Ecoclimber
Senior Member
- Messages
- 1,011
Arthritis Rheumatol. 2014 Sep;66(9):2545-57. doi: 10.1002/art.38726.
Implication of Epstein-Barr Virus Infection in Disease-Specific Autoreactive B Cell Activation in Ectopic Lymphoid Structures of Sjögren's Syndrome.
Croia C1, Astorri E, Murray-Brown W, Willis A, Brokstad KA, Sutcliffe N, Piper K, Jonsson R, Tappuni AR, Pitzalis C, Bombardieri M.
Author information
Abstract
Objective
To examine whether the B cell tropic γ-herpesvirus Epstein-Barr virus (EBV) is aberrantly expressed in its latent and lytic forms within ectopic lymphoid structures (ELS) in the salivary glands of patients with Sjögren's syndrome (SS), and to investigate the relationship between EBV dysregulation, B cell activation, in situ differentiation of autoreactive plasma cells, disease-specific autoantibody production, and cytotoxicity.
Sjögren's syndrome (SS) is characterized by lymphocytic infiltration of salivary and lacrimal glands leading to xerostomia (dry mouth) and keratoconjunctivitis sicca (dry eyes) ([1]). One of the hallmarks of SS is the formation of ectopic lymphoid structures (ELS) in the salivary glands. ELS are defined as B cell/T cell follicles with networks of stromal follicular dendritic cells (FDCs), supporting an ectopic germinal center (GC) reaction ([2-4]). ELS, which develop in ∼30–40% of patients with SS, represent functional niches whereby B cells undergo in situ activation and differentiation into autoreactive anti-Ro/La–producing plasma cells ([5]) and are strongly linked to the development of B cell lymphomas ([6, 7]).
Although the pathogenesis of SS is still unclear, it has long been postulated that viruses, including Epstein-Barr virus (EBV), human T lymphotropic virus type I, and cocksackievirus ([8-10]), play an essential role in promoting the autoimmune dysregulation typical of the disease, possibly as a result of genetic susceptibility involving the type I interferon pathway ([11, 12]). EBV, a human DNA γ-herpesvirus that establishes an asymptomatic latent infection and can cause infectious mononucleosis ([13]), has long been postulated to promote autoimmunity in several chronic autoimmune diseases ([10, 14]), primarily due to its ability to infect B cells and alter their normal differentiation program, leading to the survival of autoreactive B cell clones ([15]).
There is significant controversy concerning the question of whether EBV is implicated in the pathogenesis of SS, due to conflicting results regarding the expression of EBV proteins/nucleic acids in the salivary glands of patients with SS compared with patients with sicca or healthy control subjects. Some studies demonstrated increased EBV expression in patients with SS ([16-19]), and others failed to show differential expression between SS and control salivary glands, despite common evidence of the persistence of EBV in the salivary glands ([20-22]).
Notably, none of the previous studies attempted to investigate EBV expression in the salivary glands of patients with SS in the context of ectopic GC-like structures. This is highly relevant not only because the “GC model” is widely used to explain how EBV gains access to the memory B cell compartment ([23, 24]), but also because ectopic B cell follicles in chronic autoimmune diseases represent preferential niches of EBV latency and reactivation ([25-27]). Furthermore, despite the strong evidence linking EBV and B cell autoimmunity, it remains unclear whether EBV triggers a breach of tolerance toward disease-specific antigens within ELS.
The purpose of this study was to investigate whether the expression of latent and lytic EBV occurs preferentially in the salivary glands of patients with SS, in the context of GC-like structures and in situ differentiation of autoreactive plasma cells with specificity toward SS-associated autoantigens.
METHODS:
Latent and lytic EBV infection in the salivary glands of 28 patients with SS and 38 patients with nonspecific chronic sialadenitis (NSCS), characterized for the presence or absence of ELS, was investigated by reverse transcription-polymerase chain reaction, in situ hybridization, and immunohistochemical/immunofluorescence staining.
Glandular versus synovial production of anti-Ro 52, anti-citrullinated protein antibodies (ACPAs), and anti-EBV peptide antibodies was analyzed in situ or in vivo in human SS/SCID and human rheumatoid arthritis/SCID mouse chimeras.
RESULTS:
EBV dysregulation within inflammatory infiltrates was observed exclusively in ELS+ SS salivary gland tissue, as revealed by latent EBV infection and lytic EBV infection in B cells and plasma cells, respectively.
Conversely, epithelial latent membrane protein 2A expression was observed in both patients with SS and patients with NSCS. Importantly, perifollicular plasma cells displaying Ro 52 immunoreactivity were frequently infected by EBV. Furthermore, ELS-containing SS salivary gland tissue that was transplanted into SCID mice supported the production of anti-Ro 52/anti-La 48 and anti-EBV antibodies but not ACPAs. Analysis of CD4+ and CD8+ T cell localization and granzyme B expression demonstrated that the persistence of EBV in ELS-containing SS salivary glands was associated with follicular exclusion of CD8+ T cells and impaired CD8-mediated cytotoxicity.
DISCUSSION:
Here, we report the novel finding that a dysregulated EBV infection, characterized by the expression of both the latent and lytic forms of EBV, is a specific feature of the salivary glands of patients with SS displaying ectopic B cell follicles organized as ELS. We show that LMP-2A, a protein expressed during EBV latency that mimics the B cell receptor signaling pathway ([36]), is selectively expressed by CD20+ B cells within B cell follicles, particularly in association with FDC networks.
Furthermore, using both in situ hybridization and quantitative PCR, we show that the expression of EBERs, which are short nonpolyadenylated and nontranslated RNAs expressed during EBV latency ([13]), is strictly dependent on the formation of ELS in SS salivary glands but is virtually absent in ELS− SS or NSCS salivary glands.
Specifically, EBERs closely correlated with CD19 mRNA, a cell marker expressed by B cells and plasma cells, and AID, the enzyme required for functional B cell activation leading to somatic hypermutation and class-switch recombination of the immunoglobulin genes ([37, 38]). Accordingly, in situ hybridization for EBER demonstrated colocalization of EBER with CD20+ B cells and perifollicular CD138+ plasma cells. Thus, it appears that latent EBV infection in SS salivary glands is associated exclusively with the presence of ELS with functional GC-like structures supporting in situ B cell activation and plasma cell differentiation.
The close association of EBV with an ectopic GC response is extremely relevant, because EBV uses a GC-like growth program, which includes the expression of LMPs and EBERs, to transform latently infected B cells into proliferating blasts and convert these cells into long-lived memory B cells. Further signals, normally dependent on antigen stimulation and T cell help, favor plasma cell differentiation leading to viral reactivation ([13, 24, 39]).
Accordingly, by investigating the expression of EBV proteins involved in lytic reactivation, such as BFRF1, we demonstrated that EBV reactivation is frequently observed in ELS+ but not ELS− SS or NSCS salivary glands, where BFRF1 expression is observed exclusively in CD138+ plasma cells surrounding FDC+ B cell follicles, suggesting local differentiation within the GC ([40]). Thus, salivary gland ELS, by providing B cell help in the context of an ectopic GC response characterized by the expression of strong B cell−activating signals, such as BAFF, represent protected niches for EBV latency and reactivation. Further supporting the importance of the SS salivary gland microenvironment for EBV reactivation is the observation that saliva from patients with SS directly reactivates EBV by inducing BZLF1, an early lytic-phase EBV protein ([41]).
Overall, our work also reconciles conflicting data regarding the expression of EBV in the salivary glands of patients with SS compared with healthy individuals or patients with sicca, because previous studies failed to stratify SS salivary glands according to the presence of ectopic GC-like structures ([17-22]).
In particular, previous work primarily focused on EBV infection of salivary gland epithelial cells, which is a physiologic site of EBV latency, as we also confirmed by demonstrating LMP-2A infection in ductal (mainly) and acinar epithelial cells of SS (ELS+ and ELS−) and NSCS salivary gland tissue. Nevertheless, it is conceivable that epithelial cell infection by EBV could directly contribute to the pathogenesis of SS in genetically susceptible individuals. In this regard, aberrant type I interferon responses due to insertion/deletion and single-nucleotide polymorphisms in related genes such as IRF5 and STAT4 ([11, 12]) might result in an aberrant immune response to EBV whereby persistent viral infection may culminate in the formation of ELS. Accordingly, we recently showed that ELS can be triggered by viral infection of the submandibular glands in a novel murine model of SS-like sialadenitis ([42]).
A fundamental question regarding EBV reactivation within ELS is whether this process takes place in plasma cells bearing an autoreactive B cell receptor, thus linking EBV with the development of autoimmunity, a hypothesis that has long been formulated ([15]) but eluded formal demonstration even with sophisticated techniques ([43]).
Here, by analyzing the immunoreactivity of lesional CD138+ cells toward Ro 52, a major SS autoantigen, we show that a significant proportion of perifollicular plasma cells lytically infected with EBV display an autoreactive phenotype. Strikingly, we also show that the autoreactive profile of EBV-infected plasma cells completely differs between ELS+ SS salivary glands and ELS+ RA synovium, whereby CD138+/BFRF1+ cells are specific for Ro 52 in SS and for citrullinated antigens in RA ([27]) but not vice versa. These observations, replicated in vivo by engrafting human ELS+ SS parotid gland tissue and RA synovial tissue into SCID mice, strongly reinforce the hypothesis that in situ autoimmunity in the context of EBV reactivation within ELS is driven by local antigenic stimuli that are disease specific. Additionally, the inability of ELS to prevent autoreactive B cells from entering GCs is likely to favor the differentiation of high-affinity autoreactive plasma cells from EBV-infected B cells ([44]), which are maintained within the salivary gland microenvironment by strong plasma cell survival factors ([45]).
Another original observation in the SS/SCID mouse chimera model is that engrafted SS salivary glands with ELS sustained the production of human anti-EBV IgG antibodies, namely anti–EBNA-1 and anti-VCA, which closely associated with local anti-Ro/La production. This suggests that a humoral anti-EBV immune response takes place within ELS in the salivary glands of patients with SS and is closely linked with autoimmunity. This is particularly interesting not only because the anti-EBV antibody level is significantly increased in the serum of patients with SS ([46, 47]), but also because of molecular mimicry between SS autoantigens and EBV proteins. Interestingly, anti-Ro/La autoantibodies precipitate proteins that are complexed with EBERs ([48]), and EBNA-1 mimics Ro 52 ([49]). Moreover, cross-reactivity between α-fodrin, lipocardin, and EBV EA proteins was also described ([50]). Finally, EBV can induce the cleavage of α-fodrin, leading to induction of the antigenic 120-kd α-fodrin ([51]). Thus, our present data allow us to speculate that EBV may promote humoral autoimmunity within ELS not only by providing a survival/proliferative advantage to autoreactive B cells entering ectopic GC but also by promoting breach of self tolerance via molecular mimicry or posttranslational modification of SS-associated antigens.
Investigation of the cytotoxic immune response in SS salivary glands suggested that the dysregulated EBV infection observed in patients with ELS+ salivary glands may be favored by the follicular exclusion of cytotoxic CD4+/granzyme B–positive T cells, which accounted for only a minority of granzyme B–positive cells within B cell follicles (but represented >80% of cytotoxic cells in ELS− SS and NSCS salivary glands). In ELS+/EBV+ samples, an unusual population of CD4+/granzyme B–positive cells seemed to substitute for cytotoxic CD8+ T cells; this observation is very interesting, because circulating cytotoxic CD4+ T cells have been described in chronic viral infections, including EBV ([52]).
In summary, the current study demonstrates that in salivary gland ELS, intimate EBV–immune cell interactions take place, with profound repercussions on the aberrant B cell activation and local autoimmunity typically observed in patients with SS. Of critical relevance, it also demonstrates that this process is disease specific, indicating that following breach of tolerance for specific putative autoantigens, EBV infection contributes to the survival and perpetuation of autoreactive B cells.
Implication of Epstein-Barr Virus Infection in Disease-Specific Autoreactive B Cell Activation in Ectopic Lymphoid Structures of Sjögren's Syndrome.
Croia C1, Astorri E, Murray-Brown W, Willis A, Brokstad KA, Sutcliffe N, Piper K, Jonsson R, Tappuni AR, Pitzalis C, Bombardieri M.
Author information
Abstract
Objective
To examine whether the B cell tropic γ-herpesvirus Epstein-Barr virus (EBV) is aberrantly expressed in its latent and lytic forms within ectopic lymphoid structures (ELS) in the salivary glands of patients with Sjögren's syndrome (SS), and to investigate the relationship between EBV dysregulation, B cell activation, in situ differentiation of autoreactive plasma cells, disease-specific autoantibody production, and cytotoxicity.
Sjögren's syndrome (SS) is characterized by lymphocytic infiltration of salivary and lacrimal glands leading to xerostomia (dry mouth) and keratoconjunctivitis sicca (dry eyes) ([1]). One of the hallmarks of SS is the formation of ectopic lymphoid structures (ELS) in the salivary glands. ELS are defined as B cell/T cell follicles with networks of stromal follicular dendritic cells (FDCs), supporting an ectopic germinal center (GC) reaction ([2-4]). ELS, which develop in ∼30–40% of patients with SS, represent functional niches whereby B cells undergo in situ activation and differentiation into autoreactive anti-Ro/La–producing plasma cells ([5]) and are strongly linked to the development of B cell lymphomas ([6, 7]).
Although the pathogenesis of SS is still unclear, it has long been postulated that viruses, including Epstein-Barr virus (EBV), human T lymphotropic virus type I, and cocksackievirus ([8-10]), play an essential role in promoting the autoimmune dysregulation typical of the disease, possibly as a result of genetic susceptibility involving the type I interferon pathway ([11, 12]). EBV, a human DNA γ-herpesvirus that establishes an asymptomatic latent infection and can cause infectious mononucleosis ([13]), has long been postulated to promote autoimmunity in several chronic autoimmune diseases ([10, 14]), primarily due to its ability to infect B cells and alter their normal differentiation program, leading to the survival of autoreactive B cell clones ([15]).
There is significant controversy concerning the question of whether EBV is implicated in the pathogenesis of SS, due to conflicting results regarding the expression of EBV proteins/nucleic acids in the salivary glands of patients with SS compared with patients with sicca or healthy control subjects. Some studies demonstrated increased EBV expression in patients with SS ([16-19]), and others failed to show differential expression between SS and control salivary glands, despite common evidence of the persistence of EBV in the salivary glands ([20-22]).
Notably, none of the previous studies attempted to investigate EBV expression in the salivary glands of patients with SS in the context of ectopic GC-like structures. This is highly relevant not only because the “GC model” is widely used to explain how EBV gains access to the memory B cell compartment ([23, 24]), but also because ectopic B cell follicles in chronic autoimmune diseases represent preferential niches of EBV latency and reactivation ([25-27]). Furthermore, despite the strong evidence linking EBV and B cell autoimmunity, it remains unclear whether EBV triggers a breach of tolerance toward disease-specific antigens within ELS.
The purpose of this study was to investigate whether the expression of latent and lytic EBV occurs preferentially in the salivary glands of patients with SS, in the context of GC-like structures and in situ differentiation of autoreactive plasma cells with specificity toward SS-associated autoantigens.
METHODS:
Latent and lytic EBV infection in the salivary glands of 28 patients with SS and 38 patients with nonspecific chronic sialadenitis (NSCS), characterized for the presence or absence of ELS, was investigated by reverse transcription-polymerase chain reaction, in situ hybridization, and immunohistochemical/immunofluorescence staining.
Glandular versus synovial production of anti-Ro 52, anti-citrullinated protein antibodies (ACPAs), and anti-EBV peptide antibodies was analyzed in situ or in vivo in human SS/SCID and human rheumatoid arthritis/SCID mouse chimeras.
RESULTS:
EBV dysregulation within inflammatory infiltrates was observed exclusively in ELS+ SS salivary gland tissue, as revealed by latent EBV infection and lytic EBV infection in B cells and plasma cells, respectively.
Conversely, epithelial latent membrane protein 2A expression was observed in both patients with SS and patients with NSCS. Importantly, perifollicular plasma cells displaying Ro 52 immunoreactivity were frequently infected by EBV. Furthermore, ELS-containing SS salivary gland tissue that was transplanted into SCID mice supported the production of anti-Ro 52/anti-La 48 and anti-EBV antibodies but not ACPAs. Analysis of CD4+ and CD8+ T cell localization and granzyme B expression demonstrated that the persistence of EBV in ELS-containing SS salivary glands was associated with follicular exclusion of CD8+ T cells and impaired CD8-mediated cytotoxicity.
DISCUSSION:
Here, we report the novel finding that a dysregulated EBV infection, characterized by the expression of both the latent and lytic forms of EBV, is a specific feature of the salivary glands of patients with SS displaying ectopic B cell follicles organized as ELS. We show that LMP-2A, a protein expressed during EBV latency that mimics the B cell receptor signaling pathway ([36]), is selectively expressed by CD20+ B cells within B cell follicles, particularly in association with FDC networks.
Furthermore, using both in situ hybridization and quantitative PCR, we show that the expression of EBERs, which are short nonpolyadenylated and nontranslated RNAs expressed during EBV latency ([13]), is strictly dependent on the formation of ELS in SS salivary glands but is virtually absent in ELS− SS or NSCS salivary glands.
Specifically, EBERs closely correlated with CD19 mRNA, a cell marker expressed by B cells and plasma cells, and AID, the enzyme required for functional B cell activation leading to somatic hypermutation and class-switch recombination of the immunoglobulin genes ([37, 38]). Accordingly, in situ hybridization for EBER demonstrated colocalization of EBER with CD20+ B cells and perifollicular CD138+ plasma cells. Thus, it appears that latent EBV infection in SS salivary glands is associated exclusively with the presence of ELS with functional GC-like structures supporting in situ B cell activation and plasma cell differentiation.
The close association of EBV with an ectopic GC response is extremely relevant, because EBV uses a GC-like growth program, which includes the expression of LMPs and EBERs, to transform latently infected B cells into proliferating blasts and convert these cells into long-lived memory B cells. Further signals, normally dependent on antigen stimulation and T cell help, favor plasma cell differentiation leading to viral reactivation ([13, 24, 39]).
Accordingly, by investigating the expression of EBV proteins involved in lytic reactivation, such as BFRF1, we demonstrated that EBV reactivation is frequently observed in ELS+ but not ELS− SS or NSCS salivary glands, where BFRF1 expression is observed exclusively in CD138+ plasma cells surrounding FDC+ B cell follicles, suggesting local differentiation within the GC ([40]). Thus, salivary gland ELS, by providing B cell help in the context of an ectopic GC response characterized by the expression of strong B cell−activating signals, such as BAFF, represent protected niches for EBV latency and reactivation. Further supporting the importance of the SS salivary gland microenvironment for EBV reactivation is the observation that saliva from patients with SS directly reactivates EBV by inducing BZLF1, an early lytic-phase EBV protein ([41]).
Overall, our work also reconciles conflicting data regarding the expression of EBV in the salivary glands of patients with SS compared with healthy individuals or patients with sicca, because previous studies failed to stratify SS salivary glands according to the presence of ectopic GC-like structures ([17-22]).
In particular, previous work primarily focused on EBV infection of salivary gland epithelial cells, which is a physiologic site of EBV latency, as we also confirmed by demonstrating LMP-2A infection in ductal (mainly) and acinar epithelial cells of SS (ELS+ and ELS−) and NSCS salivary gland tissue. Nevertheless, it is conceivable that epithelial cell infection by EBV could directly contribute to the pathogenesis of SS in genetically susceptible individuals. In this regard, aberrant type I interferon responses due to insertion/deletion and single-nucleotide polymorphisms in related genes such as IRF5 and STAT4 ([11, 12]) might result in an aberrant immune response to EBV whereby persistent viral infection may culminate in the formation of ELS. Accordingly, we recently showed that ELS can be triggered by viral infection of the submandibular glands in a novel murine model of SS-like sialadenitis ([42]).
A fundamental question regarding EBV reactivation within ELS is whether this process takes place in plasma cells bearing an autoreactive B cell receptor, thus linking EBV with the development of autoimmunity, a hypothesis that has long been formulated ([15]) but eluded formal demonstration even with sophisticated techniques ([43]).
Here, by analyzing the immunoreactivity of lesional CD138+ cells toward Ro 52, a major SS autoantigen, we show that a significant proportion of perifollicular plasma cells lytically infected with EBV display an autoreactive phenotype. Strikingly, we also show that the autoreactive profile of EBV-infected plasma cells completely differs between ELS+ SS salivary glands and ELS+ RA synovium, whereby CD138+/BFRF1+ cells are specific for Ro 52 in SS and for citrullinated antigens in RA ([27]) but not vice versa. These observations, replicated in vivo by engrafting human ELS+ SS parotid gland tissue and RA synovial tissue into SCID mice, strongly reinforce the hypothesis that in situ autoimmunity in the context of EBV reactivation within ELS is driven by local antigenic stimuli that are disease specific. Additionally, the inability of ELS to prevent autoreactive B cells from entering GCs is likely to favor the differentiation of high-affinity autoreactive plasma cells from EBV-infected B cells ([44]), which are maintained within the salivary gland microenvironment by strong plasma cell survival factors ([45]).
Another original observation in the SS/SCID mouse chimera model is that engrafted SS salivary glands with ELS sustained the production of human anti-EBV IgG antibodies, namely anti–EBNA-1 and anti-VCA, which closely associated with local anti-Ro/La production. This suggests that a humoral anti-EBV immune response takes place within ELS in the salivary glands of patients with SS and is closely linked with autoimmunity. This is particularly interesting not only because the anti-EBV antibody level is significantly increased in the serum of patients with SS ([46, 47]), but also because of molecular mimicry between SS autoantigens and EBV proteins. Interestingly, anti-Ro/La autoantibodies precipitate proteins that are complexed with EBERs ([48]), and EBNA-1 mimics Ro 52 ([49]). Moreover, cross-reactivity between α-fodrin, lipocardin, and EBV EA proteins was also described ([50]). Finally, EBV can induce the cleavage of α-fodrin, leading to induction of the antigenic 120-kd α-fodrin ([51]). Thus, our present data allow us to speculate that EBV may promote humoral autoimmunity within ELS not only by providing a survival/proliferative advantage to autoreactive B cells entering ectopic GC but also by promoting breach of self tolerance via molecular mimicry or posttranslational modification of SS-associated antigens.
Investigation of the cytotoxic immune response in SS salivary glands suggested that the dysregulated EBV infection observed in patients with ELS+ salivary glands may be favored by the follicular exclusion of cytotoxic CD4+/granzyme B–positive T cells, which accounted for only a minority of granzyme B–positive cells within B cell follicles (but represented >80% of cytotoxic cells in ELS− SS and NSCS salivary glands). In ELS+/EBV+ samples, an unusual population of CD4+/granzyme B–positive cells seemed to substitute for cytotoxic CD8+ T cells; this observation is very interesting, because circulating cytotoxic CD4+ T cells have been described in chronic viral infections, including EBV ([52]).
In summary, the current study demonstrates that in salivary gland ELS, intimate EBV–immune cell interactions take place, with profound repercussions on the aberrant B cell activation and local autoimmunity typically observed in patients with SS. Of critical relevance, it also demonstrates that this process is disease specific, indicating that following breach of tolerance for specific putative autoantigens, EBV infection contributes to the survival and perpetuation of autoreactive B cells.