B-lymphocyte depletion ameliorates Sjögren's syndrome in Id3 knockout mice

Abstract

Sjögren's syndrome is an autoimmune disease in which immune cells chronically attack the lachrymal and salivary glands. The Id3 knockout mouse is a newly established animal model for primary Sjögren's syndrome. To address the role of B cells in Sjögren's syndrome and autoimmune disease, we studied the effect of CD20 monoclonal antibody treatment on the disease in Id3 knockout mice. Antibody treatment at 2-month intervals led to efficient and sustained B-cell depletion in Id3 knockout mice. A significant improvement of histopathology was observed accompanied by the recovery of saliva secretory function after CD20 antibody treatment. We further show that serum immunoglobulin G3, which is abnormally high in untreated Id3 knockout mice, was reduced after CD20 antibody treatment. This study establishes a new animal model for immunotherapy of Sjögren's symptoms and suggests a possible link between immunoglobulin G3 and disease pathology in Id3 knockout mice.

Introduction

Sjögren's syndrome is a systemic autoimmune disease characterized by chronic inflammation of the lachrymal and salivary glands and by systemic symptoms such as fatigue, arthritis, pulmonary involvement, interstitial nephritis, peripheral neuropathy and vasculitis.¹ Lymphocytic infiltration of the lachrymal and salivary glands results in impaired tear and saliva secretion. Approximately one half of patients diagnosed with Sjögren's syndrome also develop other types of rheumatic diseases such as rheumatoid arthritis or systemic lupus erythematosus. Therefore, Sjögren's syndrome is further divided into primary and secondary depending on the absence or presence of other rheumatic diseases, respectively.

Humoral immunity is tightly linked to the pathogenesis of Sjögren's syndrome. Hypergammaglobulinaemia and autoantibodies such as anti-Sjögren syndrome antigen A/B (SSA/SSB) and rheumatoid factor are common and important clinical features found in both primary and secondary Sjögren's syndrome. Recent studies reveal a correlation between specific autoantibodies reactive with α-fodrin and muscarinic (M3) to Sjögren's syndrome.^2,3 The appearance of autoantibodies seems generally to be correlated with the severity of disease symptoms.^4,5 Patients with primary Sjögren's syndrome are also at increased risk of developing B-cell non-Hodgkin's lymphoma.^6,7 Therefore, interventions aimed at controlling B-cell function could have therapeutic benefit in treating Sjögren's syndrome.

The Id3 knockout (Id3^–/–) mouse is an established animal model for primary Sjögren's syndrome.⁸ Id3, a member of the ID (inhibitor of differentiation) family of transcription factors, is a 13 000 MW nuclear protein that is up-regulated in a broad range of cell types upon stimulation with serum and many other growth factors. The primary function of Id3 is to inhibit the DNA binding of basic-helix-loop-helix (bHLH) transcription factors such as E2A.⁹ The Id3^–/– mice develop many symptoms found in primary Sjögren's syndrome patients including dry eyes and mouth.⁸ Similar to human patients, the severity of disease symptoms in Id3^–/– mice progresses with age. Significant lymphocyte infiltration begins to appear in the lachrymal and salivary glands around 6 months of age. Significantly, an increase in the frequency and severity of lymphocyte infiltration is accompanied by the appearance of autoantibodies against SSA (Ro) and SSB (La) around 12 months of age, suggesting a role for humoral immunity in disease progression in this animal model.

CD20 is a B-cell-specific surface antigen expressed on both immature and mature B cells.^10,11 CD20 monoclonal antibody (mAb) immunotherapy has been shown to be effective in treating CD20-positive lymphoproliferative diseases. Recently, CD20 mAb has also been used in the treatment of autoimmune diseases. Efficacy of CD20 mAb treatment was reported from blinded and randomized control trials on rheumatoid arthritis¹² and the US Food and Drug Administration (FDA) has approved its use for rheumatoid arthritis. Published uncontrolled series for idiopathic thrombocytopenic purpura, polyneuropathy, systemic lupus erythematosus and Wegener's vasculitis, also suggest the potential benefits of CD20 mAb therapy.^13,14

To explore the effectiveness of B-cell depletion in Sjögren's syndrome, we assessed whether CD20 mAb had beneficial effects in Id3^–/– mice. We observed efficient depletion of B cells after CD20 mAb treatment of Id3^–/– mice. Treated Id3^–/– mice showed improvement of objective parameters of disease activity, including a decrease in cell infiltration in the extraorbital lachrymal and submandibular glands in Id3^–/– mice. Our studies further suggested a possible role for immunoglobulin G3 (IgG3) in pathogenesis.

Materials and methods

Mice and immunotherapy

Generation of Id3^–/– mice has been previously described.¹⁵ The Id3^–/– mice were maintained on either 129 and C57BL/6 mixed background or C57BL/6 background after an 11-generation backcross. Both genetic backgrounds produced disease phenotypes with similar kinetics,⁸ (Zhuang, unpublished data) and were therefore used in the current study as a single group of Id3^–/– mice. Age-matched C57BL/6 mice were used as wild-type controls. Mouse anti-mouse CD20-specific mAb were produced as described elsewhere.¹¹ Anti-mouse CD20 and isotype-matched control mAbs in 200 μl phosphate-buffered saline were injected intraperitoneally. To determined the effects on B-cell depletion by CD20, 2- to 3-month-old Id3^–/– mice were injected with either 100 μg IgG2a isotype CD20 mAb (MB20-11) or a control IgG2a mAb. For the evaluation of efficacy of CD20 mAb treatment, 6-month-old Id3^–/– mice or wild-type mice were injected with either IgG2a isotype CD20 mAb (MB20-11; 250 μg each time) or a control IgG2a mAb at 2-month intervals. Eye bleeding and saliva tests were performed at monthly intervals. The second and third rounds of mAb injection were carried out immediately after the saliva test and eye bleeding. All animal procedures were approved by the Duke University Animal Use and Care Committee.

Saliva secretion test

Saliva secretion was examined as described elsewhere.⁸ Mice were anaesthetized with avertin before the assay. Pilocarpine hydrochloride (Sigma, St. Louis, MO) was dissolved in double-distilled water and injected intraperitoneally (0·5 mg/g body weight) to stimulate tear and saliva production. Saliva was collected with a 20 μl microcapillary pipette immediately after pilocarpine injection for a duration of 9 min. Saliva secretory volumes were normalized to body weight.

Flow cytometry analysis

Single-cell suspensions of lymphocytes from the spleen, peripheral lymph nodes (paired inguinal) and peripheral blood were prepared in ice-cold phosphate-buffered saline supplemented with 5% bovine calf serum. Splenocytes were depleted of erythrocytes by ammonium chloride lysis before use. All suspensions were counted with a haemocytometer, and 10⁶ cells were stained immediately with a combination of a fluorescein isothiocyanate (FITC)-conjugated antibody, a phycoerythrin (PE)-conjugated antibody, an allophycocyanin (APC)-conjugated antibody, and 7-amino-actinomycin D (7AAD) (Molecular Probes, Eugene, OR). Cells were washed once with phosphate-buffered saline containing bovine calf serum and analysed on a FACScalibur flow cytometer (Becton Dickinson, San Jose, CA). Data from 10⁵ cells were collected and analysed using the Flowjo program (TreeStar, Inc.). Antibodies used in this study included FITC-conjugated anti-B220 (RA3-6B2; Caltag, Burlingame, CA), FITC-conjugated anti-mouse IgM (Caltag), PE-conjugated Thy1.2 (5a-8; Caltag), APC-conjugated B220 (RA3-6B2; Caltag) antibodies. Dead and damaged cells were labelled with 7AAD, and were eliminated from the analysis.

Serum immunoglobulin isotype-specific ELISA

Immunoglobulin levels in diluted serum were determined by an isotype-specific enzyme-linked immunosorbent assay (ELISA) as described previously.¹⁶ Briefly, ELISA plates were coated with antibodies against mouse immunoglobulins (Southern Biotechnology Associates, Inc., Birmingham, AL). A standard curve was generated using affinity-purified mAbs against mouse IgM, IgG1, IgG2a, IgG2b and IgG3 (Southern Biotechnology Associates, Inc.). The immunoglobulin concentration for each individual sample was determined by comparing the mean optical density values from duplicate wells to the standard curve.

Histology

Histology sections were prepared from frozen tissues and stained with haematoxylin & eosin (H&E). The following criteria were used to score infiltrate foci for each mouse as described previously.⁸ One infiltrate focus is defined as 50 or more nucleated cells in the cluster. Two random measurements were taken per tissue section. Focus score is the total infiltrate foci of histological sections prepared from one facial side, including one extraorbital lachrymal gland and one submandibular gland, of the mouse.

Statistical analysis

Data shown in graphs are means ± SEM. The Student's t-test was used for verifying significant differences between samples.

Results

Elevated serum immunoglobulin levels in 6-month-old Id3^–/– mice

We showed in our earlier studies that a significant increase in frequency of lymphocyte infiltrations in lachrymal and salivary glands occurred around 6 months of age.⁸ To further evaluate humoral immunity in Id3^–/– mice, we examined serum immunoglobulin levels in 6-month-old Id3^–/– mice. Among five immunoglobulin isotypes examined, IgG2a, IgG3 and IgM levels were significantly increased in Id3^–/– mice compared to age-matched wild-type controls (Fig. 1). This observation suggests that gland tissue pathology in Id3^–/– mice is accompanied by elevated levels of serum immunoglobulins.

Figure 1.

Serum immunoglobulin levels in 6-month-old Id3^–/– mice. Sera from 6-month-old Id3^–/– and wild-type mice were analysed by ELISA for IgM, IgG1, IgG2a, IgG2b and IgG3 levels. Significant differences between Id3^–/– and wild-type mice are indicated by *P < 0·05 or **P < 0·01. No significant difference was observed for IgG1 or IgG2b.

B-cell depletion after a single dose of CD20 mAb in 2- to 3-month-old Id3^–/– mice

The CD20 mAb treatment effectively eliminates circulating B cells.^11,17 We first determined the kinetics of B-cell depletion after CD20 mAb treatment in Id3^–/– mice. Efficient depletion of B cells from blood, lymph node and spleen was observed in Id3^–/– mice 7 days after a single intraperitoneal dose of 100 μg CD20 mAb (Fig. 2). The numbers of IgM⁺ B220⁺ B cells in blood, spleen and lymph nodes after CD20 mAb treatment for both wild-type and Id3^–/– mice were reduced more than 90% in comparison with control mAb-treated groups. Specifically, circulating B cells were reduced from 3·4 × 10³ ± 0·4 × 10³/μl and 5·0 × 10³ ± 0·7 × 10³/μl in untreated to 0·4 × 10² ± 0·1 × 10²/μl and 1·2 × 10² ± 0·2 × 10²/μl in CD20 mAb-treated wild-type and Id3^–/– mice, respectively. Meanwhile, circulating or tissue T-cell numbers were unchanged (data not shown). Thus, CD20 mAb treatment effectively and specifically depleted B cells in Id3^–/– mice just as it did in wild-type control mice.

Figure 2.

Depletion of tissue B cells following CD20 mAb treatment. B220⁺ B cells from blood, peripheral lymph nodes (PLN), and spleen were analysed by FACS 1 week after treatment with either CD20 or control mAb. Histograms of B220 staining are representative of at least three mice in each genotype group treated with CD20 mAb (MB20-11, 100 μg; solid line) or isotype-matched control mAb (CTL, 100 μg; shadow). Percentages of B220⁺ and B220^– B cells in the CD20 treated samples are indicated in each plot.

B-cell depletion following CD20 mAb treatment of 6-month-old Id3^–/– mice

To evaluate the efficacy of CD20 mAb in disease treatment, we chose 6-month-old Id3^–/– mice for the test and age-matched wild-type mice as controls. Antibody treatment and assay procedures are outlined in Fig. 3(a). The experiment was terminated for histopathology assessment 5 months after the initiation of antibody treatment when mice had reached 11 months of age. We collected blood samples at monthly intervals after mAb infusion. Circulating B cells were reduced to undetectable levels for up to 1-month post CD20 mAb treatment and were detected at low levels by 2 months post CD20 mAb treatment (Fig. 3b). In contrast, circulating T cells were not affected by CD20 mAb treatment (data not shown). Additional mAb treatment at 2-month intervals was sufficient to keep circulating B cells at low levels during the course of treatment.

Figure 3.

Sustained depletion of circulating B cells in mice treated with CD20 mAb. (a) Schedule of mAb injection and assays performed during 5-month period. Six-month-old Id3^–/– or wild-type mice were injected with CD20 or control mAb (250 μg each time) at 2-month intervals. Serum immunoglobulin and saliva secretion were assayed at the time-points indicated in the table. Mice were killed 5 months after the initiation of the mAb treatment for final histology assessment of gland tissues. (b) IgM⁺ B220⁺ cells from peripheral blood were determined by flow cytometry analysis at monthly intervals. The second and third rounds of mAb injection were carried out after blood sampling. Number shown is IgM⁺ B220⁺ cell numbers (× 10³/µl) in blood. The P-values for changes in B-cell numbers between the starting point and 1, 2, 3, or 4 months after the first-round CD20 mAb injection were 0·0001, 0·05, 0·0001, and 0·001 for wild-type and 0·0001, 0·0005, 0001, and 0·005 for Id3^–/– mice, respectively.

Decrease of serum immunoglobulin levels following B-cell depletion

Serum immunoglobulin levels were followed after CD20 mAb treatment. Id3^–/– mice had increased IgG2a, IgG3 and IgM levels before the treatment (Fig. 1). Both IgG2a and IgM levels showed a decrease after 4 months of treatment (Fig. 4a). However, there was no significant difference between CD20 and control mAb treatment for these two immunoglobulin isotypes at the end of treatment, suggesting that the changes are not linked to CD20 mAb-mediated B-cell depletion. The level of IgG1 was also reduced in both Id3^–/– and wild-type control groups following CD20 mAb treatment. The most dramatic effect of CD20 mAb treatment was on the levels of serum IgG3 (Fig. 4a,b). Serum IgG3 levels were decreased on average 22-fold from 391 μg/ml at the beginning to 18 μg/ml at the end of CD20 mAb treatment in Id3^–/– mice. In contrast, serum IgG3 levels in Id3^–/– mice treated with control mAb were only reduced three-fold and remained higher than those in wild-type mice at the end of treatment.

Figure 4.

Serum immunoglobulin levels in Id3^–/– mice treated with CD20 mAb. (a) Serum levels of IgG1, IgG2a, IgG3 and IgM were determined by ELISA at monthly intervals following antibody treatment of wild-type (dash line) and Id3^–/– mice (solid line) as outlined in Fig. 3(a). Top and bottom panels are CD20 and control antibody treatments, respectively. Each data point is the average of at least four mice. Student's t-tests were performed to evaluate changes in serum immunoglobulin levels before versus after 4 months of CD20 mAb treatment of Id3^–/– mice (IgG1, NS; IgG2a, P < 0·05; IgG3, P < 0·005; IgM, NS). (b) Serum IgG3 levels are compared before and after the 4-month treatment: wild-type mice, open bars; Id3^–/– mice, solid bars; *P < 0·05, **P < 0·005, NS indicates not significant.

Improvement of saliva secretion following CD20 mAb treatment

The volume of saliva secretion upon pilocarpine stimulation was also determined before and after the completion of mAb treatment. We found that Id3^–/– mice produced significantly less saliva than wild-type controls before mAb treatment. Saliva secretion was significantly increased in Id3^–/– mice at the end of 4 months of CD20 mAb treatment (Fig. 5). In contrast, no significant improvement was observed among Id3^–/– mice treated with control mAb.

Figure 5.

Improvement of saliva secretory function following CD20 mAb treatment. Saliva secretory volume (in μl/g body weight) upon pilocarpine stimulation was determined before and after 4-month mAb treatment. A significant difference was observed between wild-type and Id3^–/– mice before mAb treatment (P < 0·05). A significant increase in saliva volume was observed among Id3^–/– mice treated with CD20 mAb (P < 0·05) but not with control mAb (NS as not significant).

Reduction of gland lymphocyte infiltrates following CD20 mAb treatment

A hallmark of disease symptoms in Id3^–/– mice is the appearance of infiltrating foci of lymphocytes in lachrymal and submandibular glands beginning around 6 months of age. We stopped the experiment 5 months after the initiation of mAb treatment and evaluated the histopathology of lachrymal and submandibular glands (Fig. 6). Each mouse had received three rounds of mAb treatment at the time of analysis (Fig. 3). Consistent with the previous report,⁸ we observed a significant amount of infiltrating foci among 6-month-old Id3^–/– mice (Fig. 6b). The number of foci in Id3^–/– mice receiving control mAbs was not significantly different between samples collected at the beginning and the end of treatment. In contrast, Id3^–/– mice receiving CD20 mAb treatment showed significantly reduced numbers of infiltrating foci at the end of treatment. In fact, only one focus was found in all the gland tissues examined among four CD20-treated Id3^–/– mice (Fig. 6a–c). Wild-type mice did not show any foci in the lachrymal and submandibular glands throughout the course of CD20 or control mAb treatment. Thus, CD20 mAb treatment can effectively ameliorate the pathological symptoms in Id3^–/– mice.

Figure 6.

Histopathology assessment of submandibular and extraorbital lachrymal glands. (a) Representative photos of haematoxylin and eosin (H&E) staining of submandibular gland (SMG, upper panel) and extraorbital lachrymal glands (LG, lower panel). Samples are from a 6-month-old wild-type mouse (i and v), a 6-month-old Id3^–/– mouse (ii and vi), an 11-month-old Id3^–/– mouse treated CD20 mAb (iii and vii), and an 11-month-old Id3^–/– mouse treated with control mAb (iv and viii). Arrows indicate infiltrating foci in tissue sections. Photos were taken with the original magnification of 10 × 4 lenses (i–iv) or 10 × 10 lenses (v–viii), respectively. (b) Summary of focus analysis. Focus score represents the total number of infiltrate foci counted in the histological sections of submandibular and extraorbital lachrymal glands from one facial side. One focus is defined as a cluster of at least 50 mononucleated cells. Samples include 6-month-old wild-type mice, 6-month-old Id3^–/– mice, 11-month-old wild-type mice treated with CD20 mAb, 11-month-old Id3^–/– mice treated with CD20 mAb, 11-month-old wild-type mice treated with control mAb, and 11-month-old Id3^–/– mice treated with control mAb.

Discussion

Currently, there is no known cure for Sjögren's syndrome, nor is there a specific treatment to restore gland function. The available treatments are generally aimed at alleviating dryness symptoms and related morbidities.¹⁸ More recent approaches to therapy for Sjögren's syndrome include the use of muscarinic (M3) receptor agonists to increase glandular excretion. Systemic manifestations are treated with steroids, non-steroidal anti-inflammatory drugs, disease-modifying anti-rheumatic drugs (DMARDs), hydroxychloroquine, or cytotoxic agents. These therapies are marginally effective and have potentially serious side-effects.

In this study, B-cell depletion by CD20 mAb treatment of Id3^–/– mice improved the secretory function of salivary glands and reduced the frequency of cell infiltrations in lachrymal and salivary glands. Previously, we demonstrated that lymphocytes, in particular T lymphocytes, play an essential role in the development of the disease.⁸ The results from the current study strongly suggest that B cells play a major role in the pathogenesis of Sjögren's syndrome in Id3^–/– mice. The disease symptoms may not only be controlled but also be reversed by CD20 mAb treatment. This finding is consistent with the idea that autoreactive T cells may be involved in initiating the diseases at a young age, whereas autoreactive B cells may exacerbate the disease symptoms at an older age.

Recently, an open-label phase II study was undertaken to investigate the safety and efficacy of Rituximab (a chimeric human IgG1 anti-human CD20 mAb) in the treatment of patients with early (maximum 4 years from disease onset), active primary Sjögren's syndrome and patients with primary Sjögren's syndrome associated with mucosa-associated lymphoid tissue (MALT-type) lymphoma.¹⁹ In this phase II study, selective depletion of B cells led to improvement of both subjective and objective parameters of disease activity, including salivary and lachrymal gland function, in patients¹⁹. The efficacy of Rituximab was also suggested in several case reports of clinical studies.^20,21 However, a full-scale evaluation of the efficacy of Rituximab in treating primary Sjögren's syndrome is still lacking. Id3^–/– mice provide a unique animal model for further testing the effect of CD20 mAb on primary Sjögren's syndrome. We started CD20 mAb treatment when Id3^–/– mice had reached 6 months of age and already showed prominent signs of histopathologies in gland tissues. Both disease symptoms and age of Id3^–/– mice resemble an early stage of disease development in human patients. Our data suggest that CD20 mAb treatment may not only prevent further development of disease symptoms but also ameliorate disease symptoms. Therefore, this animal model is valuable for understanding the mechanism by which Rituximab controls and/or reverts disease symptoms in Sjögren's syndrome patients.

This study revealed significantly increased serum IgM, IgG2a and IgG3 in 6-month-old Id3^–/– mice. CD20 mAb treatment leads to a specific and significant reduction of serum IgG3 and improvement of disease symptoms. In a human study, levels of IgG and IgM did not change during CD20 mAb treatment.¹⁹ Previous studies involving CD20 mAb treatment of younger mice showed no effect on serum immunoglobulin levels.¹⁷ The effect of CD20 mAb treatment on IgG3 in Id3^–/– mice could be unique to this disease model. Perhaps IgG3-producing B cells in Id3^–/– mice are particularly sensitive to CD20 treatment. This does not exclude the possibility that the function rather than the number of IgG3-secreting B cells is affected by CD20 treatment. IgG3 is the major mouse IgG isotype produced in response to TI-2 antigens. Mouse IgG3 demonstrates superior binding to polysaccharide antigens, which results in enhanced activation of effector function, including complement activation and Fc receptor binding.²² Although the exact role for IgG3 in Sjögren's syndrome still remains to be clarified and the observed phenomenon may be unique to this animal model, our study of Id3^–/– mice provides a possible link between mouse IgG3 and the pathogenesis of autoimmune diseases. Future studies aimed at characterizing autoantibodies in this animal model are clearly needed for understanding the role of humoral immunity in disease pathogenesis.