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 Erythematosus¹

The (NZB × NZW)F₁ (BWF₁) strain of mice spontaneously develop systemic autoimmune disorders resembling systemic lupus erythematosus (SLE)³ in humans (1, 2). It is characterized by production of a variety of IgG autoantibodies and massive deposition of immune complexes in glomeruli in the kidney. More than 95% of these mice die from renal failure before 12 mo of age. A marked mononuclear cell infiltration in the target organs, including the kidney and lung, is also observed in aged BWF₁ mice. We previously demonstrated that B lymphocyte chemoattractant (BLC; CXCL13) was highly and ectopically expressed by myeloid dendritic cells in the target organs including the thymus, kidney, and lung in aged BWF₁ mice, but not in aged NZB and NZW mice, and that B1 cells were preferentially chemoattracted toward BLC (3, 4).

B1 cells are a specialized cell population that are distinguished from conventional B cells (B2 cells) by their origin, cell surface phenotype, unique tissue distribution, and capacity for self-renewal, and have also been considered to be involved in autoantibody production in the development of autoimmune diseases (5, 6, 7). Involvement of B1 cells in IgM-mediated autoimmune diseases such as autoimmune hemolytic anemia was also demonstrated (8, 9). However, it remains to be elucidated whether B1 cells class-switch from IgM to IgG in the development of autoimmune diseases. Accumulating data also suggest that B1 cells contribute to the innate immunity by producing natural IgM Ab in the circulation and by secreting IgA Ab in the intestinal mucosa (10, 11, 12, 13). It has been recently reported that BLC is essential for B1 cell homing to the peritoneal cavity and for body cavity immunity (14). In the present study, we investigated a role of BLC expression in aged BWF₁ on aberrant B1 cell trafficking, including B1 cell homing to the peritoneal cavity. We found that B1 cell homing to the peritoneal cavity was impaired and preferential B1 cell trafficking to the target organs in aged BWF₁ mice. We also demonstrated that the number of BLC-producing peritoneal macrophages, which were reported to be a major cell source for BLC in the peritoneal cavity (14), was markedly decreased in aged BWF₁ mice. Pathological significance of abnormal B1 cell trafficking in the development of murine lupus is discussed.

(NZB × NZW)F₁ (BWF₁), NZB, NZW, and BALB/c mice, originally obtained from the Shizuoka Laboratory Animal Center (Shizuoka, Japan), were maintained in our animal facility at University of Tokyo. Female BWF₁ and BALB/c mice aged 8–10 wk were used as young mice, and BWF₁ mice aged 8–12 mo with moderate-to-severe proteinuria and BALB/c mice aged 8 mo were used as aged mice.

Rat mAbs specific for mouse CD11b (M1/70), CD11c (HL-3), B220/CD45R (RA3-6B2), CD5 (53-7.3RRH), MHC class II (ER-TR3), and CD16/32 (2.4G2) were purchased from BD PharMingen (San Diego, CA). Anti-F4/80 (CI: A3-1) mAb and goat anti-BLC polyclonal Ab (pAb) were purchased from Serotec (Raleigh, CA) and Genzyme Techne (Minneapolis, MN), respectively.

Peritoneal lavage cells were isolated by flushing the peritoneal cavity with 4–6 ml of RPMI 1640 supplemented with 5% FCS. Cell suspensions prepared from mouse spleen and lymph nodes were passed through nylon mesh. Peripheral blood leukocytes (PBLk) were obtained by centrifugation on Lymphorite-M (Cedarlane, Hornby, Ontario, Canada). RBC in the spleen and PBLk were lysed by ammonium chloride solution.

Flow-cytometric analyses of peritoneal cells, spleen cells, lymphoid cells, and PBLk were performed using an Epics Elite cell sorter (Coulter Electronics, Hialeah, FL) as described previously (3). After blocking FcR with anti-CD16/32 for 10 min, peritoneal cells were stained with FITC-conjugated anti-CD11b mAb and PE-conjugated anti-CD11c mAb, and biotin-conjugated anti-B220, F4/80, or class II mAb followed by allophycocyanin-streptavidin (BD PharMingen). CD11b^highCD11c^lowF4/80⁺ macrophages were sorted to >95% purity using an Epics Elite cell sorter. Sorted macrophages were spun down onto a slide glass and stained with Giemsa solution (Merck, Tokyo, Japan). In some experiments, peritoneal cells were strained with FITC-conjugated anti-CD5 mAb, PE-conjugated anti-B220 mAb, and biotin-conjugated anti-F4/80 mAb followed by allophycocyanin-streptavidin. CD5⁺B220⁺F4/80⁻ B1 cells in the peritoneal cavity and CD5⁻B220⁺F4/80⁻ B2 cells in the spleen were sorted on an Epics Elite cell sorter. More than 95% of B1 and B2 cells were CD19 positive.

Cell labeling of B1 cells with CFSE (Molecular Probes, Eugene, OR) and B2 cells with CMTMR (Molecular Probes) was performed according to the manufacturer’s instructions. These cells were all labeled in 10 μM at 37°C for 15 min. The cell viability after CFSE and CMTMR staining and before injection was >99% in dye exclusion tests using trypan blue. The mixture of 4 × 10⁶ CFSE-labeled peritoneal B1 cells and/or 4 × 10⁶ CMTMR-labeled splenic B2 cells suspended in 0.2 ml of PBS were injected i.v. into BWF₁ or BALB/c mice. In some experiments, the mixture of B1 cells from aged BWF₁ mice labeled with CFSE and B1 cells from young BWF₁ mice labeled with CMTMR suspended in 0.2 ml of PBS was injected i.v. into young BWF₁ mice. Mice were sacrificed 24 h after i.v. injection and subjected to flow-cytometric and immunofluorescent analysis. The number of cells localized in the various organs was calculated by the percentage of labeled cells in each organ, and the percentage of recruited cells to the total number of injected cells was presented using an Epics Elite cell sorter. In nonlymphoid organs, 10 nonconsecutive microscopic fields of cryostat section were examined, and the numbers of CFSE-labeled B1 cells and CMTMR-labeled B2 cells in each field were counted. The mean percentage of B1 cells among total number of fluorescent cells was presented.

For immunofluorescent analysis of target organs, spleen, lung, kidney, and thymus from aged BWF₁ mice were fixed in paraformaldehyde, embedded in Tissue-Tek OCT compound (Miles, Elkhart, IN), and then frozen in liquid nitrogen. Six-micron cryostat sections were then incubated with biotin-conjugated anti-B220 mAb followed by allophycocyanin-streptavidin. Avidin/biotin blocking solution (Vector Laboratories, Burlingame, CA) was used to decrease nonspecific staining. Finally, the sections were analyzed by Olympus IX-70 confocal laser-scanning microscope system (Olympus Optical, Tokyo, Japan).

Peritoneal macrophages obtained by cytospin centrifugation were stained with goat anti-BLC pAb, followed by biotinylated rabbit anti-goat IgG (DAKO, Carpinteria, CA) and HRP-labeled streptavidin (DAKO). Avidin/biotin blocking solution and normal rabbit IgG were used to decrease nonspecific staining.

Total RNA was isolated from peritoneal cells by using RNAzol B (Tel-Test, Friendswood, TX) according to manufacturer’s instructions. It was then reversely transcribed into cDNA using Superscript II preamplification kit (Invitrogen, Carlsbad, CA) and amplified with specific oligonucleotide primers. The sense and the antisense primers used were as follows: BLC primer, 5′-TCTCTCCAGGCCACGGTATTCT-3′ and 5′-ACCATTTGGCACGAGGATTCAC-3′, and GAPDH primer, 5′-AGTATGACTCCACTCACGGCAA-3′ and 5′-TCTCGCTCCTGGAAGATGGT-3′. PCR was performed by thermal cycler for 35 cycles of 94°C for 30 s, 58°C for 45 s, and 72°C for 45 s, and final extension was done at 72°C for 10 min. The PCR products of GAPDH and BLC were examined by 2.5% agarose gel electrophoresis. Real-time quantitative PCR analysis was performed by using ABI 7700 sequence detector system (PE Applied Biosystems, Foster City, CA). FAM-labeled primers were used as target hybridization probes for BLC and GAPDH (BLC, 5′-CATCATAGTTCGGATTCAAGTTACGCCCCC-3′; GAPDH, 5′-AAGGGACACAGTCAAGGCCGAGAAT-3′). The thermal cycling condition included 50°C for 2 min and 95°C for 10 min, followed by 45 cycles of amplification at 95°C for 15 s and 55°C for 1.5 min for denaturing and annealing, respectively. PCR were run in triplicate. BLC quantity was normalized by the level of GAPDH.

Statistical analysis was performed using Student’s t test. The 95% confidence limit was taken as significant.

We conducted B1 cell homing experiments where CFSE-labeled B1 or CMTMR-labeled B2 cells were injected i.v. into young or aged BWF₁ mice to examine the change in B1 cell trafficking during the development of lupus nephritis in these mice. A significant number of B1 cells homed to the peritoneal cavity in young BWF₁ mice when injected i.v. with CFSE-labeled B1 cells (Fig. 1,A). However, in aged BWF₁ mice, B1 cells failed to home to the peritoneal cavity (Fig. 1,B). In contrast, B1 cells did home to the peritoneal cavity in aged BALB/c mice as well as in young BALB/c mice (Fig. 1, C and D) We next investigated the accumulation of B1 cells in the milky spots, which are reported to be the entry sites for B1 cells into the peritoneum (14). The decreased number of CFSE-labeled B1 cells accumulated in the omentum milky spots in aged BWF₁ mice compared with young mice (Fig. 2,A). In contrast, B1 cell accumulation in the omentum milky spots was intact in aged BALB/c mice (Fig. 2,B). B1 cell homing to the peritoneal cavity was also normal in aged BALB/c mice. A similar number of B1 cells was detected in the peritoneal cavity and spleen in aged BALB/c mice compared with that in young BWF₁ mice. Furthermore, there was no difference in homing ability to the peritoneal cavity between B1 cells obtained from young BWF₁ mice and those from aged mice. A similar number of B1 cells from either origin was detected in the milky spots and peritoneal cavity when injected together into young BWF₁ mice (Fig. 2 C).

When the mixture of CFSE-labeled B1 cells and CMTMR-labeled B2 cells was injected i.v. into the same aged BWF₁ mice, B1 cells were preferentially recruited in the cellular infiltrates (B220 positive) in the target organs such as the kidney, lung, and thymus, whereas they were similarly recruited to B cell follicles in the spleen (Fig. 3,A). The mean percentages of B1 cells among total number of fluorescent cells were 50.15 ± 3.44, 62.33 ± 10.60 (p < 0.05), 88.81 ± 6.88 (p < 0.001), and 83.00 ± 3.91 (p < 0.001) in the spleen, thymus, kidney, and lung, respectively (Fig. 3 B). These data confirmed that the percentage of B1 cells recruited to the target organs was significantly higher than that of B2 cells in aged BWF₁ mice.

To elucidate the mechanism for defective B1 cell homing, RT-PCR and real-time quantitative PCR analysis on BLC expression in the peritoneum were performed. BLC was expressed in the peritoneal cells in both young and aged BWF₁ mice, although BLC gene expression was significantly lower in aged BWF₁ mice than in young mice (Fig. 4,A, a and b). It contrasted with our previous study demonstrating that BLC gene expression was markedly increased in the target organs including the kidney, lung, and thymus in aged BWF₁ mice (3). RT-PCR analysis on highly purified leukocyte subpopulations demonstrated that BLC was highly expressed in peritoneal CD11b^highCD11c^low cells (Fig. 4,B). Sorted CD11b^highCD11c^low cells showed a morphological appearance of macrophages and expressed F4/80, a cell surface marker for macrophage while they were low in class II expression and negative for B220 (Fig. 5,A). These findings were consistent with previous reports (14). Furthermore, immunocytochemical analysis confirmed BLC protein expression by peritoneal macrophages both in young and aged BWF₁ mice (Fig. 5 B).

Although the number of total peritoneal cells and B1 cells in the peritoneal cavity was significantly increased in aged BWF₁ mice compared with those in young BWF₁ mice (Fig. 6, A and B), the number of peritoneal macrophages in aged BWF₁ mice was markedly reduced compared with those in young BWF₁ mice (Fig. 6 C).

We demonstrated that B1 cells failed to home to the peritoneal cavity in aged BWF₁ mice, whereas a significant number of B1 cells did home in young BWF₁ mice. This was also supported by the fact that markedly decreased number of B1 cells appeared in the milky spots in aged BWF₁ mice when injected i.v. These results are not attributed to mere age-dependent phenomena, because B1 cell homing to the peritoneal cavity was intact in aged BALB/c mice. There was also no difference in homing ability between B1 cells from young mice and those from aged mice. Instead of homing to the peritoneal cavity, B1 cells were recruited to the target organs such as the lung, kidney, and thymus in aged BWF₁ mice. This is probably due to ectopic and high expression of BLC in these organs as demonstrated in our previous study (3) and also due to reduced number of BLC-producing macrophages in the peritoneum in aged BWF₁ mice. In agreement with previous findings (14), the peritoneal macrophage was a major cell source for BLC in the peritoneal cavity both in young and aged BWF₁ mice. We found that the number of macrophages in the peritoneal cavity was markedly decreased in aged BWF₁ mice compared with those in young BWF₁ mice, whereas the number of B1 cells in the peritoneal cavity was increased in aged BWF₁ mice. Increased B1 cells and decreased macrophages in the peritoneal cavity of aged BWF₁ mice are possibly due to enhanced IL-10 expression in the peritoneal cells in aged BWF₁ mice (data not shown), because it is reported that IL-10 produced by B1 cells plays a role as an autocrine growth factor for B1 cells and as a negative regulator for macrophages (15). Enhanced B1 cell generation by increased expression of IL-10 in the peritoneal cavity may overwhelm decreased B1 cell homing to the peritoneal cavity. In contrast, this situation in the peritoneal cavity in aged BWF₁ mice would facilitate B1 cell exit from the peritoneal cavity.

Coinjection of B1 and B2 cells into aged BWF₁ mice resulted in preferential recruitment of B1 cells in the target organs such as the thymus, kidney, and lung, whereas a similar number of B1 and B2 cells was recruited to the spleen. These results are also well explained by our previous findings that B1 cells are more efficiently chemoattracted toward BLC than B2 cells (3). Because BLC expression in high endothelial venules is reported in chronically activated lymphoid tissues such as tonsils and Peyer’s patches (16, 17), BLC expression in high endothelial venules in inflammatory cellular infiltrates is also possibly involved in preferential B1 cell recruitment to the target organs. Chemotaxis assay showed that B1 cells migrated toward stromal cell-derived factor 1 (SDF-1) and secondary lymphoid tissue chemokine (SLC) in addition to BLC, and that only BLC showed preferential chemotactic activity to B1 cells (data not shown). It has been reported recently that SDF-1 expression is enhanced in the glomeruli in the kidney in aged BWF₁ mice (18). However, RT-PCR analysis showed no increase in the level of SDF-1 expression in the target organs between young and aged BWF₁ mice (data not shown). SLC is another chemokine that may possibly affect B1 cell homing to the peritoneal cavity. However, little SLC expression was detected in peritoneal cells both in young and aged BWF₁ mice (data not shown). Constitutive SLC expression was observed in the target organs such as the kidney as well as in lymphoid tissues in both young and aged BWF₁ mice. These results suggest that preferential recruitment of B1 cells in the target organs such as the kidney and lung compared with B2 cells is most likely attributed to high and ectopic expression of BLC in these organs. It is unlikely that migrated cells in the target organs were contaminated macrophages or other cell type, because the purity of B1 cells (CD5⁺B220⁺F4/80⁻) was >95%. It was also confirmed that all CD5⁺B220⁺F4/80⁻ B cells were CD19 positive. We, therefore, favor the idea that ectopic high expression of BLC in the target organs and decreased number of BLC-producing macrophages in the peritoneal cavity contribute to aberrant B1 cell trafficking in aged BWF₁ mice.

Although we do not know the physiological significance of B1 cell homing to the peritoneal cavity at this moment, it is possibly required to maintain the level of IgM natural Abs, because B1 cells differentiate into IgM plasmablasts in the presence of Ag-primed myeloid dendritic cells or peritoneal macrophages (19). Natural IgM Abs, which are secreted mainly by B1 cells, constitute most of circulating IgM and protect systemic bacterial infection (20, 21, 22). Unlike mammalian DNA, bacterial DNA has potent immunostimulatory effects that lead to polyclonal B cell activation as well as the production of specific Abs in mice (23). It is reported that bacterial DNA induces anti-dsDNA Ab cross-reactive to mammalian dsDNA only in autoimmune-prone mice such as in BWF₁ mice (24). IgG anti-DNA Ab are a serologic hallmark for SLE and important mediators for kidney damage (25, 26). Furthermore, accumulating data suggest that B1 cells play an important role in mucosal immunity in the gut (13). Macpherson et al. (12) suggested that IgA produced by B1 cells in the gut contributed to the defense against direct penetration of commensal bacteria into the systemic circulation. Therefore, it is tempting to speculate that aberrant B1 cell trafficking in aged BWF₁ mice may result in penetration of pathogens into systemic circulation and induction of vigorous IgG anti-DNA Ab production that cross-reacts with mammalian DNA.

It has been a matter of debate whether B1 cells are involved in IgG autoantibody production in aged BWF₁ mice because of the lack of firm evidence for IgG class switching in vivo. Recent study has suggested that CXCR5⁺CD4⁺ T cells designated as follicular Th cells enhanced IgG and IgA response in the absence of APCs (16, 27). Preliminary experiments showed that the number of CXCR5⁺CD4⁺ T cells with the phenotype of follicular Th cells was increased in aged BWF₁ mice, and that CD4⁺ T cells obtained from aged BWF₁ mice enhanced IgG Ab production by B1 cells (data not shown). We are now extensively investigating the role of CXCR5⁺CD4⁺ T cells in IgG autoantibody production by B1 cells.

Collectively, we have demonstrated that B1 cells fail to home to the peritoneal cavity and are preferentially recruited to the target organs in aged BWF₁ mice developing murine lupus. These findings would provide a new insight into the pathological significance of B1 cells in the development of SLE and a novel way for the regulation of murine lupus by targeting B1 cell trafficking.

We are very grateful to Dr. Kaoru Hamada for helpful comments.