Review article
Immunopathogenesis of oral lichen planus
L. J. Walsh^ ^ N. W. Savage^ T. Ishii^ and ^G. J. Seymour^ 'Department ol Dermatology, University ot Pennsytvanta, Phttadetphia, USA and 'Immunopathology Unit, Department ot Dentistry, University ol Queensland, Brisbane, Austratia
Walsh LJ, Savage NW, Ishii T, Seymour GJ. Immunopathogenesis of oral lichen planus, J Oral Pathol Med 1990; 19: 389-96. Oral lichen planus (LP) is a common mucosal disorder in which cell mediated immunity is thought to play a major role. In this paper, a unifying hypothesis which attempts to integrate cellular and molecular signals in the loeal immune response in oral LP is presented. In this model, modified keratinocyte surface antigens are the target for the cytotoxic cell response which characterizes oral LP, whereas mast cells and antigen presenting Langerhans cells are key cellular elements in the evolving lesion. It has been established that mast cell degranulation induces adhesion molecule expression on endothelium whieh facilitates lymphocyte homing to the tissues. These adhesive interactions between lymphocytes and keratinocytes are postulated to be itnportant detenitinants in the effector phase of the lesion. Cytokines produced by both lymphocytes atid keratinocytes which influence the local iinmune response could promote chronicity. Accordingly, modulation of immunologie events is a potential therapeutic approach for oral LP,
Accepted for publication July 16, 1990
Oral lichen planus (LP) is a relatively corntiton disorder of unknown etiology which may affect tnucosal and/or skin surfaces, Cornpared with skin lesions, mucosal affections are far trtore chronic in nature, and often persist for many years (1, 2), Because of the rnorbidity associated with these oral lesions and their propensity for malignant development (3), oral LP occupies an important place in the practice of oral pathology and rnedicine. Oral LP has a variety of clinical presentations, including the reticular form in which characteristic Wickam's striae are seen (Fig, 1), Of the various clinically recognizable types of oral LP, papular and uleerative lesiotis tend to occur early in the disease whereas atrophic and plaque-like variants generally appear later (1, 2). While the initial presence of papular lesions is predictive of complete remission (2), the influence of other clinical factors on the course of the disease is unclear, and remains a topic for further study.
dominant cell type is the T lytnphocyte (CD3 + ), although the relative proportion of helper-inducer (CD4 + ) and suppressor-cytotoxic (CD8-I-) T cells is variable (17-21), This variation may be related to disease phase, as reported for epidermal LP (22), with CD4-|-and CD8-I-T cell populations predominating in early and late lesions, respectively. An increase in the number (17, 18, 23) and/or activation status (24, 25) of antigen presetiting Langerhans cells (LC) in both the comiective tissue and the epithelium is a consistent finding in oral LP. It is strikitig that the immunohistology of oral LP resetnbles that described for delayed-type hypersensitivity (DTH) reactions and chronic graft-versus-host disease (GVHD) (26-28). Furthermore, oral lesions of GVHD frequently resemble oral LP (Fig. 3). In view of the central role of T lymphocytes as effector cells in both these lesions, a key role for cell mediated immunity in the pathogenesis of oral LP is likely.
With regard to patietit factors, available evidence supports the view that LP is a predetermined condition or diathesis, rather than a simple cause and effect disorder (I, 4). Serologic typing studies have demonstrated an association between the HLA-DR 1 tissue type and
both drug-related and idiopathic lichenoid lesions (5, 6), consistent with the view that a genetic predisposition or susceptibility to the development of LP exists. Nurnerous faetors appear capable of aggravating or precipitating oral LP (Table 1), however it is unclear how such diverse influences elicit the disorder. This paper reviews the evidence for atl imtnunopathogenesis for oral LP and considers imtnune mechanisms which may be critical in the development and progression of this entity. Immunohistologic aspects of oral lichen planus As seen iti routine histologie exatnination, the pathognomonic features of oral LP include a subepithelial tnononuclear cell inftltrate containing occasional plasma cells, together with variable nurnbers of intraepithelial lymphocytes and apoptotic bodies (Fig, 2A), Destruction of the basetnent membrane and close apposition of lymphocytes and basal keratinocytes are characteristic findings (Fig. 2B), Alterations in epithelial architecture (atrophy, acanthosis, hyperkeratosis) are frequently observed. Analysis of the lesional cell populations has revealed that the pre-
Key words: adhesion molecules; lichen planus, immunology, pathology: lymphocytes; mast cells. L. J. Walsh, Dental School, Turbot Street, Brisbane, Qld. 4000, Australia.
Recent studies have provided direct evidence for the central role of T lymphocytes in the evolution of epidermal LP. In lesional skin transplanted onto athymic mice (which lack functional T cells), characteristic histologic features of LP disappear (29, 30). Similar
390
WALSH et cd.
changes occur in explants of lesional skin maintained in organ culture (29). These findings indicate that migration of T lymphocytes and other elements of the cellular immune response is a critical event in the pathogenesis of LP. Further support for this view has heen provided
the evidence that antigenic change in skin or mucosa is a critical event in A major obstacle to elucidating the in LP (1, 4, 35), the DTH reaction may situ microenvironments in oral lichen- provide a useful mode! to examine the oid lesions is related in part to the sam- initiation phase or oral LP. It is well pling process used, in that biopsy mate- established that antigen presenting cells rial obtained for diagnostic purposes re- are required for cell mediated itiimune by the elegant studies of SHIOHARA and presents a single timepoint in the response to antigens in an epithelial micolleagues which demonstrated that LP- development of the lesion. Thus, it is lieu (36). Both LC in the epithelium and like skin lesions could be induced in likely that the majority of "routine" oral dendritic cells in the submucosa are posyngeneic mice by local transfer of LP material comprises lesions in the ef- tent stimulators of T lymphocytes CD4-I-T lymphocyte clones with cyto- fector phase, in which gross microscopic (37-39). Interactions between these antoxic capabilities (30-32). Remarkably, changes have occurred and overt clinical tigen presenting cells and T cells may these clones were autoreactive in that pathology is present. As such, the dis- occur locally (in the submucosa) or in they were specific for self la antigens, tinction between primary immunologie the lymph node paracortices. In both the murine equivalent of human major events and epiphenomena in this mate- sites, clustering of T cells about dendrithistocompatibility (MHC) Class II rial is difficult to make. However, some ic cells occurs. This process is mediated HLA-DR antigens. This latter finding insight into early events (the "initiation by adhesive interactions between cell suggests that autoreactive CD4-I-cells phase") of oral LP may conceivably be surface molecules, specifically intercelmay contribute to cytotoxicity directed gained from existing information regar- lular adhesion molecule-1 (ICAM-l, against oral mueosal keratinocytes, ding leukocyte trafficking in normal CDS4) on LC (40) and lymphocyte which are frequently observed to ex- and pathological oral mucosa and epi- function-associated antigen-1 (LFA-1, press HLA-DR antigens in lesions of dermis. CDlla) on T cells (41). The effector oral LP. Despite evidence for humoral cells generated by this response include autoimmunity and cellular hypersentivicytoxic T lymphocytes (42), which may ty in some cases of oral LP (33, 34), Epithelial cells be either CD8 + (MHC Class I rethe exact contribution of cell-mediated stricted) or CD4-|-(MHC Class II reIn view of the similarities between oral autoimmunity remains to be deter- LP and the DTH reaction (26-28), and stricted). mined. Because of their extensive dendrites, LC are ideally positioned to respond to antigens which bind to keratinocytes (contact sensitizers, allergens, drugs) or are expressed on the keratinoeyte surfaee membrane (MHC and viral antigens). In this context, the eell mediated response initiated by LC will likely include cytotoxic cells directed against keratinocytes, as seen in GVHD and severe DTH reactions (28). Involvement of viral antigens in oral LP has been suggested (see Ref. 1 for review). The joint observations that mucosal nerve Evolution of oral lichen planus
Fig. 1. Reticular lesions of oral lichen planus distributed on the buccal mucosa of a 68-yrold woman. Table 1. Factors which precipitate or aggra vate oral lichen planus Systemic medications Non-steroidal anti-inflammatory drugs, Antihyperlcnsivc agents, gold salts, Angiotensin-converting enzyme Inhibitors (I, 4, 7-10) Local factors Dental materials Amalgam/metals (10-11) Composite resin (12) Cyanoacrylate (13) Trauma Surgery (14) Stress (15, 16) Viral infection (I)
'A^Hi f/g. 2. Histology of oral LP. A, dense subepithelial mononuclear cell infiltrate is present. X 10. B, lymphocytes in lamina propria and epithelium are a conspicuous feature. Basal cell liquefaction foci are apparent, x 40.
Fig. 3. Erosive lichenoid lesions affecting gingiva and labial mucosa in patient with chronie graft-versus-host disease. This individual received bone marrow transplant from HLAidentical sibling during treatment for acute myeloid leukemia.
Oral lichen planus
391
fibers closely appose basal keratinocytes (Fig. 4B), and that keratinoeyte expression of receptors (CD21) for Epstein Barr virus is upregulated in lesions of oral LP (Fig. 4A) suggest that low-grade and/or persistent infection of epithelial cells with Herpes group viruses may be a possible etiologic factor in LP.
ri^iffrsi^?
Mast cells
In addition to their association with vascular elements and epithelial cells, there is accumulating evidence that mucosal nerves closely appose mast cells (43^5). Chronic stimulation of these nerves, together with the release of neuropeptides such as substance P, induce mast cell degranulation (46-47). The significance of this event in altering leukocyte trafficking is seen in the ohservation that mast cell degranulation induces expression of endothelial leukocyte adhesion molecule-l (ELAM-1), a glycoprotein which is critically required for leukocyte adhesion to the luminal surface of endothelial cells (48). Thus, mast cells serve as gatekeepers of the microvascular bed, and provide a link between neural networks and immune cell trafficking (48, 49). Mast cells contain immunoreactive tumor necrosis factor alpha (TNF-a) (Fig. 4C), which is released upon degranulation. This cytokine upregulates the expression of endothelial eell adhesion molecules, including ELAM-1 and ICAM-l (Fig. 4D, E), together with the recently described vascular cell adhesion molecule-1 (VCAM1) (50, 51). In oral LP, mast cell degranulation (Fig. 4 F-H) may be promoted by drugs implicated in lichenoid eruptions, such as thiazides (52), or by neuropeptides secreted from nerves affeeted fig. '#. Immunohistology of oral LP, showing by electrical potentials (galvanic efcells involved in putative induction phase. All sections except F and G stained using fects), trauma (surgery), infection (Herimmunoperoxidase. A, CD21 (EBV receptor) pes group viruses), or psychologic expression by keratinocytes in oral LP. AntiCR2 antibody. X 40. B. mucosal nerves (ar- stress. In this fashion, the interaction rows) in close proximity to basal keratinobetween nerves and mast cells may pro. _ _ ^ « . . . «_ cytes identified by expression of neural cell vide a common pathway through which A . . . J j m l . . ^ p * - •*"'^^'*-^^ adhesion molecule (NCAM). NKH-I certain agents known to aggravate LP '• t " , " » • • • • : ' ..3».. .^Af^ antibody, X40. C, immunoreaetive tumor neexert their effect. Release of cytokines crosis factor in mast cells in normal human such as TNF-a from mast cells and conoral mucosa. B154.2 antibody, x 100. D, expression of intercellular adhesion molecule-l sequential induction of adhesion mole(ICAM-l) on endothelial cells in oral LP cule expression may therefore comprise Anti-ICAM antibody, x 40. E. expression of an important event in the initiation of endothelial leukocyte adhesion molecule-l the lymphocytic infiltration which char(ELAM-1) on endothelial cells in orat LP. «
Immunopathogenesis of oral lichen planus
L. J. Walsh^ ^ N. W. Savage^ T. Ishii^ and ^G. J. Seymour^ 'Department ol Dermatology, University ot Pennsytvanta, Phttadetphia, USA and 'Immunopathology Unit, Department ot Dentistry, University ol Queensland, Brisbane, Austratia
Walsh LJ, Savage NW, Ishii T, Seymour GJ. Immunopathogenesis of oral lichen planus, J Oral Pathol Med 1990; 19: 389-96. Oral lichen planus (LP) is a common mucosal disorder in which cell mediated immunity is thought to play a major role. In this paper, a unifying hypothesis which attempts to integrate cellular and molecular signals in the loeal immune response in oral LP is presented. In this model, modified keratinocyte surface antigens are the target for the cytotoxic cell response which characterizes oral LP, whereas mast cells and antigen presenting Langerhans cells are key cellular elements in the evolving lesion. It has been established that mast cell degranulation induces adhesion molecule expression on endothelium whieh facilitates lymphocyte homing to the tissues. These adhesive interactions between lymphocytes and keratinocytes are postulated to be itnportant detenitinants in the effector phase of the lesion. Cytokines produced by both lymphocytes atid keratinocytes which influence the local iinmune response could promote chronicity. Accordingly, modulation of immunologie events is a potential therapeutic approach for oral LP,
Accepted for publication July 16, 1990
Oral lichen planus (LP) is a relatively corntiton disorder of unknown etiology which may affect tnucosal and/or skin surfaces, Cornpared with skin lesions, mucosal affections are far trtore chronic in nature, and often persist for many years (1, 2), Because of the rnorbidity associated with these oral lesions and their propensity for malignant development (3), oral LP occupies an important place in the practice of oral pathology and rnedicine. Oral LP has a variety of clinical presentations, including the reticular form in which characteristic Wickam's striae are seen (Fig, 1), Of the various clinically recognizable types of oral LP, papular and uleerative lesiotis tend to occur early in the disease whereas atrophic and plaque-like variants generally appear later (1, 2). While the initial presence of papular lesions is predictive of complete remission (2), the influence of other clinical factors on the course of the disease is unclear, and remains a topic for further study.
dominant cell type is the T lytnphocyte (CD3 + ), although the relative proportion of helper-inducer (CD4 + ) and suppressor-cytotoxic (CD8-I-) T cells is variable (17-21), This variation may be related to disease phase, as reported for epidermal LP (22), with CD4-|-and CD8-I-T cell populations predominating in early and late lesions, respectively. An increase in the number (17, 18, 23) and/or activation status (24, 25) of antigen presetiting Langerhans cells (LC) in both the comiective tissue and the epithelium is a consistent finding in oral LP. It is strikitig that the immunohistology of oral LP resetnbles that described for delayed-type hypersensitivity (DTH) reactions and chronic graft-versus-host disease (GVHD) (26-28). Furthermore, oral lesions of GVHD frequently resemble oral LP (Fig. 3). In view of the central role of T lymphocytes as effector cells in both these lesions, a key role for cell mediated immunity in the pathogenesis of oral LP is likely.
With regard to patietit factors, available evidence supports the view that LP is a predetermined condition or diathesis, rather than a simple cause and effect disorder (I, 4). Serologic typing studies have demonstrated an association between the HLA-DR 1 tissue type and
both drug-related and idiopathic lichenoid lesions (5, 6), consistent with the view that a genetic predisposition or susceptibility to the development of LP exists. Nurnerous faetors appear capable of aggravating or precipitating oral LP (Table 1), however it is unclear how such diverse influences elicit the disorder. This paper reviews the evidence for atl imtnunopathogenesis for oral LP and considers imtnune mechanisms which may be critical in the development and progression of this entity. Immunohistologic aspects of oral lichen planus As seen iti routine histologie exatnination, the pathognomonic features of oral LP include a subepithelial tnononuclear cell inftltrate containing occasional plasma cells, together with variable nurnbers of intraepithelial lymphocytes and apoptotic bodies (Fig, 2A), Destruction of the basetnent membrane and close apposition of lymphocytes and basal keratinocytes are characteristic findings (Fig. 2B), Alterations in epithelial architecture (atrophy, acanthosis, hyperkeratosis) are frequently observed. Analysis of the lesional cell populations has revealed that the pre-
Key words: adhesion molecules; lichen planus, immunology, pathology: lymphocytes; mast cells. L. J. Walsh, Dental School, Turbot Street, Brisbane, Qld. 4000, Australia.
Recent studies have provided direct evidence for the central role of T lymphocytes in the evolution of epidermal LP. In lesional skin transplanted onto athymic mice (which lack functional T cells), characteristic histologic features of LP disappear (29, 30). Similar
390
WALSH et cd.
changes occur in explants of lesional skin maintained in organ culture (29). These findings indicate that migration of T lymphocytes and other elements of the cellular immune response is a critical event in the pathogenesis of LP. Further support for this view has heen provided
the evidence that antigenic change in skin or mucosa is a critical event in A major obstacle to elucidating the in LP (1, 4, 35), the DTH reaction may situ microenvironments in oral lichen- provide a useful mode! to examine the oid lesions is related in part to the sam- initiation phase or oral LP. It is well pling process used, in that biopsy mate- established that antigen presenting cells rial obtained for diagnostic purposes re- are required for cell mediated itiimune by the elegant studies of SHIOHARA and presents a single timepoint in the response to antigens in an epithelial micolleagues which demonstrated that LP- development of the lesion. Thus, it is lieu (36). Both LC in the epithelium and like skin lesions could be induced in likely that the majority of "routine" oral dendritic cells in the submucosa are posyngeneic mice by local transfer of LP material comprises lesions in the ef- tent stimulators of T lymphocytes CD4-I-T lymphocyte clones with cyto- fector phase, in which gross microscopic (37-39). Interactions between these antoxic capabilities (30-32). Remarkably, changes have occurred and overt clinical tigen presenting cells and T cells may these clones were autoreactive in that pathology is present. As such, the dis- occur locally (in the submucosa) or in they were specific for self la antigens, tinction between primary immunologie the lymph node paracortices. In both the murine equivalent of human major events and epiphenomena in this mate- sites, clustering of T cells about dendrithistocompatibility (MHC) Class II rial is difficult to make. However, some ic cells occurs. This process is mediated HLA-DR antigens. This latter finding insight into early events (the "initiation by adhesive interactions between cell suggests that autoreactive CD4-I-cells phase") of oral LP may conceivably be surface molecules, specifically intercelmay contribute to cytotoxicity directed gained from existing information regar- lular adhesion molecule-1 (ICAM-l, against oral mueosal keratinocytes, ding leukocyte trafficking in normal CDS4) on LC (40) and lymphocyte which are frequently observed to ex- and pathological oral mucosa and epi- function-associated antigen-1 (LFA-1, press HLA-DR antigens in lesions of dermis. CDlla) on T cells (41). The effector oral LP. Despite evidence for humoral cells generated by this response include autoimmunity and cellular hypersentivicytoxic T lymphocytes (42), which may ty in some cases of oral LP (33, 34), Epithelial cells be either CD8 + (MHC Class I rethe exact contribution of cell-mediated stricted) or CD4-|-(MHC Class II reIn view of the similarities between oral autoimmunity remains to be deter- LP and the DTH reaction (26-28), and stricted). mined. Because of their extensive dendrites, LC are ideally positioned to respond to antigens which bind to keratinocytes (contact sensitizers, allergens, drugs) or are expressed on the keratinoeyte surfaee membrane (MHC and viral antigens). In this context, the eell mediated response initiated by LC will likely include cytotoxic cells directed against keratinocytes, as seen in GVHD and severe DTH reactions (28). Involvement of viral antigens in oral LP has been suggested (see Ref. 1 for review). The joint observations that mucosal nerve Evolution of oral lichen planus
Fig. 1. Reticular lesions of oral lichen planus distributed on the buccal mucosa of a 68-yrold woman. Table 1. Factors which precipitate or aggra vate oral lichen planus Systemic medications Non-steroidal anti-inflammatory drugs, Antihyperlcnsivc agents, gold salts, Angiotensin-converting enzyme Inhibitors (I, 4, 7-10) Local factors Dental materials Amalgam/metals (10-11) Composite resin (12) Cyanoacrylate (13) Trauma Surgery (14) Stress (15, 16) Viral infection (I)
'A^Hi f/g. 2. Histology of oral LP. A, dense subepithelial mononuclear cell infiltrate is present. X 10. B, lymphocytes in lamina propria and epithelium are a conspicuous feature. Basal cell liquefaction foci are apparent, x 40.
Fig. 3. Erosive lichenoid lesions affecting gingiva and labial mucosa in patient with chronie graft-versus-host disease. This individual received bone marrow transplant from HLAidentical sibling during treatment for acute myeloid leukemia.
Oral lichen planus
391
fibers closely appose basal keratinocytes (Fig. 4B), and that keratinoeyte expression of receptors (CD21) for Epstein Barr virus is upregulated in lesions of oral LP (Fig. 4A) suggest that low-grade and/or persistent infection of epithelial cells with Herpes group viruses may be a possible etiologic factor in LP.
ri^iffrsi^?
Mast cells
In addition to their association with vascular elements and epithelial cells, there is accumulating evidence that mucosal nerves closely appose mast cells (43^5). Chronic stimulation of these nerves, together with the release of neuropeptides such as substance P, induce mast cell degranulation (46-47). The significance of this event in altering leukocyte trafficking is seen in the ohservation that mast cell degranulation induces expression of endothelial leukocyte adhesion molecule-l (ELAM-1), a glycoprotein which is critically required for leukocyte adhesion to the luminal surface of endothelial cells (48). Thus, mast cells serve as gatekeepers of the microvascular bed, and provide a link between neural networks and immune cell trafficking (48, 49). Mast cells contain immunoreactive tumor necrosis factor alpha (TNF-a) (Fig. 4C), which is released upon degranulation. This cytokine upregulates the expression of endothelial eell adhesion molecules, including ELAM-1 and ICAM-l (Fig. 4D, E), together with the recently described vascular cell adhesion molecule-1 (VCAM1) (50, 51). In oral LP, mast cell degranulation (Fig. 4 F-H) may be promoted by drugs implicated in lichenoid eruptions, such as thiazides (52), or by neuropeptides secreted from nerves affeeted fig. '#. Immunohistology of oral LP, showing by electrical potentials (galvanic efcells involved in putative induction phase. All sections except F and G stained using fects), trauma (surgery), infection (Herimmunoperoxidase. A, CD21 (EBV receptor) pes group viruses), or psychologic expression by keratinocytes in oral LP. AntiCR2 antibody. X 40. B. mucosal nerves (ar- stress. In this fashion, the interaction rows) in close proximity to basal keratinobetween nerves and mast cells may pro. _ _ ^ « . . . «_ cytes identified by expression of neural cell vide a common pathway through which A . . . J j m l . . ^ p * - •*"'^^'*-^^ adhesion molecule (NCAM). NKH-I certain agents known to aggravate LP '• t " , " » • • • • : ' ..3».. .^Af^ antibody, X40. C, immunoreaetive tumor neexert their effect. Release of cytokines crosis factor in mast cells in normal human such as TNF-a from mast cells and conoral mucosa. B154.2 antibody, x 100. D, expression of intercellular adhesion molecule-l sequential induction of adhesion mole(ICAM-l) on endothelial cells in oral LP cule expression may therefore comprise Anti-ICAM antibody, x 40. E. expression of an important event in the initiation of endothelial leukocyte adhesion molecule-l the lymphocytic infiltration which char(ELAM-1) on endothelial cells in orat LP. «
