SEARCH

SEARCH BY CITATION

Keywords:

  • Sjögren's syndrome;
  • B lymphocytes;
  • BAFF ;
  • Flt3 ligand;
  • non-Hodgkin's lymphoma

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Criteria and new tools for the diagnosis
  5. B-lymphocyte subpopulations
  6. Intrinsic B-cell defects in primary SS
  7. Disturbances of cytokines in primary SS
  8. B-cell proliferation in primary SS
  9. Anti-B-cell-targeted therapy in primary SS
  10. Conclusion
  11. Author contributions
  12. References

The autoimmune exocrinopathy Sjögren's syndrome (SS) is characterized by mononuclear cell (MNC) infiltrates of exocrine glands and overactivity of B lymphocytes. Although T cells have long been perceived as the prime effectors, increasing evidence indicates that the key role is rather served by B cells. Among related abnormalities are rheumatoid factor (RF), anti-SSA/Ro, and anti-SSB/La antibodies (Ab). Also, supporting this view is our finding of an increase in the number of circulating naïve mature B (Bm) cells, with a reciprocal decrease in that of memory B cells. Furthermore, a ratio of Bm2-plus-Bm2′ cells to early Bm5-plus-late Bm5 above 5 is diagnostic. This variation partly reflects the migration of active memory B cells into the exocrine glands of the patients, as well as into their skin. More recently, the B-cell-activating factor of the TNF family (BAFF) has been endorsed with a pivotal role in B-cell survival and hence implicated in the pathogenesis of autoimmunity. In practice, B cells have turned quite attractive as a target for biotherapy. For example, treatment with anti-CD20 Ab has afforded some benefits in this disease, while BAFF blockers are still on the way, but should expand our armamentarium for treating SS. With such B-cell-directed biotherapies in mind, we delineate herein the distinguishing traits of B lymphocytes in SS.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Criteria and new tools for the diagnosis
  5. B-lymphocyte subpopulations
  6. Intrinsic B-cell defects in primary SS
  7. Disturbances of cytokines in primary SS
  8. B-cell proliferation in primary SS
  9. Anti-B-cell-targeted therapy in primary SS
  10. Conclusion
  11. Author contributions
  12. References

Among the diseases often missed in oral medicine is Sjögren's syndrome (SS), coined as an autoimmune exocrinopathy by Moutsopoulos in 1994 (Moutsopoulos, 1994). It may occur alone as primary SS or against a background of connective disease as secondary SS (Moutsopoulos et al, 1979). Although xerostomia resulting from the involvement of salivary glands (SG), and xerophthalmia due to that of lacrimal glands, are usually prominent (Figure 1), this nosological entity refers to a number of multifaceted presentations with a large variety of clinical complications and biological manifestations. Indeed, its spectrum extends from an organ-specific disorder through a systemic process that threatens the musculoskeletal and nervous systems or the lungs, kidneys, and blood vessels (Tzioufas and Voulgarelis, 2007). Another trait of primary SS is the profusion of anti-self-antibodies (Abs) (Sheldon, 2004), for example, RF, anti-sicca syndrome (SS)A/Ro and anti-SSB/La Ab, IgA-containing immune complexes (Bendaoud et al, 1991), and hypersialylation of IgA (Basset et al, 2000). The patients display a biased distribution of mature B-lymphocyte subsets in the peripheral blood (PB), and a group of them are at risk of developing B-cell non-Hodgkin's lymphoma (NH). Such polymorphism explains the delay in the diagnosis, associated with underestimation of the patients' concerns made by the dentist, which discourages patients from taking medical advice as oral symptoms arise. There is, thus, every likelihood that the prevalence of primary SS is far higher than estimated. Clearly, lymphocytes are influential in the disease, but controversy persists over which type of lymphocytes, T or B, initiates its happening. Beyond the paradigm that T cells maintain strict control over B cells, new evidence has sparked a great deal of interest in the possibility that the leading part is played by these cells (Cornec et al, 2012; Youinou et al, 2012). Not only do they produce auto-Abs, but they also accomplish various tasks, such as the synthesis of a flurry of cytokines and the presentation of antigen (Ag). Additional breakthroughs have occurred in their study, including the dissection of B-cell developmental stages, the profiling of B-cell-related gene expression using microarray technology (Devauchelle-Pensec et al, 2010), and the discovery of two ligands for the TNF receptor family: one is the B-cell-activating factor (BAFF) and the other a proliferation-inducing ligand (APRIL).

image

Figure 1. Clinical manifestations of xerophthalmia and xerostomia in patient with primary Sjögren's syndrome

Download figure to PowerPoint

In sum, peri-epithelial MNC infiltration of damaged exocrine glands and hyperactivity of B lymphocytes document the autoimmune nature of primary SS. Thus, with B-cell-targeted therapy as a hint, we venture into the centrality of B cells in the pathogenesis and analyze the process that underpins their proliferation in primary SS.

Criteria and new tools for the diagnosis

  1. Top of page
  2. Abstract
  3. Introduction
  4. Criteria and new tools for the diagnosis
  5. B-lymphocyte subpopulations
  6. Intrinsic B-cell defects in primary SS
  7. Disturbances of cytokines in primary SS
  8. B-cell proliferation in primary SS
  9. Anti-B-cell-targeted therapy in primary SS
  10. Conclusion
  11. Author contributions
  12. References

Availability of potentially active drugs, most notably anti-lymphocyte monoclonal Abs (mAb), has made a priority of the early diagnosis of primary SS. The trouble is that such an issue should not be simple to address, given that as many as 11 criterion sets have hitherto been published for the classification of the disease. The latest set, still in use, has been defined by the American-European Community Study Group (Vitali et al, 2002) and subsequently revised by the Sjögren's International Collaborative Clinical Alliance (Shiboski et al, 2012). Both sets are based upon subjective complaints of dryness and objective marks of dysfunction.

One may easily guess that the diagnosis proves difficult in some patients and delayed accordingly. Interestingly, plenty of data have brought about new criteria, prompting the suggestion that they might be of help in the diagnosis, for example, measurement of an ever-growing number of salivary biomarkers (Hu et al, 2010), ultrasonography of the SG (Cornec et al, 2013), and possibly, abnormalities of the distribution of circulating B-lymphocyte subsets and/or specificities of the SG lymphoid infiltrate.

B-lymphocyte subpopulations

  1. Top of page
  2. Abstract
  3. Introduction
  4. Criteria and new tools for the diagnosis
  5. B-lymphocyte subpopulations
  6. Intrinsic B-cell defects in primary SS
  7. Disturbances of cytokines in primary SS
  8. B-cell proliferation in primary SS
  9. Anti-B-cell-targeted therapy in primary SS
  10. Conclusion
  11. Author contributions
  12. References

Ontogenic classification

Dating back into the 1980s, a T-cell marker has been discovered on the B cells from patients with chronic lymphocytic leukemia (Boumsell et al, 1980), subsequently identified as CD5, and detected on a fraction of normal B lymphocytes. Accordingly, the B-lymphocyte population has been subdivided into a major CD5-non-expressing B2 subpopulation and a minor CD5-expressing B1 subpopulation. These latter cells accumulate in leukemia and produce multispecific Ab, directed, among other targets, against non-organ-specific auto-Ag.

The novelty in the interpretation of the B-cell abnormalities in the PB and SG is to trace their ontogeny (Table 1). They originate in the bone marrow (BM) as pro-B cells that end up by rearranging their immunoglobulin (Ig) genes productively and thereby evolve to pre-B cells. In turn, they exit the BM, and, settle down in the spleen, to generate type-1 (BT1) and type-2 (BT2) transitional B lymphocytes (Palanichamy et al, 2009). At the start, BT1 cells present as CD20+ CD5+ CD10+/− CD21+/− CD23+/− IgM+/− IgD+ CD38+, and once their cognate Ag encountered, they differentiate into BT2 cells as CD20+ CD5+/− CD21++ CD23+/− IgM++ IgD++ CD38+/−. Then, they shift into circulating lymphocytes that get organized in the lymphoid follicles (FO) as germinal centers (GC) or progress as non-circulating B lymphocytes doomed to populate the splenic marginal zone (MZ).

Table 1. Summary of the main phenotypic features of human B cells during their ontogeny
Transitional type 1CD20+CD5+CD10+/−CD21+/−CD23+/−IgM+IgD+CD38+
Transitional type 2CD20+CD5+/−CD21highCD23+/−IgMhighIgDhighCD38+/− 
Mature naïve B cellsCD20+CD10CD24+/−CD23+IgD+CD27CD38+/− 
Marginal zone B cellsCD20+CD5CD23CD21highIgMhighIgD+/−CD1chigh 
Germinal center B cellsCD20+IgDCD38+CD10+CD27IgM  

The strength of B-cell Ag receptor (BCR)-evoked signals (Saito et al, 2003) settles whether immature BT2 cells move to the FO or to the MZ (Saito et al, 2003). In there, maturation of a GC launches an intense proliferation phase. Based on the relative expression of IgD and CD38 (Pascual et al, 1994), developmental stages of mature B (Bm) lymphocytes follow one another. The sequence begins at the border of GC with IgD+ CD38+ naïve Bm1 cells and terminates at their egress with IgD CD38+++ plasmablasts or IgD CD38+ early-memory Bm5 cells evolving locally to IgD+ CD38 late-memory Bm5 cells or IgD CD38+++ plasmablasts (Odendahl et al, 2000). They get back to the BM where they differentiate into long-lived plasma cells (PC). Throughout this ontogenesis, a few specimens of each subset leak into the circulation.

Should the second option be taken, MZ B cells may express mutated Ig genes, but lack activation-induced cytidine deaminase (AID), suggesting that they have passed through a GC (Willenbrock et al, 2005). On the other hand, expression of sphingosine-1-phosphate receptor-1 may retain B lymphocytes within the MZ (Cinamon et al, 2004), by overcoming their attraction toward the GC by CXCL13 (Reif et al, 2002). This chemokine is expressed inside the FO (Amft and Bowman, 2001), and its receptor upregulated (Hansen et al, 2005) in B cells of primary SS patients (Figure 2). Also, the Toll-like receptors (TLR) are likely to be involved in the process. Originally described as receptors that bind microbial Ag, they proved to bind auto-Ag as well (Marshak-Rothstein, 2006). It may therefore be anticipated that innate TLR-dependent pathways, most notably those initiated by TLR9 on B cells, are required for tolerance checkpoints to be efficient. In this respect, we have shown that TLR participate in the development of MZ B lymphocytes from immature BT2 cells (Guerrier et al, 2012b). Also, interesting is our finding of mRNA for TLR9 in clusters of B cells infiltrating the SG of primary SS patients (Guerrier et al, 2012a).

image

Figure 2. B-cell classification according to their ontogenic state. From the transitional type 1 (T1) and T2 B cells, two options depend on the B-cell receptor (BCR)-evoked signal and the downstream Notch 2 proteins: germinal center (GC) B cells driven by the B-cell-attracting (BCA)-1 chemokine (or CXCL13) and marginal zone B cells with mutations but without activation-induced cytidine deaminase (AID)

Download figure to PowerPoint

Odd distribution of B-cell subsets in the blood of SS patients

It has long been known that in SS patients, the number of B1 lymphocytes is elevated in the PB as well as the SG, particularly in the presence of a monoclonal Ig in the serum (Youinou et al, 1988). By being translocated to the membrane, the CD5 molecules convey the SH2-containing phosphatase 1 to the vicinity of the BCR. Because the threshold of the response is raised, potentially deleterious cells do not proliferate, but are induced into anergy. Moreover, CD5 brings the recombination-activating gene (RAG)-encoded enzymes into play and thereby enables rearrangement of V, D, and J segments for the BCR heavy and light chains. In so doing, RAG enables the random emergence of autoreactive B cells (Hillion et al, 2005).

It has since been established that in addition to the disturbance in the B1 and B2 compartments, most B-cell subsets are disorganized. In particular, memory B cells escape the circulation (Hansen et al, 2002) and accumulate inside epithelial organs, in position to establish ectopic GC (Daridon et al, 2006). In contrast, the proportions of transitional mature (but naïve) B cells are increased in the PB (d'Arbonneau et al, 2006). Why Bm cells are abnormally distributed in the PB of primary SS patients remains unresolved (Bohnhorst et al, 2001; Hansen et al, 2002). Of particular interest (Binard et al, 2009) is that a ratio of Bm2-plus-Bm2′ cells to early Bm5-plus-late Bm5 cells equal to or superior to 5 is diagnostic for primary SS (Figure 3), compared with healthy controls and with disease controls afflicted with rheumatoid arthritis (RA) or systemic lupus erythematosus (SLE).

image

Figure 3. B-cell subsets in patients with primary Sjögren's syndrome (pSS). Bm2 + Bm2′/eBm5 + Bm5 ratio is increased in pSS patients compared with systemic lupus and rheumatoid arthritis patients and healthy controls. pSS, primary Sjögren's syndrome; SLE, systemic lupus erythematosus; RA, rheumatoid arthritis (Binard et al, 2009)

Download figure to PowerPoint

CD27 and IgD staining reveals a reduction in the number of circulating CD27+ IgD+ IgM+ memory B cells, without isotype switching, whereas the surface expression of HLA-DR, CD38, and CD95 is enhanced and that of CD21 and CXCR5 reduced (Hansen et al, 2009). Such phenotype reflects abnormal differentiation and censoring mechanisms within secondary lymphoid organs, ectopic GC. GC-derived CD27+ IgD IgM memory B cells are retained in SG and proportionally diminished in the PB.

Analysis of SG-infiltrating B lymphocytes

Chemokine receptor–ligand interactions, most notably CXCL12-CXCR4 and CXCL13-CXCR5 interactions, give us a clue as to why B cells escape from the PB to colonize the glands (Bohnhorst et al, 2001). In the tissues, MNC progressively substitutes for epithelial cells (EC). The ensuing histopathological damages spread from mild infiltrates (that do not significantly affect the tissue architecture) to diffuse infiltrates (that cause a concomitant loss of normal structures).

However, we ignore the fine mechanism of accumulation of MNC in the tissue. Parenthetically, what we realized was that lowering of circulating CD6-expressing B cells may result from their migration into the SG. This might be simply explained by the ability of the CD6 molecule to encourage lymphocytes to cross endothelial barriers. Consistent with this view is our finding of a abnormally high level of CD166, one of the ligands for CD6, on EC of the SG (Alonso et al, 2010).

Immunological cells interfere with the local secretion of interferon (IFN)-related pathway-activating cytokines and releasing auto-Ab inside the SG. Their systemic hyperactivity notwithstanding, a minority of B lymphocytes are seen near the EC (Bodeutsch et al, 1993), along with a handful of PC (Tapinos et al, 1998) and virtually no macrophages (Stott et al, 1998). In this area, up to 80% of MNC are activated CD4+ lymphocytes (Xanthou et al, 1999) and 10% granzyme-containing cytotoxic CD8+ T lymphocytes. The respective proportions of macrophages, T lymphocytes, B lymphocytes, and dendritic cells reflect the severity of the lesions (Christodoulou et al, 2010). That is, T cells predominate in mild lesions, whereas it is B cells in those advanced, with decreased and increased numbers of macrophages and dendritic cells, respectively. The decline of T-cell population is related to a decrease in the number of CD4+ T cells, inasmuch as the number of infiltrating CD8+ T cells does not change.

The SG CD20+ CD27+ memory B cells are hyper-reactive, as suggested by their striking expression of auto-Ab (Hansen et al, 2002). Locally, IgG- and IgM-making PC predominate in the patients' SG, whereas the main isotype is IgA in those of the controls (Speight et al, 1990; Matthews et al, 1993). B lymphocytes infiltrating SG may be driven to initiate oligoclonal or monoclonal proliferations (Pablos et al, 1994; Jordan et al, 1995; Stott et al, 1998) and release RF and anti-SSA/Ro and anti-SSB/La Ab in the tissue (Halse et al, 1999; Salomonsson et al, 2003). Finally, B-cell clusters exist in one-third of the SG samples, particularly when titers of RF are high, serum levels of IgG augmented, and focus scores elevated (Jonsson et al, 2007).

The current view holds that GC are common in the SG of patients with primary SS. Nevertheless, we have observed that genuine GC of B cells are less frequent than hitherto acknowledged. Indeed, the B-cell-associated CD10 and CD38 markers for GC resident cells are absent from most of these clusters of B cells and the transcripts for AID undetectable in most of the laser-dissected aggregates of B cells (Le Pottier et al, 2009b). Our interpretation is that such aggregates do not necessarily consist of functional GC (Salomonsson et al, 2003). Additionally, BT2 cells have been identified in these GC-looking aggregates (Daridon et al, 2007a).

Based on the fact that xerosis cutis is a common complain of these patients, it occurred to us that skin biopsies might be diagnostic for the disease, and indeed, skin samples documented lymphocytic infiltration in eight of the 12 patients tested (Roguedas et al, 2010). Ironically, two patients with abnormal skin biopsy had normal SG biopsies. Their skin was infiltrated with memory CD10 CD20+ CD27+ IgD and immature CD20+ CD24+ B lymphocytes. Thus, it seems reasonable to suggest use of serial skin biopsies to monitor the effects of biotherapy in a prospective cohort of patients.

Intrinsic B-cell defects in primary SS

  1. Top of page
  2. Abstract
  3. Introduction
  4. Criteria and new tools for the diagnosis
  5. B-lymphocyte subpopulations
  6. Intrinsic B-cell defects in primary SS
  7. Disturbances of cytokines in primary SS
  8. B-cell proliferation in primary SS
  9. Anti-B-cell-targeted therapy in primary SS
  10. Conclusion
  11. Author contributions
  12. References

Although the reasons for hyperactivity of B cells in primary SS are elusive, at least three intrinsic abnormalities have been identified. First, preswitch Ig transcripts are retained in circulating memory B cells, regardless of the advent of postswitch Ig transcripts (Hansen et al, 2004). Second, kinetics of the BCR translocation into the lipid-rich membrane signaling microdomains, called lipid rafts (LR), is impaired in patients' B cells (d'Arbonneau et al, 2006). This default maintains the receptors within the LR and sustains the signal from the BCR. Third, using the single-cell technology on B lymphocytes eluted from the SG and triple reverse-transcriptase polymerase chain reactions, these B cells have been shown to contain mRNA for BAFF.

Disturbances of cytokines in primary SS

  1. Top of page
  2. Abstract
  3. Introduction
  4. Criteria and new tools for the diagnosis
  5. B-lymphocyte subpopulations
  6. Intrinsic B-cell defects in primary SS
  7. Disturbances of cytokines in primary SS
  8. B-cell proliferation in primary SS
  9. Anti-B-cell-targeted therapy in primary SS
  10. Conclusion
  11. Author contributions
  12. References

Recent discoveries have unveiled new aspects of B-cell-derived cytokines, such as IFN-γ and interleukin (IL)-4 that modulate the response (Youinou et al, 2009) Given the kinetics of B-cell generation and the cytokine profile of B lymphocytes, a T helper (Th) 1 phenotype may be imprinted by B effector (Be) 1 cells via their expression of IL-2 and IFN-γ. This is reinforced by an IFN- γ/IFN-γ receptor autocrine loop. Conversely, Th2 cells polarize naïve B lymphocytes into Be2 cells, which produce IL-4 and IL-6. The Moutsopoulos' group has cleverly forwarded the concept that the balance between Th1 and Th2 cells correlates with the progress of immunopathological lesions in the SG (Mitsias et al, 2002). In other words, the Th1/Th2 balance parallels the progress of lesions in the tissues. Their interpretation agrees with our finding, in the SG, that Be1 accompany Th1, whereas Be2 accompany Th2 (Daridon et al, 2007b). Not only do the serum cytokine patterns distinguish patients from controls, but they also differentiate one patient from another, depending on extraglandular complications (Szodoray et al, 2004).

Overall, B-cell-derived cytokines fall into three families: pro-inflammatory cytokines (e.g. IL-1, IL-6, TNF-α and lymphotoxins), immunosuppressive cytokines (e.g. transforming growth factor-β and IL-10), and hematopoietic growth factors (e.g. IL-17 and granulocyte/monocytes–colony-stimulating factor). SS patients have more IL-14 in their MNC compared with the controls (Shen et al, 2009). This B-cell growth factor has, however, been less extensively studied in humans than in mice. Those mice transgenic (Tg) for IL-14 develop clinical and immunological features of pSS. Noteworthy is that these abnormalities follow one another from early hypergammaglobulinemia to B-cell lymphoma, in the same order that occurs in the patients.

IL-6 is influential as well [reviewed in: (Youinou and Jamin, 2009)]. It promotes Th17 cells and activates B cells in an autocrine manner. The IL-6-induced Th17 cells orchestrate the development of GC by autoreactive lymphocytes (Hsu et al, 2008), as recently shown to take place in pathological SG (Le Pottier et al, 2009b). We have also provided evidence that IL-6 favors the expression of RAG and therefore secondary Ab gene rearrangements as well as auto-Ab synthesis (Hillion et al, 2007).

B-lymphocyte activator factor belonging to the TNF family, also known as BLyS (for B-lymphocyte stimulator), has been described in the late 1990s (Schneider et al, 1999), and since recognized as one of the most influential factors for maturation, tolerization, and transformation of B cells. In addition to BAFF and APRIL, the TNF family comprises three receptors: B-cell maturation Ag (for BCMA), transmembrane activator and calcium interactor (for TACI), and BR3 (also known as BAFF receptor). BAFF interacts chiefly with BR3, but can bind to the other two receptors, whereas APRIL is restricted to BCMA and TACI (Mackay and Browning, 2002).

Because of the reduced competitiveness of auto-Ag-engaged B cells, due to their increased dependance on BAFF (Lesley et al, 2004), an excess of BAFF rescues unwanted autoreactive B cells from tolerizing deletion (Mackay et al, 1999) and leads them to forbidden FO and MZ niches (Thien et al, 2004). The problem is that B cells are particularly prone to polyclonal activation in the MZ. Consequently, preventing autoreactive B cells from colonizing this area is paramount, so that aberrant production of BAFF constitutes a serious threat.

Multiple findings have highlighted the leading part of BAFF in the pathogenesis of primary SS. First, BAFF-Tg mice develop autoimmune traits reminiscent of the disease (Groom et al, 2002). Considering patients with primary SS, high serum levels of BAFF are found in their SG (d'Arbonneau et al, 2006), their saliva (Pers et al, 2005a), and their sera (Pers et al, 2005b). However, there exists the issue of why the serum concentrations of BAFF remains within, or below, normal range in a proportion of autoimmune patients. Moreover, increased serum concentrations of BAFF have been associated with the production of RF by some (Bosello et al, 2008), but not other authors (Becker-Merok et al, 2006), or with that of anti-nuclear Ab by some (Mariette et al, 2003), but not other authors (Collins et al, 2006). Additionally, estimates of BAFF fluctuate with changes in inflammatory activity, extent of the damages, and disease classification criteria chosen by the investigators. Such awareness prompted us to set up an enzyme-linked immunosorbent assay for BAFF (Le Pottier et al, 2009a). Our intuition appeared to be correct when our in-house assay detected high levels of BAFF in virtually all sera from patients with primary SS. Importantly, BAFF enhances the membrane expression of CD19, by upregulating the B-cell-specific transcription factor Pax5 (Hase et al, 2004). This has since been confirmed by our finding that in addition to augmenting their number, the patients' circulating Bm2/Bm2′ cells carry more CD19 molecules than those of the controls (d'Arbonneau et al, 2006). An inference is that the prolonged residency of the BCR in the LR was associated with high CD19 expression on sorted Bm2/Bm2′ cells from patients and correlated with serum levels of BAFF. In SG, BAFF is markedly expressed by EC, activated T lymphocytes, and by B lymphocytes. One consequence is that auto-Ab isotypes switch outside and inside the GC (Daridon et al, 2007b; Le Pottier et al, 2009b). Thus, not only does BAFF favor the expansion of BT2 and MZ-like B cells, to the expenses of FO cells in the SG, but it also regulates their reappearance in the SG after therapeutic B-cell depletion.

B-cell proliferation in primary SS

  1. Top of page
  2. Abstract
  3. Introduction
  4. Criteria and new tools for the diagnosis
  5. B-lymphocyte subpopulations
  6. Intrinsic B-cell defects in primary SS
  7. Disturbances of cytokines in primary SS
  8. B-cell proliferation in primary SS
  9. Anti-B-cell-targeted therapy in primary SS
  10. Conclusion
  11. Author contributions
  12. References

Aberrant expression of BAFF by infiltrating MNC and structural EC has been implicated in expansion of autoreactive B lymphocytes, their altered differentiation, generation of GC, and development of NHL. Autoreactive BT1 and BT2 cells move to an area that is reminiscent of the splenic MZ and behave as a fast track to autoimmunity. In the SG, they are likely to aggregate within the SG, where accumulate a majority of BT2 and MZ-like B lymphocytes (Daridon et al, 2006). They may serve as reservoirs for autoreactive B cells. Given the long-lasting simulation with auto-Ag, they proliferate (Friedman, 1991), as illustrated when MZ-lacking BAFF-Tg mice develop nephritis, but not sialadenitis (Fletcher et al, 2006). For autoimmune epitheliitis to occur (Moutsopoulos, 1994), MZ B cells would thus be necessary. Transformation of naïve B cells into mature proliferating cells through non-malignant pseudolymphoma follows a continuum.

Among all autoimmune diseases, primary SS displays the highest incidence of malignant lymphoproliferative disorders [reviewed in: (Youinou et al, 2010)]. Recent meta-analysis has calculated that the standardized incidence risk to develop NHL varies from 9 to 48 in primary SS, from 3 to 7 in SLE, and from 1.5 to 4 in RA (Smedby et al, 2006). Although various types of lymphoma have been described, diffuse large B-cell lymphoma (DLBCL) represents almost 30% of the cases and mucosa-associated lymphoid tissue (MALT) lymphoma, almost 70% (Voulgarelis et al, 1999). These latter primary low-grade lymphomas are preferentially extranodal MZ, usually located in the major SG, and restricted to stages I and II. The poorest prognosis is that of DLBCL.

Several clinical and serological markers have been claimed to herald the development of NHL (Table 2). For example, a fivefold increased risk of NHL has been associated with splenomegaly, persistent enlargement of parotid glands, lymphadenopathy (Kassan et al, 1978; Anaya et al, 1996; Sutcliffe et al, 1998), palpable purpura (Ioannidis et al, 2002), mixed cryoglobulinemia (Tzioufas et al, 1996), low serum levels of C4 (Ramos-Casals et al, 2005), neutropenia (Baimpa et al, 2009), or lymphocytopenia (Theander et al, 2006).

Table 2. Clinical and serological markers associated with high risk of lymphoma development in primary Sjögren's syndrome
Splenomegaly
Persistent enlargement of parotid glands
Lymphadenopathy
Palpable purpura
Cryoglobulinemia
Low levels of C4
High levels of Flt3 ligand
Lymphocytopenia
Neutropenia
Anemia
Monoclonal component
High levels of β2-microglobulin
Hypogammaglobulinemia

Retrospective analyses of early SG biopsy samples collected for diagnosis have also concluded that the presence of GC would be predictive of NHL (Theander et al, 2011). We have, ourselves, demonstrated that high serum levels of Fms-like tyrosine kinase 3 ligand, a cytokine implicated in ontogenesis of B cells, and proliferation of hematological malignancies, herald the development of a NHL in primary SS (Tobon et al, 2010) and Tobón GJ, Saraux A, Gottenberg JE, Quartuccio L, Fabris M, Seror R, Devauchelle-Pensec V, Morel J, Rist S, Mariette X, De Vita S, Youinou P, Pers JO, manuscript in preparation).

Anti-B-cell-targeted therapy in primary SS

  1. Top of page
  2. Abstract
  3. Introduction
  4. Criteria and new tools for the diagnosis
  5. B-lymphocyte subpopulations
  6. Intrinsic B-cell defects in primary SS
  7. Disturbances of cytokines in primary SS
  8. B-cell proliferation in primary SS
  9. Anti-B-cell-targeted therapy in primary SS
  10. Conclusion
  11. Author contributions
  12. References

The most widely studied target molecule for B-cell-related therapy is CD20 (Table 3). This Ag is a hydrophobic 35-kDa transmembrane protein detected on pre-B cells, mature B cells (Einfeld et al, 1988; Valentine et al, 1989), and over 90% of B cells in NHL (Anderson et al, 1984).

Table 3. Potential B-cell targets and treatments used for B-cell depletion in patients with Sjögren's syndrome
Direct targeting of B cells
CD-20 antigen
 Rituximab (chimeric monoclonal antibody)
 Ocrelizumab (humanized monoclonal antibody)
 Ofatumumab (human monoclonal antibody)
 Veltuzumab (humanized monoclonal antibody)
 TRU-015 (engineered protein)
CD-22 antigen
 Epratuzumab (humanized monoclonal antibody)
Indirect targeting of B cells
  1. BAFF, B-lymphocyte activator factor belonging to the TNF family; TACI, transmembrane activator and calcium modulator and cyclophilin ligand interactor.

BAFF
 Belimumab (LymphoStat B: fully human monoclonal antibody against BAFF)
BAFF receptors
 Anti-BR3
 Atacicept (IgG Fc fused to the extracellular TACI receptor domain)
 Briobacept/BR3-Fc (IgG Fc fused to the extracellular BAFF receptor (BR3) domain)

So far, rituximab (RTX) is the only anti-CD20 mAb tested as a treatment for primary SS. Retrospective analyses (Gottenberg et al, 2005; Galarza et al, 2008), open-label studies (Pijpe et al, 2005; Devauchelle-Pensec et al, 2007), and randomized double-blind placebo-controlled trials [(Meijer et al, 2010) Saraux et al, manuscript in preparation] show a certain efficacy of RTX for at least 6–9 months in patients with active primary SS, whose subjective complaints and objective abnormalities were ameliorated. Sicca symptoms abated in some of them, as well as B-cell-associated laboratory abnormalities. Exocrine functions and extraglandular manifestations responded to RTX in early disease. Similarly, retreatment with RTX gives good clinical responses. To confirm these promising results, additional randomized placebo-controlled trials are presently in progress in France, referred to as TEARS, and in the United Kingdom, referred to as TRACTISS.

Examination of the local consequences of RTX-induced B-cell depletion has provided new insights into the pathogenesis of B-cell abnormalities of primary SS. This drug reduces lymphocytic infiltrates (Pers et al, 2007), amends lymphoepithelial lesions, and erases, at least partially, ectopic GC (Seror et al, 2007). The ensuing repopulation reproduces the stage-by-stage ontogeny of B lymphocytes. After B cells are killed, their reconstitution starts with a predominance of naïve B cells, while memory B cells are lacking. In contrast, BT1 and memory B cells were the first to repopulate the SGs. In other words, the local environment contributes to the abnormal fate of B cells.

Another therapeutic approach aims at facing the deleterious effects of BAFF on B cells. These can be repressed with anti-BAFF or anti-BR3 mAb or with BR3 or TACI decoy fusion proteins. Selective BAFF blockers prevent BAFF from interacting with its receptors, while APRIL remains free for binding to TACI and BCMA. In contrast, non-selective BAFF blockers abolish interactions of BAFF and APRIL with the three receptors. To date, there exists one drug in this class (Dass et al, 2008), which is a human Ig Fc protein fused to the extracellular domain of TACI. Variability in the distribution of the forms of BAFF denotes the potential of patients to resist BAFF antagonists. Treatment of B cells with anti-TACI mAb inhibits proliferation in vitro, and activation of a chimeric receptor including the intracellular domain of TACI induces apoptosis (Seshasayee et al, 2003). These results emphasize how necessary TACI is for homeostasis of B cells. Some therapeutic efficacy of the BAFF antagonists has been observed in patients with SLE. In primary SS, two studies are currently on their way to evaluate its effects.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Criteria and new tools for the diagnosis
  5. B-lymphocyte subpopulations
  6. Intrinsic B-cell defects in primary SS
  7. Disturbances of cytokines in primary SS
  8. B-cell proliferation in primary SS
  9. Anti-B-cell-targeted therapy in primary SS
  10. Conclusion
  11. Author contributions
  12. References

No doubt, primary SS is unequaled as a model to incriminate B cells in the genesis of autoimmune states. Despite a number of caveats in their prosecution, B cells should plea guilty as the main lymphocyte type in the disease. Such a verdict does not negate an important role for T cells. As discussed throughout this review, several B-cell abnormalities have been characterized in the PB and the SG. They may help to follow up B cell-targeted therapies. Several groups of experts are working on BAFF. By the time their results are obtained, BAFF antagonists will be used. Other B-cell-derived cytokines, such as IL-6 and IL-14, are also expected to open new therapeutic avenues. This area is now so worthy of pursuit that the earliest diagnosis as possible by the dentist has become crucial.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Criteria and new tools for the diagnosis
  5. B-lymphocyte subpopulations
  6. Intrinsic B-cell defects in primary SS
  7. Disturbances of cytokines in primary SS
  8. B-cell proliferation in primary SS
  9. Anti-B-cell-targeted therapy in primary SS
  10. Conclusion
  11. Author contributions
  12. References