| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Institut National de la Santé et de la Recherche Médicale Unité 131, Institut Paris-Sud sur les Cytokines, Clamart, France;
Unité d’Immunologie Virale and
Unité de Chimie Organique, Institut Pasteur, Paris, France;
Service d’Anatomie et de Cytologie Pathologiques, UPRESA 3102, Hôpital Robert Debré, Paris, France; and
¶ DNAX Research Institute, Palo Alto, CA
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
9 anti-phosphatidylcholine autoantibodies, some of which express the RidA Id (3, 10, 11). B1a lymphocytes play an important role in the production of pathogenic autoantibodies (2, 12, 13). They are also an important source of mucosal IgA (6, 14, 15). Defining the mechanisms involved in the generation, migration, and differentiation of B1a lymphocytes into Ab-producing cells is thus crucial to our understanding of innate and mucosal immunity and of B lymphocyte-mediated autoimmune diseases. (New Zealand Black x New Zealand White)F1 (NZB/W) hybrid mice spontaneously develop a severe autoimmune disease that closely mimics human systemic lupus erythematosus (SLE) (16, 17). These mice have Abs that react against their own nuclear molecules, including anti-dsDNA IgG, leading to progressive and fatal glomerulonephritis. Abnormal expansion of the autoreactive B1a lymphocyte population in the peritoneal cavity (PerC) and in the spleen is one of the features of NZB/W mice (12, 13, 18). This abnormality is due to the NZB genetic background, because NZB mice also display increased numbers of B1a lymphocytes (13, 17, 18). In NZB/W mice, the elimination of PerB cells by the i.p. injection of water delays the onset and reduces the severity of lupus nephritis (19). The introduction of an X-linked immunodeficient gene (xid), causing a preferential depletion of the B1 cell subpopulation, prevents autoimmunity in NZB/W mice (20). PerB1 lymphocytes also play an important role in a B cell receptor-transgenic model of IgM-mediated autoimmune hemolytic anemia (21).
IL-10 stimulates the growth of normal and malignant CD5+ B lymphocytes (22, 23). Because B1a lymphocytes are considered to be the main source of B cell-derived IL-10 in the mouse, this cytokine may act as an autocrine growth factor for these cells (24). In normal mice, IL-10 neutralization leads to the depletion of PerB1a lymphocytes (25). IL-10 may be involved in the abnormal expansion of the PerB1a lymphocyte compartment in NZB/W mice because the neutralization of IL-10 prevents the onset of the autoimmune disease (26). A similar mechanism may occur in human SLE, in which IL-10 is produced in abnormally large amounts by B lymphocytes (27) and contributes to the production of autoantibodies (28). This mechanism may explain the rapid improvement of clinical symptoms in patients treated with an anti-IL-10 mAb (29).
Chemokines also affect PerB lymphocytes. In vitro, B lymphocyte chemoattractant (BLC/CXCL13) attracts PerB1a lymphocytes from both normal and NZB/W mice (4, 30, 31, 32, 33), and it is required for B1 cell homing to body cavities (33). Since BLC is produced in the cellular infiltrates of kidneys in aged NZB/W mice, it may contribute to the accumulation of activated B1a lymphocytes in inflamed tissues in such mice (31). Stromal cell-derived factor 1 (SDF-1/CXCL12) is involved in the local accumulation of B1a lymphocytes in the PerC of normal mice and in their self-renewal at this site: mesothelial cells constitutively produce SDF-1 and the in vivo neutralization of SDF-1 depletes the PerB1 lymphocyte compartment (4). Several mechanisms contribute to the effects of SDF-1 on PerB lymphocytes: SDF-1 attracts these cells, stimulates their proliferation, and promotes their survival (4). All effects of SDF-1 on PerB lymphocytes depend on the constitutive production of IL-10 by peritoneal cells and their intensity is modulated by the local concentration of IL-10 (32). Therefore, the role of IL-10 in the self-renewal of PerB1a lymphocytes in normal mice consists of sensitizing these cells to SDF-1. This raises the possibility that in NZB/W mice, the role of IL-10 in the autoimmune process is to facilitate the SDF-1-dependent expansion of the autoreactive B lymphocyte population. If this were the case, neutralization of SDF-1 in vivo should improve autoimmune symptoms in NZB/W mice. These hypotheses were tested in the present study.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
NZB, NZW, BALB/c, and C57BL/6 mice were purchased from IFFA-CREDO (L’Abresle, France). NZB/W F1 mice were bred from NZB females and NZW males. Only NZB/W female mice were used in this study. IL-9-transgenic (Tg5) and control (FVB) mice were kindly provided by J. C. Renauld (34).
Treatment of mice
For preventive treatment, groups of 10 NZB/W mice were treated with either 1B13A, a rat IgG1 mAb specific for murine IL-10R, or K15C, a mouse IgG2a mAb specific for SDF-1. As a control, five NZB/W mice were treated with GL113, a rat IgG1 recognizing Escherichia coli 
-galactosidase, and five mice received a mouse IgG2a anti-Haemophilus influenzae A mAb (Clonatec, Paris, France). Similar findings were obtained with these two control mAbs. From three wk of age, mice were injected with 250 µg of 1B13A, 100 µg of K15C, or with the same dose of the corresponding isotype-matched control mAbs. Intraperitoneal injections were given twice per week for 30 wk. For curative treatment, groups of eight NZB/W mice with mild proteinuria (2+, 100 mg/dl) were randomized and treated blindly with either the K15C anti-SDF-1 mAb or the isotype control (100 µg/injection) injected i.p. twice a week for 10 wk.
Evaluation of renal disease
Proteinuria was titrated using Multistix 8 SG (Bayer Diagnostics, Puteaux, France) on a 0–4+ scale, corresponding to the following approximate protein concentrations: 0, negative or trace; 1+, 30 mg/dl; 2+, 100 mg/dl; 3+, 300 mg/dl; and 4+, 
2000 mg/dl. Mice were considered to have severe proteinuria if two consecutive urine samples scored 3+ (300 mg/dl). The severity of glomerulonephritis was graded according to the World Health Organization (WHO) classification and scored according to the morphologic index of glomerular activity (35). For immunofluorescence studies, frozen kidney sections were stained with a FITC-conjugated rabbit anti-human IgG Ab (Dakopatts, Glostrup, Denmark) or with the biotinylated-K15C anti-SDF-1 mAb followed by FITC-streptavidin (Tebu, Le Perray-en-Yvelines, France). Binding of the purified rat anti-mouse CD34 mAb (clone RAM34; BD Biosciences, Rungis, France) was revealed using the rhodamine (tetramethylrhodamine isothiocyanate)-conjugated F(ab')2 donkey anti-rat IgG (H + L; Jackson ImmunoResearch, West Grove, PA). Binding of the polyclonal rabbit anti-human podocin Ab was revealed using the tetramethylrhodamine isothiocyanate-conjugated F(ab')2 goat anti-rabbit IgG (H + L; Jackson ImmunoResearch) (36).
ELISA and serological assays
Serum Abs specific for dsDNA were quantified by ELISA as previously described (26). Anti-dsDNA Ab titers were expressed in micrograms per milliliter and were determined from a standard curve established with the F4.1 mouse anti-DNA mAb. The dsDNA was obtained from Sigma-Aldrich (St. Louis, MO) and the F4.1 anti-DNA mAb was kindly provided by T. Terninck (Institut Pasteur, Paris France). Serum Abs bearing the RidA Id related to VH11/V9-type Abs were quantified by ELISA (10). Capture was performed with the rat RidA mAb (kindly provided by S. Kawaguchi, Shimane Medical University, Izumo, Japan) and RidA Id+ Abs were titrated using 1 µg/ml biotinylated RidA mAb. RidA Id+ titers were expressed as units per milliliter using a reference-positive standard of pooled serum from untreated 7-mo-old NZB/W female mice. A 1/200 dilution of this standard serum was arbitrarily assigned a value of 100 U/ml. Total concentrations of IgM and IgG were assessed as previously described (37). Standard curves were obtained with murine IgM or IgG (Sigma-Aldrich).
Cell preparation and mAbs for flow cytometry studies
Freshly isolated cells from 7- to 10-wk-old females or from mAb-treated mice were stained to determine the proportions of B1a, B1b, and B2 cells by four-color flow cytometry (FACScan; BD Biosciences) using the following mAbs: PE- or allophycocyanin-labeled rat anti-Mac-1 (clone M1/70), PE-labeled rat anti-CD43 (clone S7), PE-cyanin 5-labeled rat anti-CD5 (clone 53-7.3), and FITC- or PE-conjugated rat anti-CD19 (clone 1D3). Activated T lymphocytes were characterized using the following mAbs: FITC-conjugated rat anti-CD69 (clone H1.2F3), PE-labeled rat anti-CD25 (clone PC61), and PE-cyanin 5-labeled rat anti-CD4 (clone RM4-5). CXCR4 expression was studied using the PE-labeled mouse anti-human CXCR4 mAb (clone 12G5) and the corresponding isotype-matched control mAb (mouse IgG2a, clone G155-178) as previously described (38). All mAbs were purchased from BD Biosciences.
Functional evaluation of chemokine receptors
For cytoskeleton rearrangement studies, peritoneal cells were stained with fluorescent mAbs, incubated at 37°C for 2 h in RPMI 1640 medium supplemented with 20 mM HEPES at 6 x 106 cells/ml. The change in mean fluorescence intensity induced by SDF-1 (250 ng/ml) was monitored by flow cytometry as previously described (4, 32). For cell migration experiments, the Transwell system (Corning Costar, Brumath, France) was used as previously described (32), except that 3–5 x 105 cells were added to the upper chamber. Unless specified, SDF-1 was used at a concentration of 1 µg/ml and IL-10 (R&D Systems, Minneapolis, MN) at a concentration of 100 ng/ml. The anti-IL-10R and control Abs (R&D Systems) were used at a concentration of 10 µg/ml. The chemokine was added to the lower chamber only, whereas IL-10 and the Abs were added to both the upper and lower chambers. Cells recovered in the lower chamber were characterized and counted by flow cytometry. The fraction of PerB1a cells migrating across the membrane was calculated as follows: number of PerB1a lymphocytes migrating to the lower chamber/number of PerB1a lymphocytes added to the upper chamber at the start of the assay.
Proliferation and survival assays
For proliferation and survival experiments, the CFSE and the 3,3'-dihexylocarbocyanin iodide dye assays were performed as previously described (4, 32). SDF-1 and IL-10 were used at a concentration of 1 µg/ml and 100 ng/ml, respectively.
Statistical analysis
Results were compared using the Mann-Whitney U test or the log rank test for proteinuria and survival curves.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We analyzed the chemotactic response to SDF-1 of PerB1a lymphocytes from NZB/W mice. This response was compared with that of three strains of normal mice and of IL-9-transgenic mice which display an expansion of the PerB1b lymphocyte compartment (34). SDF-1 stimulated the migration of PerB1a lymphocytes from all the strains tested. However, the fraction of PerB1a lymphocytes migrating in response to SDF-1 was higher in NZB/W mice than in any other strain (Fig. 1A). This difference was observed for all concentrations of SDF-1. As little as 50 ng/ml SDF-1 was as efficient in triggering migration of NZB/W cells as 1 µg/ml SDF-1 was in control mice (Fig. 1B). We further investigated the response of PerB1a lymphocytes by comparing SDF-1-induced cytoskeleton rearrangement in NZB/W mice and BALB/c mice. The level of SDF-1-induced actin polymerization in PerB1a lymphocytes was about twice as high in NZB/W mice than in control mice. PerB1b and PerB2 lymphocytes from NZB/W mice also responded to SDF-1, but this response was weaker than that of PerB1a lymphocytes and it did not differ between NZB/W and control mice (Fig. 1C).
|
We evaluated the contribution of the NZB and NZW genetic backgrounds to the hypersensitivity to SDF-1 of NZB/W hybrids by comparing SDF-1-induced cytoskeleton rearrangement in NZB and NZW PerB1a lymphocytes. Stronger SDF-1-triggered actin polymerization was observed in NZB than in NZW or BALB/c cells. The level of actin polymerization was even higher in NZB than in NZB/W cells (Fig. 1D). Therefore, the hypersensitivity to SDF-1 of PerB1a lymphocytes in NZB/W mice is due to the NZB rather than to the NZW genetic background.
To test whether increased sensitivity to SDF-1 in NZB and NZB/W mice could be related to a different level of expression of the SDF-1 receptor, we analyzed by flow cytometry the expression of CXCR4. Higher numbers of PerB1a lymphocytes expressed CXCR4 in NZB/W mice than in NZW or BALB/c mice. However, NZB/W mice also expressed this receptor at a higher level than NZB mice. The increased expression of CXCR4 was not restricted to B1a lymphocytes in NZB/W mice because CD5- B lymphocytes also expressed higher levels of CXCR4 in NZB/W mice than in NZB, NZW, or BALB/c mice (Table I).
|
We tested whether the deregulated response to SDF-1 of PerB1a cells in NZB/W mice remained sensitive to IL-10. Addition of IL-10 enhanced the response to SDF-1 of PerB1a lymphocytes from normal mice. It also strongly enhanced the response of cells from NZB/W mice. The increased response of NZB/W B1a cells to SDF-1 persisted in the presence of 100 ng/ml IL-10: >70% of the cells migrated in response to SDF-1 vs <30% for the other four strains tested in parallel (Fig. 1A). Neither IL-5, IL-6, nor IL-9 potentiated the response of NZB/W PerB1a lymphocytes to SDF-1 (data not shown). PerB1a lymphocytes constitutively produce IL-10 in normal and NZB/W mice (Refs. 24, 32, 39 and data not shown). To investigate whether this autocrine IL-10 production influenced SDF-1 effects, the SDF-1-induced migration of NZB/W PerB1a lymphocytes was tested in the presence of an anti-IL-10R Ab or a control Ab. The neutralization of endogenous IL-10 decreased the chemotactic response to SDF-1 (Fig. 1E). The SDF-1-induced cytoskeleton rearrangement in NZB/W PerB1a lymphocytes was also reduced by the anti-IL-10R Ab (data not shown). Overall, these data show that PerB1a lymphocytes from NZB/W mice are hypersensitive to the effect of SDF-1. This abnormality is not observed with other B lymphocyte subpopulations. Despite this increased sensitivity, the level of response of NZB/W PerB1a lymphocytes to SDF-1 remains modulated by IL-10, either produced endogenously or added to the cells.
SDF-1 and IL-10 stimulate the proliferation and improve the survival of NZB/W PerB1a lymphocytes
The effects of SDF-1 and IL-10 on the proliferation and survival of NZB/W PerB1a lymphocytes were investigated. SDF-1 increased the proliferation of NZB/W PerB1a lymphocytes and prevented their spontaneous apoptosis. The level of proliferation and protection from cell death induced by SDF-1 was higher in NZB/W mice than in BALB/c mice. Addition of IL-10 potentiated the effects of SDF-1 in both BALB/c and NZB/W mice. In this condition, the ability of SDF-1 to stimulate proliferation and to prevent cell death remained stronger in NZB/W mice than in BALB/c mice (Fig. 2).
|
Production of SDF-1 was evaluated by immunofluorescence studies in tissues of NZB/W mice. SDF-1 was produced by mesothelial cells in the PerC and by stromal cells in the spleen (Fig. 3). However, this production did not significantly differ from that of BALB/c mice, either in terms of intensity or distribution (data not shown). The presence of SDF-1 was analyzed in kidneys of 10-, 22-, and 32-wk-old NZB/W mice (n = 3 in each group). At each age, three BALB/c mice were tested as controls. Regardless of their age, the staining with the anti-SDF-1 mAb was weak or absent in BALB/c mice (Fig. 4A). There was a strong staining of SDF-1 in the glomeruli of 22- and 32-wk-old NZB/W mice (Fig. 4B). This staining was specific, as it disappeared when the anti-SDF-1 mAb was omitted (data not shown) or when a 100-fold molar excess of SDF-1 was added to the mAb before incubation with tissue sections (Fig. 4C). In contrast to older NZB/W mice, 10-wk-old NZB/W mice did not express larger amounts of SDF-1 than matched controls (Fig. 4D).
|
|
SDF-1 neutralization early in life prevents proteinuria and death in NZB/W mice
We evaluated the contribution of SDF-1 and IL-10 to the development of lupus by treating NZB/W mice with either an anti-IL-10R mAb or an anti-SDF-1 mAb from week 3 to week 32 of life. Control mice were treated with isotype-matched Abs. Treatment with the anti-IL-10R mAb delayed proteinuria (p < 0.002) and death (p < 0.02) with respect to control mice. Treatment with the anti-SDF-1 mAb also delayed the onset of proteinuria and prevented death (p < 0.002 vs control mice; Fig. 5, A and B). We determined in these mice the serum concentration of anti-dsDNA IgG autoantibodies. In week 20, this concentration was higher in control mice than in the other two groups (p < 0.01) and it continued to increase until death. In anti-IL-10R mAb-treated mice, the concentration of anti-dsDNA IgG remained stable between weeks 20 and 24, but increased between weeks 24 and 30. In the anti-SDF-1 mAb-treated group, the concentration of anti-dsDNA IgG remained low and stable up to week 30 (Fig. 5C). In contrast to anti-dsDNA IgG, the serum concentration of total IgG did not differ between controls and mice treated with either the anti-IL-10R mAb or the anti-SDF-1 mAb (Fig. 5D). Therefore, early neutralization of either IL-10 or SDF-1 effects prevents the onset of autoimmunity in NZB/W mice.
|
We then investigated whether the administration of an anti-SDF-1 mAb initiated later, in NZB/W mice with already established nephritis, was able to reverse glomerular damages. Mice with grade 2+ proteinuria were randomized to receive either a control mAb or an anti-SDF-1 mAb for 10 wk. In mice treated with the control mAb, proteinuria continued to increase during the next 10 wk of follow-up. One mice died 6 wk after treatment initiation. In contrast, proteinuria rapidly decreased in mice treated with the anti-SDF-1 mAb. The difference between the two groups became significant as early as 2 wk after treatment initiation and remained so until the end of the follow-up (Fig. 6A). Ten weeks after treatment initiation, six of seven mice had 300 mg protein/dl in the control group, whereas six of eight mice had <100 mg protein/dl in the anti-SDF-1 mAb-treated group. Pathological changes in the kidneys were analyzed at the end of treatment in six mice of each group. In the control group, diffuse proliferative glomerulonephritis (grade IV of the WHO classification) was observed in five of the six mice, with large deposits extending to the peripheral capillary wall. In the anti-SDF-1 mAb-treated group, only one mouse had a grade IV glomerulonephritis. Glomerular changes were milder in the other five mice: lesions were purely mesangial, with minimal deposits and little or no cellular infiltration, and there was no abnormality in the capillary walls (grade IIA and IIB; Fig. 6, B and C, and Table II). The activity index of glomerular lesions was 11.0 ± 2.1 in control mice, whereas it was 3.1 ± 1.4 in anti-SDF-1 mAb-treated mice (p = 0.01). Immunofluorescence staining of kidney sections showed abundant granular IgG deposits on the subendothelial capillary wall in five of the six control mice along with mesangial deposits. In five of the six anti-SDF-1 mAb-treated mice, there was either no IgG deposits or a few deposits limited to the mesangial area (Fig. 6, D and E).
|
|
In the above experiment, the fraction of CD4+ T lymphocytes expressing CD69 or CD25 in the spleen and in the peripheral lymph nodes was compared at the end of the 10-wk follow-up between anti-SDF-1 mAb-treated mice and controls. Similar findings were observed for both markers and in both secondary lymphoid organs: the fraction of activated CD4+ T lymphocytes was significantly lower in anti-SDF-1 mAb-treated mice than in controls (Fig. 7, A and B).
|
The regression of renal symptoms in mice treated with the anti-SDF-1 mAb was correlated with changes in circulating anti-dsDNA IgG concentrations. Eight weeks after treatment initiation, the serum concentration of autoantibodies was lower in anti-SDF-1 mAb-treated mice than in controls (Fig. 7C). This decrease was specific to autoreactive IgG because the serum concentration of total IgG did not differ between the two groups (Fig. 7D). Contrasting with total IgG, the concentration of circulating IgM significantly decreased during treatment with the anti-SDF-1 mAb (Fig. 7E). As circulating IgM mostly originate from B1a lymphocytes (7, 8), this suggested that SDF-1 neutralization in NZB/W mice preferentially targeted the B1a lymphocyte compartment. This hypothesis was further tested by two different approaches. First, we evaluated the effect of the anti-SDF-1 mAb administration on the serum concentration of B1a lymphocyte-derived RidA Id+ IgM. The concentration of RidA Id+ Abs was lower in anti-SDF-1 mAb-treated mice than in controls (Fig. 7F). This difference appeared early during treatment, as it was already observed 2 wk after treatment initiation. The effect of SDF-1 neutralization on B lymphocyte compartments was directly determined by enumerating at the end of treatment B1a, B1b, and B2 cells in the PerC and in the spleen. In both locations, the number of B1a lymphocytes was significantly lower in anti-SDF-1 mAb-treated mice than in controls. A similar finding was observed for B1b lymphocytes. In contrast, SDF-1 neutralization did not alter the number of B2 cells recovered from the PerC or from the spleen (Fig. 7, G and H). Therefore, regression of glomerular damages in NZB/W mice treated with an anti-SDF-1 mAb is associated with a large decrease in the level of circulating autoantibodies and with a contraction of the B1a lymphocyte compartment.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several functional abnormalities of the B1a lymphocyte subpopulation have been reported in NZB/W mice (7, 17, 39, 40, 41, 42). The increased sensitivity of PerB1a lymphocytes to SDF-1 provides another example of the abnormal behavior of this subpopulation in NZB/W mice. As in humans and in normal mice (4, 43), mesothelial cells from the PerC of NZB/W mice produce SDF-1. Given the role of SDF-1 in the persistence of B1a lymphocytes in the PerC (4), the increased sensitivity of PerB1a lymphocytes to SDF-1 may account for their massive accumulation in NZB/W mice. The finding that only B1a lymphocytes are hypersensitive to SDF-1 is consistent with the observation that only the B1a cell subpopulation is abnormally expanded in NZB/W mice (Refs. 12, 13, 18 and data not shown). The strong response of PerB1a lymphocytes from NZB/W mice to the chemoattractant effect of SDF-1 may increase the retention of B1a lymphocytes in the PerC, an environment favorable for their survival and expansion. The effects of SDF-1 on the proliferation of PerB1a cells and on their survival were also stronger in NZB/W than in BALB/c mice. These combined effects may all contribute to the abnormal expansion of the B1a lymphocyte population in the peritoneal cavity of NZB/W mice. A similar phenomenon may account for the large accumulation of B1a lymphocytes in the spleen of NZB/W mice where SDF-1 is produced by stromal cells. This would explain why the number of B1a lymphocytes decreases in the spleen of NZB/W mice treated with the anti-SDF-1 mAb.
As in NZB/W mice, B1a lymphocytes are hypersensitive to SDF-1 in NZB mice. In contrast, the response of PerB1a lymphocytes is within the normal range in NZW mice. Consistent with these in vitro findings, massive accumulation of B1a lymphocytes is observed in the PerC of NZB mice, but not in NZW mice (Refs. 13, 17, 18 and data not shown). This indicates that the abnormal sensitivity to SDF-1 of NZB/W mice originates from the NZB genetic background. The absence of overt lupus in NZB mice indicates that expansion of the B1a lymphocyte subpopulation is not sufficient to explain autoimmune disease in NZB/W mice and that NZW-derived genetic abnormalities also contribute to emergence of the disease (44, 45). The molecular mechanism of the increased sensitivity to SDF-1 in NZB and NZB/W mice is unclear. The fraction of PerB1a lymphocytes expressing CXCR4 is higher in NZB/W mice than in NZW or BALB/c mice. However, it is also higher than in NZB mice, which express this receptor at the same level as control mice. Moreover, the increased expression of CXCR4 by B lymphocytes from NZB/W mice is not restricted to B1a lymphocytes and extends to CD5- B cells, which are not abnormally sensitive to SDF-1. Therefore, the hypersensitivity to SDF-1 of B1a lymphocytes from NZB and NZB/W mice cannot be accounted for by an increased expression of CXCR4 only, and it should involved postreceptor events. Such a dissociation between expression level and function of CXCR4 has been already described in a number of conditions in humans (46, 47, 48). Since IL-10 acts in synergy with SDF-1, one hypothesis would be that the increased response to SDF-1 actually reflects an hypersensitivity to IL-10. However, there is no increased response in NZB/W mice to IL-10 alone, arguing against such a hypothesis. Rather, the abnormal behavior of B1a cells may be due to events downstream of CXCR4 itself.
The positive interaction between IL-10 and SDF-1 provides a unifying mechanism accounting for the combined contributions of IL-10 and SDF-1 to the pathophysiology of lupus in NZB/W mice. The involvement of IL-10 in the development of murine lupus was demonstrated by administering an anti-IL-10 or an anti-IL-10R mAb to NZB/W mice early in life. Such treatment delayed the onset of the disease (Ref. 26 and this work). We now demonstrate that the treatment of NZB/W mice with a neutralizing anti-SDF-1 mAb also prevents development of the autoimmune disease. This treatment inhibited autoantibody production as long as it was administered, delaying both the onset of proteinuria and death. The ability to prevent lupus by antagonizing either IL-10 or SDF-1 effects argues in favor of the hypothesis that the combined action of IL-10 and SDF-1 on B1a lymphocytes, demonstrated in vitro, also occurs in vivo in NZB/W mice.
Possibly, a similar interaction between IL-10 and SDF-1 may be involved in other forms of IL-10-dependent B1a-related immune disorders. NZB mice produce anti-RBC, anti-lymphocyte, and anti-ssDNA IgM autoantibodies, leading to hemolytic anemia and mild glomerulonephritis. When aged, they develop IL-10-dependent CD5+ B cell lymphomas with a high frequency (23, 49). Fas-deficient antierythrocyte Ig transgenic mice develop autoimmune hemolytic anemia, and activation of B1 lymphocytes by IL-10 is required in this process. However, the effect of IL-10 in this model may be indirect (50, 51, 52). It would be interesting to determine in these conditions whether neutralization of SDF-1 influences B1-related immunological abnormalities to the same extent as IL-10 neutralization.
When the anti-SDF-1 mAb was administered to NZB/W mice suffering from established lupus nephritis, the symptoms of renal disease rapidly improved. Proteinuria started to decrease as early as 2 wk after the initiation of treatment. In week 10, kidney lesions were minimal and mostly inactive in anti-SDF-1 mAb-treated mice and there were few anti-dsDNA autoantibodies. This is, to our knowledge, the first effective treatment of lupus nephritis by a chemokine antagonist.
Several mechanisms may have contributed to the reversal of nephritis. In normal mice, SDF-1 is involved in the homing of plasma cells (53, 54), raising the hypothesis that the beneficial effect of SDF-1 neutralization in NZB/W mice resulted from the impaired function or survival of terminally differentiated B lymphocytes. However, such a mechanism is unlikely to explain our findings because the anti-SDF-1 mAb treatment selectively inhibited the concentration of circulating anti-dsDNA IgG, while sparing total IgG levels. Therefore, if SDF-1 neutralization directly counteracts the production of IgG by plasma cells, this effect should be restricted to the autoreactive subpopulation of B cells.
Circulating IgM mainly originate from the B1a lymphocyte subpopulation (7, 8); this is particularly true for IgM expressing the RidA Id (10). In contrast, circulating IgG are mainly derived from B2 lymphocytes. Analyses of circulating IgM, RidA Id+ IgM, and IgG suggests that SDF-1 neutralization selectively targeted the B1a cell subpopulation, sparing B2 lymphocytes. This was directly shown by enumerating B lymphocyte subpopulations in the PerC and in the spleen of treated mice. At both sites, SDF-1 neutralization decreased the numbers of B1a lymphocytes, whereas B2 lymphocyte numbers remained unchanged. The ability of SDF-1 neutralization to decrease the size of the B1a lymphocyte compartment and to improve the autoimmune disease supports the idea, already raised from previous studies (19, 20, 52), that the abnormal expansion of the B1a cell subpopulation plays a significant role in the autoimmune process. Since the number of B1a lymphocytes decreased simultaneously with the concentration of circulating autoantibodies, this provides a simple explanation for the reversal of nephritis in anti-SDF-1 mAb-treated mice. However, B1a lymphocytes may also act as APC, triggering the activation of autoreactive T cells, which are involved in tissue damage in murine lupus (40, 55, 56, 57). In humans, SDF-1 contributes to T lymphocyte activation (58, 59). SDF-1 neutralization in NZB/W mice strongly decreased the level of activation of T lymphocytes in secondary lymphoid organs, and this may also have contributed to improve renal symptoms. It is also possible that SDF-1 in NZB/W mice targeted other cells in addition to B1a and T lymphocytes. For instance, SDF-1 neutralization in such mice decreased the number of peritoneal macrophages (data not shown). An effect of SDF-1 on macrophages or on other APC may also contribute to the disease improvement.
Interestingly, SDF-1 was produced in situ in the inflamed glomeruli. In NZB/W mice with overt nephritis, podocytes and, to a lesser extent, endothelial and mesangial cells produced large amounts of SDF-1, whereas BALB/c mice did not. This deregulation of SDF-1 production seems to be related to the development of nephritis because it was not observed in young NZB/W mice. SDF-1 was initially considered as a homeostatic chemokine, produced constitutively in uninflamed tissues (60). Our results, as well as some recent studies in other conditions, challenge this view (61, 62, 63). Combined with the hypersensitivity of B1a lymphocytes in NZB/W mice, SDF-1 production in the glomeruli suggests a mechanism of recruitment of autoreactive B lymphocytes to inflamed glomeruli, in which these cells may produce autoantibodies. Pathological changes in kidneys of NZB/W mice with nephritis are associated with an increased in situ production of SDF-1 and BLC (31), two homeostatic chemokines important in the homing and expansion of the B1a cell subpopulation (4, 33). The monocyte chemoattractant protein 1 (CCL2) chemokine may also be involved in the nephritis of the MRL/lpr lupus model (64). Another potential consequence of SDF-1 production in the glomeruli is direct tissue damage. Human podocytes express CXCR4 in vitro and their activation by SDF-1 increases NADPH-oxidase activity, leading to the production of reactive oxygen intermediates, which may be involved in glomerular damage (65). A similar phenomenon may occur in glomeruli of NZB/W mice.
Therefore, the reversal of nephritis in NZB/W mice treated with an anti-SDF-1 mAb may result from a dual systemic and local effect. SDF-1 neutralization decreases B1a lymphocyte numbers in the PerC and in the spleen, the level of circulating autoantibodies, and the intensity of T lymphocyte activation in secondary lymphoid organs. In the inflamed kidneys, neutralization of locally produced SDF-1 may prevent homing of autoreactive effector cells as well as SDF-1-induced glomerular damage. The role of IL-10 in tuning the SDF-1 response of B1a lymphocytes enlightens the mechanism by which the neutralization of this cytokine improves disease symptoms. The complicity between SDF-1 and IL-10 is a key element of the mechanism by which pathogenic autoantibodies are produced, and it is a target for therapeutic effect in a murine model of SLE. The rapid improvement of disease symptoms in SLE patients treated with an anti-IL-10 mAb suggests that a similar phenomenon may occur in this condition and that antagonists of SDF-1 or of its receptor, CXCR4, may also be of value in the treatment of human lupus.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Dynavax Technologies, 717 Potter Street, Suite 100, Berkeley, CA 94710.
3 Address correspondence and reprint requests to Dr. Dominique Emilie, Institut National de la Santé et de la Recherche Médicale Unité 131, Institut Paris-Sud sur les Cytokines, 32 rue des Carnets, 92140 Clamart, France. E-mail address: emilie{at}ipsc.u-psud.fr
4 Abbreviations used in this paper: PerB, peritoneal B; BLC, B lymphocyte chemoattractant; PerC, peritoneal cavity; SDF-1, stromal cell-derived factor 1; SLE, systemic lupus erythematosus.
Received for publication September 30, 2002. Accepted for publication January 10, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
9 B cell antigen receptor drives generation of CD5+ B cells both in vivo and in vitro. J. Immunol. 164:4586.B-inducing kinase (NIK) causes defects in secondary lymphoid tissue chemokine receptor signaling and homing of peritoneal cells to the gut-associated lymphatic tissue system. J. Exp. Med. 191:1477.
receptor-deficient mice. J. Immunol. 167:4686./
T cells in antierythrocyte immunoglobulin transgenic mice. J. Exp. Med. 192:1577. 1 on synovial T cells contributes to their accumulation within the rheumatoid synovium. J. Immunol. 165:3423.Related articles in The JI:
This article has been cited by other articles:
|
|
 |
|
  A. Wang, A.-M. Fairhurst, K. Tus, S. Subramanian, Y. Liu, F. Lin, P. Igarashi, X. J. Zhou, F. Batteux, D. Wong, et al. CXCR4/CXCL12 Hyperexpression Plays a Pivotal Role in the Pathogenesis of Lupus J. Immunol., April 1, 2009; 182(7): 4448 - 4458. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Meiron, Y. Zohar, R. Anunu, G. Wildbaum, and N. Karin CXCL12 (SDF-1{alpha}) suppresses ongoing experimental autoimmune encephalomyelitis by selecting antigen-specific regulatory T cells J. Exp. Med., October 27, 2008; 205(11): 2643 - 2655. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Balabanian, B. Lagane, J. L. Pablos, L. Laurent, T. Planchenault, O. Verola, C. Lebbe, D. Kerob, A. Dupuy, O. Hermine, et al. WHIM syndromes with different genetic anomalies are accounted for by impaired CXCR4 desensitization to CXCL12 Blood, March 15, 2005; 105(6): 2449 - 2457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Sela, R. Hershkoviz, L. Cahalon, O. Lider, and E. Mozes Down-Regulation of Stromal Cell-Derived Factor-1{alpha}-Induced T Cell Chemotaxis by a Peptide Based on the Complementarity-Determining Region 1 of an Anti-DNA Autoantibody via Up-Regulation of TGF-{beta} Secretion J. Immunol., January 1, 2005; 174(1): 302 - 309. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Zou, B. Barnett, H. Safah, V. F. LaRussa, M. Evdemon-Hogan, P. Mottram, S. Wei, O. David, T. J. Curiel, and W. Zou Bone Marrow Is a Reservoir for CD4+CD25+ Regulatory T Cells that Traffic through CXCL12/CXCR4 Signals Cancer Res., November 15, 2004; 64(22): 8451 - 8455. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Tarlinton and P. D. Hodgkin Targeting Plasma Cells in Autoimmune Diseases J. Exp. Med., June 7, 2004; 199(11): 1451 - 1454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Hecht, R. Hershkoviz, S. Shivtiel, T. Lapidot, I. R. Cohen, O. Lider, and L. Cahalon Heparin-disaccharide affects T cells: inhibition of NF-{kappa}B activation, cell migration, and modulation of intracellular signaling J. Leukoc. Biol., June 1, 2004; 75(6): 1139 - 1146. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Atencio, H. Amano, S. Izui, and B. L. Kotzin Separation of the New Zealand Black Genetic Contribution to Lupus from New Zealand Black Determined Expansions of Marginal Zone B and B1a Cells J. Immunol., April 1, 2004; 172(7): 4159 - 4166. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ito, S. Ishikawa, T. Sato, K. Akadegawa, H. Yurino, M. Kitabatake, S. Hontsu, T. Ezaki, H. Kimura, and K. Matsushima Defective B1 Cell Homing to the Peritoneal Cavity and Preferential Recruitment of B1 Cells in the Target Organs in a Murine Model for Systemic Lupus Erythematosus J. Immunol., March 15, 2004; 172(6): 3628 - 3634. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), C57BL/6 (
), FVB (
), Tg5 (
), and NZB/W (
) mice was evaluated. Results (mean ± SEM) are expressed as the percentage of input PerB1a lymphocytes that migrated to the lower chamber (three to seven experiments). *, p < 0.05 and **, p < 0.005, as compared with all other strains. B, Chemotaxis of NZB/W (
) and from BALB/c (
), NZW (
), and NZB (•) mice were tested for SDF-1-induced cytoskeleton rearrangement. Results (mean ± SEM) are from three to five independent experiments. *, p < 0.05, vs BALB/c and NZW mice; #, p < 0.05, vs NZB/W mice. E, SDF-1-induced chemotaxis of NZB/W PerB1a cells in the presence of an anti-IL-10R Ab or a control Ab was evaluated. Results (mean ± SEM) are from three independent experiments. *, p < 0.05, vs cells migrating in the presence of the control Ab.
) or an anti-SDF-1 mAb (
), an anti-SDF-1 (•), or a control (
) mAb. Sera were collected at 20, 24, and 30 wk of age. Each circle represents an individual mouse and lines show median values. 
, End of treatment.
