface receptors on the cells and by SEM observations of cell surface architecture. MATERIALS

AND METHODS

Tissue for this study came from eight patients with oral lichen planus (four with the form and four with the nonerosive form). Buccal mucosabiopsy specimenswere subdivided for morphologic and immunologic studies. Light microscopy was used to confirm clinical diagnosis. Fresh tissue for immunologic rosette tests was frozen and stored at -70” C. until needed. Normal spleen, lymph node, and thymus were prepared in a similar manner for use as controls. erosive

40

Regezi,

Deegan,

and Hayward

Oral Surg. July, 1978

Fig. 4. Scanning electron micrograph of lymphocytes covering connective tissue papillae from Patient 6. (Magnification, x 1,200.) Inset: surface morphology of macrophage seen in infiltrate of Patient 4. (Magnification, X7,500.)

Sheeperythrocytes were incubated with appropriate concentrationsof IgG, IgM, and complement as previously described.16The resultant erythrocytes served as reagent cells (IgGEA, IgMEA, IgMEAC) to test for B cells and macrophages.Washed unsensitized erythrocytes (E) were also used as reagent cells. Although expected to be negative, IgMEA and E tests were run as negative controls to rule out nonspecific erythrocyte adherence. Rosette assaysfor these tests were done on air dried 8 p frozen sections. Tissueslayered with IgGEA, IgMEA, IgMEAC, and E were incubated for 30 minutes at room temperature. Nonadherentreagenterythrocytes were washedoff with cold buffered saline solution. The sections were then fixed, stained with hematoxylin and eosin, and examined by both brightfield and darkfield microscopy. Adherence of erythrocytes was assessedaccording to a scale of 0 to 4+. Rosette assaysfor T cells were performed on frozen sectionsaccording to the method of Brubaker,and Whiteside. Moist sectionswere layered with 1 per cent suspensionof sheeperythrocytes in dextrose-gelatin-veronal-buffered saline with calcium and magnesiumadded. The erythrocytes were previously treated with 2-aminoethylisothiouronium bromide (E-AET).21 After an incubation period of 30 minutes in a humidity chamber, the sections were washed with buffer, fixed with 3 per cent glutaraldehyde, washed, and stainedwith hematoxylin and eosin. Fresh tissue for SEM was fixed in 2.5 per cent glutaraldehyde. With the use of microdissection instruments, the epithelium was mechanically teasedaway from the supporting connective tissue. The tissue was then postfixed in osmium tetroxide and dehydrated in a graded seriesof alcohol and amyl acetate. The tissue was then transferred to a critical point drying apparatusfor drying.16 The tissue was coated with gold in a glow-

Volume 46 Number I

Lichen planus 49

Fig. 5. Scanning electron micrograph of lymphocytic infiltrate from Patient 6. Note relatively smooth cell surfaces. Magnification, x 500.)

dischargecoater and examined in the SEM. In one case, lymphocytes were teasedout of the biopsy specimeninto a Petri dish containing buffer. The cells were then centrifuged and resuspendedin 2.5 per cent glutaraldehyde. They were then postfixed, dehydrated, dried, gold coated, and examined as described above. RESULTS

Light microscopic confirmation of all lichen planus biopsieswas made. Morphologically, the mononuclear infiltrates were composed primarily of lymphocytes with small numbersof other mononuclearcells. Plasmacells were noted in small numbersin biopsies from Cases1, 4, and 8. Results of the rosette assaysare illustrated in Table I. All rosette tests with IgMEA, IgMEAC, and E were negative. Cases1, 3, and 4 showed rosette formation with IgGEA reagent; all others were negative (Fig. 1). Resultswith E-AET were positive in all patients (Fig. 2) except Patient 8, in whom there was insufficient tissue for complete testing. In control tissuesIgMEAC formed rosettes in follicular areasof spleen and lymph node. IgGEA formed rosettesin red pulp of spleenand medullary areasof lymph node and E-AET formed rosettes with thymocytes in thymic tissue (Fig. 2). The surface architecture of the lymphocytes was relatively uniform for all specimens. Lymphocytes were readily found on the surface of the connective tissue after the epithebum was stripped off (Fig. 3). Groups of lymphocytes were seenas grapelike clusters on connective tissue papillae (Fig. 4) or as isolated cells covering broad areasin a cobblestone fashion (Fig. 5). The individual cells had smooth surfaceswith little or no tendency toward the formation of villi. Proportionately larger cells with blunted cytoplasmic projections were seenin tissue from Patient 4 and were considered to be macrophages(Fig.

50

Regezi, Deegan, and Hayward

Oral Surg. July, 1978

Fig. 6. Scanning electron micrograph of cells teased out of biopsy specimen from Patient 5. Smooth lymphocyte to the right, RBC at the bottom, and crenated RBC’s at top and right. Magnification, x I .500.)

4). In the casewhere mononuclearcells were studied in suspensionafter being teasedout of the biopsy specimen, smooth-surfacedlymphocytes and erythrocytes dominated the cell types seen (Fig. 6). Infrequently seen were large ruffled cells (macrophages)and crenated erythrocytes. DISCUSSION

It is apparent from the immunologic data that the mononuclearinfiltrate in oral lichen planus is composedalmost exclusively of T cells. The cells were positively identified in tissue, both directly by the formation of rosettes with AET-treated erythrocytes and indirectly by the lack of rosette formation with IgGEA (with exceptions) and IgMEAC. The formation of rosetteswith IgGEA in Cases1, 3, and 4 is interpreted as identification of Fc receptors on macrophages.The presenceof macrophagesin two of these cases(1 and 4) can be explained in part as being representative of a nonspecific inflammatory cell infiltrate associatedwith ulceration near the biopsy site. Their presencemay alsobe due to a role in the processingof antigen in a cell-mediated responseimportant in the pathogenesisof lichen planus.’ The finding of smooth-surfacedcells in the lichen planus infiltrate is consideredonly circumstantial evidence of a T cell origin becauseof the controversy currently surrounding surface identification of lymphocytes. However, theseobservationsare supportive of the immunologic data and add an interesting facet to cell identification. Final judgement of the SEM studiesmust wait until adequatecontrol tissue(B cell diseaseof mucousmembrane) can be found and studied in a similar manner. The humoral immune system (B cells) apparently plays only a minor role in the

Volume 46

Number 1

Lichen phus 51

development oflichen planus, since Bcellsareuncommonly seen intheinfiltrate83 9and antigen-antibody complexes areinconsistently found inthebasement membranearea.229 23 The finding of an infiltrate composed almost exclusively of T cells demonstrates the importance of the cell-mediated immune system in the pathogenesis of this disease. It is suggestive of an immune condition similar to cell-mediated reaction seen in graft rejection and delayed hypersensitivity. This cell-mediated reaction in lichen planus could theoretically be directed against an epithelial antigen(s). 25 This would also be consistent with ultrastructural observations which have shown that primary changes are found in basal cells.ld3 Unfortunately, initial attempts to prove this hypothesis have been unproductive. Investigators have attempted, unsuccessfully, to show that peripheral lymphocytes from patients with lichen planus are cytotoxic for autologous cultured epithelial cells.26 Attempts at lymphocyte transformation with extracts of epithelium have been negative.” Additional studies in these areas should be done (especially with lymphocytes extracted from lichen planus lesions) in order to further understand the etiology and pathogenesis of lichen planus. SUMMARY

In this investigation biopsies from patients with oral lichen planus were evaluated for the presence of T cells, B cells, and macrophages in the submucosal infiltrate. Erythrocyte rosette assays on frozen sections with the reagent cells, IgGEA, IgMEA, IgMEAC, E, and E-AET demonstrated that the infiltrate was composed primarily of T cells. SEM showed that the cells had smooth surfaces, which suggested a T cell origin also. The results supported the hypothesis that lichen planus is a disease mediated by thymus-dependent lymphocytes. REFERENCES I. Ebner, H., and Gebhart, W.: Epidermal Changes in Lichen Planus, J. Cutan. Pathol. 3: 167-179, 1976. 2. El Labban, N. G., and Kramer, I. R. H. : Light and Electron Microscopic Study of Liquefaction Degeneration in Oral Lichen Planus, Arch. Oral Biol. 20: 653-657, 1975. 3. Shklar, G., Flynn, E., and Szabo, G.: Basement Membiane Alteration in Oral Lichen Planus, J. Invest. Dermatol. 70: 45-50, 1978. 4. Walker, D. M.: The Inflammatory Infiltrate in Lichen Planus Lesions, J. Oral Pathol. 5: 277-286, 1976. 5. Edelson, R. L., Smith, R. W., Frank, M. M., and Green, I.: Identification of Subpopulations of Mononuclear Cells in Cutaneous Infiltrates, J. Invest. Dermatol. 61: 82-89, 1973. 6. Whiteside, T. L.: Reactivity of Anti-Human Brain Serum with Human Lymphocytes, Am. J. Pathol. 86: I-16, 1978. 7. Brubaker, D. B., and Whiteside, T. L.: Localization of Human T Lymphocytes in Tissue Sections by a Rosetting Technique, Am. J. Pathol. 86: 323-333, 1978. 8. Walker, D. M.: Identification of Subpopulations of Lymphocytes and Macrophages in the Infiltrate of Lichen Planus Lesions of Skin and Oral Mucosa, Br. J. Dermatol. 94: 529-534, 197.5. 9. Tan, R., Byrom, N., and Hayes, J.: A Method of Liberating Living Cells from the Dermal Infiltrate, Br. J. Dermatol. 93: 271-275, 1975. 10. Polliack. A., Lampcn, N., Clarkson, B., and DeHarven, E.: Identification of Human B and T Lymphocytes by Scanning Electron Microscopy, J. Exp. Med. 138: 607-624, 1973. 1 I. Alexander, E., and Wetzel, B.: Human Lymphocytes: Similarity of B and T Cell Surface Morphology, Science 188: 732-734, 1975. 12. Lin, P., Cooper, A., and Wortis, H.: Scanning Electron Microscopy of Human T Cell and B Cell Rosettes, N. Engl. J. Med. 289: 548-551, 1973. 13. Luk, S., Napajaroonsri, C., and Simon, G.: The Architecture of the Normal Lymph Node and Hemolymph Node, Lab. Invest. 29: 258-265, 1973. 14. Weiss, L.: A Scanning Electron Microscopic Study of the Spleen, Blood 43: 665-674, 1974. 15. Bentwich, Z., Polliack, A., and Douglas, S. D.: Surface Markers and Other Characteristics of the Lymphocyte in Chronic Lymphocytic Leukemia, Isr. J. Med. Sci. 12: 304324, 1976.

52

Regezi, Deegan, and Hayward

16. Deegan, M. J., Cossman, J., Chosney, B. T., and Schnitzer, B.: Hairy Cell Leukemia, An immunologic and Ultrastructural Study, Cancer 38: 1952- 1961, 1976. 17. Domagala, W., Emeson, E., Greenwald, E., and Koss, L.: A Scanning Electron Microscopic and Immunologic Study of B-Cell Lymphosarcoma Cells in Cerebrospinal Fluid, Cancer 40: 716-720, 1977. IS. Polhack, A., Siegal, F., Clarkson, B., Fu, S., Winchester, R., Lampen, N., Siegal, M., and DeHarven, E.: A Scanning Electron Microscopy and Immunological Study of 84 Cases of Lymphocytic Leukemia and Related Lymphoproliferative Disorders, Stand. J. Haematol. 15: 359-376, 1975. 19. Greenberg, B., Peter, C., Glassy, F., and MacKenzie, M.: A Case of T Cell Lymphoma with Convoluted Lymphocytes, Cancer 38: I602- 1607, 19??. 20. Van Ewijk, W., Brons, N., and Rozing, J.: Scanning Electron Microscopy of Homing and Recirculating Lymphocyte Populations, Cell. Immunol. 19: 245-261, 1975. 21. Kaplan, M., and Clark, C.: An Improved Rosetting Assay for Detection of Human T Lymphocytes, J. Immunol. Meth. 5: 13 I - 135, 1974. 22. Abell, E., Presbury, D., Marks, R., and Ramnarain, D.: The Diagnostic Significance of Immunoglobulin and Fibrin Deposition in Lichen Planus, Br. J. Dermatol. 93: 17-27, 1975. 23. Tuffanelh, D.: Cutaneous Immunopathology: Recent Observations, J. Invest. Dermatol. 65: 143- 153, 1975. 24. Ebner, H.: Untersuchungen uber die Cellulare Zusammensetzung des Lichen Ruber Planus Infiltrates, Arch. Dermatol. Forsch. 247: 309-318, 1973. 25. Dolby, A., and Slade, M.: Failure to Detect Immune Epithelial Cytolysis in Oral Lichen Planus, J. Dent. Res. 52: 985, 1973. 26. Walker, D.: Immunological Processes Involving the Oral Mucosa in Lichen Planus, Proc. R. Sot. Med. 69: 7-8, 1976. Reprint requests to: Joseph A. Regezi, D.D.S., M.S. Department of Oral Pathology School of Dentistry University of Michigan Ann Arbor, Mich. 48109