CD22 × Siglec-G Double-Deficient Mice Have Massively Increased B1 Cell Numbers and Develop Systemic Autoimmunity

Julia Jellusova, Ute Wellmann, Kerstin Amann, Thomas H. Winkler and Lars Nitschke
J Immunol April 1, 2010, 184 (7) 3618-3627; DOI: https://doi.org/10.4049/jimmunol.0902711

Abstract

CD22 and Siglec-G are inhibitory coreceptors for BCR-mediated signaling. Although CD22-deficient mice show increased calcium signaling in their conventional B2 cells and a quite normal B cell maturation, Siglec-G–deficient mice have increased calcium mobilization just in B1 cells and show a large expansion of the B1 cell population. Neither CD22-deficient, nor Siglec-G–deficient mice on a pure C57BL/6 or BALB/c background, respectively, develop autoimmunity. Using Siglec-G × CD22 double-deficient mice, we addressed whether Siglec-G and CD22 have redundant functions. Siglec-G × CD22 double-deficient mice show elevated calcium responses in both B1 cells and B2 cells, increased serum IgM levels and an enlarged population of B1 cells. The enlargement of B1 cell numbers is even higher than in Siglecg−/− mice. This expansion seems to happen at the expense of B2 cells, which are reduced in absolute cell numbers, but show an activated phenotype. Furthermore, Siglec-G × CD22 double-deficient mice show a diminished immune response to both thymus-dependent and thymus-independent type II Ags. In contrast, B cells from Siglec-G × CD22 double-deficient mice exhibit a hyperproliferative response to stimulation with several TLR ligands. Aged Siglec-G × CD22 double-deficient mice spontaneously develop anti-DNA and antinuclear autoantibodies. These resulted in a moderate form of immune complex glomerulonephritis. These results show that Siglec-G and CD22 have partly compensatory functions and together are crucial in maintaining the B cell tolerance.

Bcell receptor transmitted signaling critically influences the development and function of B cells. Engagement of the BCR by Ag binding can result in different outcomes. It may drive the B cell to apoptosis, to anergy, or activate it to trigger Ab responses. Alternations in the strength or quality of the BCR-mediated signal may result in autoimmunity or immunoinsufficiency. To ensure adequate responses, normal B cells possess several coreceptors, which can influence the BCR-induced signal positively (1) or negatively (2). One of such regulating coreceptors is CD22, which is present on the surface of B cells from the pre-B cell stage on and lost only after differentiation to plasma cells. It is expressed on all mature B cell populations including the B2 cells, marginal zone (MZ) B cells and B1 cells. CD22 belongs to the familiy of Siglecs (sialic acid-binding Ig-like lectins) and specifically binds α2,6 linked sialic acids (3). These sugars are present on the surface of many cells, including lymphocytes and cytokine-activated endothelial cells (4, 5). CD22 contains ITIM sequences in its cytoplasmic tail to which SHP-1 is recruited after anti-IgM stimulation (6). The study of CD22-deficient mice has provided some insight into the biological role of this molecule. The main phenotypic feature of CD22-deficient mice is their increased BCR-induced calcium signaling in B2 cells (7–10). Even though defective inhibitory signaling is often associated with the breakdown of tolerance, no signs of autoimmunity were observed in CD22-deficient mice on a pure C57BL/6 background (7). This was in contrast to CD22-deficient mice on a mixed 129 × C57BL/6 background, which developed high-affinity autoantibodies at an older age (11). Overall, CD22-deficient mice showed a normal B cell development with the exceptions of defective homing of recirculating mature B cells to the bone marrow (12) and a decrease in MZ B cells in the spleen (13).

MZ B cells are a special splenic B cell population that inhabit the red pulp/white pulp junction and are characterized by high expression of CD21, CD1d, and low-to-intermediate expression of CD23. Because of their proximity to the marginal sinus, they are among the first B cells to encounter blood borne pathogens in the spleen (14). They were shown to play an important role in thymus independent responses to polysaccharide Ags such as TNP-Ficoll. On binding of these Ags, MZ B cells leave their location, migrate to the red pulp where they become short-lived plasma cells secreting specific Abs (14). Consistent with the decrease in MZ B cells, CD22-deficient mice show a decreased immune response to TNP-Ficoll (7, 9, 13).

Siglec-G is a recently identified new inhibitory coreceptor on B cells. Siglec-G also belongs to the Siglec family and is distantly related to CD22 (6). Siglec-G is expressed in a B cell-restriced pattern, with the highest expression levels on B1 cells. Outside the B cell lineage, Siglec-G is expressed only at low levels (15, 16). When overexpressed in B cell lines, Siglec-G can inhibit BCR-induced Ca2+ signaling (16). This inhibition of signaling is thought to be dependent on SHP-1 binding to Siglec-G, a biochemical mechanism that has so far only been demonstrated for its human ortholog Sigec-10 (17). Siglec-G–deficient mice showed that Siglec-G is a specific inhibitory coreceptor for B1 cells, as Siglec-G–deficient mice had enhanced calcium mobilization in B1 cells, but normal calcium signaling in B2 cells (16). Siglec-G deficiency resulted in highly increased B1 cell numbers, whereas other B cell populations did not seem to be affected. Athough Siglec-G–deficient mice showed fairly normal immune responses, their IgM serum levels which are known to be produced largely by B1 cells, were greatly increased. Also, no emergence of spontaneous high-affinity IgG autoantibodies was observed in Siglec-G–deficient mice (15, 16).

B1 cells represent only a small fraction of the splenic B cells, but are the dominant population of B cells in the pleural and peritoneal cavities. Their localization enables them to be the first to meet pathogens that crossed the gut epithelium. The restricted BCR repertoire of B1 cells favors specificities for glycolipids or carbohydrate motifs that are shared by many pathogens. The produced Igs are mainly of the IgM isotype, are weakly self-reactive and critically contribute to the pool of natural Abs that are present in sera independently of previous infections. The B1 cells in the peritoneal cavity can be further divided into two subpopulations, the B1a cells that express CD5 and B1b cells that are CD5 negative but are otherwise identical in their surface phenotype (IgMhi, CD43+, B220lo). Like MZ B cells, B1 cells contribute to thymus-independent immune responses (18). The expansion of the B1 population in mice is often associated with autoimmunity. Increased numbers of B1 cells are found in several mouse strains [NZB, NZB/W (19)], which are used as models for autoimmune diseases. Also some mutant mouse strains with enhanced B cell signaling, such as SHP-1−/− (20) or Lyn−/− mice (21), show increased B1a cell numbers and signs of autoimmunity.

Neither increased B1 cell numbers plus defective negative signaling in Siglec-G–deficient mice nor hyperresponsive B2 cells of CD22-deficient mice results in autoimmunity. Furthermore, several B cell populations expressing Siglec-G or CD22 do not seem to be greatly affected when one of the molecules is missing. Therefore, a partial redundancy of these two receptors could take place. We have bred Siglec-G × CD22 double-deficient mice to test this possibility. Siglec-G × CD22 double-deficient mice display an enhanced phenotype in comparison with the combined Siglecg−/− and Cd22−/− phenotypes. They show highly increased numbers of B1 cells even more than Siglecg−/− mice, an overall activated B cell phenotype, diminished adaptive responses to TNP-Ficoll and NP-OVA, hyperproliferation in response to TLR ligands and spontaneous production of IgG autoantibodies with glomerulonephritis in aged animals.

Materials and Methods

Mice

To obtain Siglec-G × CD22 double-deficient mice we have crossed Siglecg−/− (BALB/c background) and a Cd22−/− (C57BL/6 background) mice. Several mice with independent crossovers between the linked Cd22 and Siglecg genes were obtained and crossed to obtain double-deficient mice. Control mice, Siglecg−/−, Cd22−/−, and mice carrying three mutant alleles were chosen from littermates or were age-matched to Siglec-G × CD22 double-deficient mice. Experiments were performed in accordance with the German law for protection of animals, after approval by the animal welfare committee.

Cell preparation and flow cytometry

Single-cell suspensions of lymph nodes (LNs) (cervical, inguinal, mesenteric), bone marrow, spleen, and cells obtained from peritoneal lavage were prepared in PBS and 5% FCS. Blood was obtained from tail vein or from penetrating the heart and collected in PBS containing heparin (Roche, Basel, Switzerland). After erythrocyte lysis cells were washed and incubated for 20 min at 4°C with combinations of the following Abs (conjugated to PE, FITC, biotin, allophycocyanin, cy5, or PE-cy5): anti-B220 (eBioscience, San Diego, CA), B7.2 (eBioscience), anti-CD19 (our hybridoma), anti-CD21 (our hybridoma), anti-CD23 (eBioscience), anti-CD43 (BD Biosciences, San Jose, CA), anti-CD5 (eBioscience), anti-CD25 (eBioscience), anti-CD69 (eBioscience), anti–c-kit (eBioscience), anti-IgM (Jackson ImmunoResearch Laboratories, West Grove, PA), and anti-MHCII (BD Pharmingen, San Diego, CA). The stainings were performed in PBS with 0.1% (w/v) BSA, 0.05% (w/v) sodium azide, and saturating concentrations of 2.4G2 (our hybridoma) to block Fc-receptors. Biotinylated Abs were detected using streptavidin Cy5.5 (BD Biosciences). To detect erythrocyte bound IgM, heparinated blood was stained with anti-IgM (Caltag Laboratories, Burlingame, CA). Data were acquired on a four-color flow cytometer (FACSCalibur; BD Biosciences). For analysis, cells were gated to include only living lymphocytes as judged by forward and side scatter parameters. Total cell numbers of living cells were determined by trypan blue staining.

Immunizations and ELISA

NP-OVA (100 μg/mouse, alum precipitated) or TNP-Ficoll (10 μg/mouse in PBS) was i.p. injected into adult age-matched mice. Mice immunized with TNP-Ficoll were bled on d 0, d 5, d 8, and d 13. Levels of TNP specific Abs were determined by ELISA with TNP-BSA (10 μg/ml) coated maxisorb plates (Nunc, Naperville, IL). NP-OVA immunized mice were bled on d 0, d 7, d 14, and d 21. For a secondary immune response, mice were injected again on d 14 with 100 μg NP-OVA. NP-specific Igs were determined by ELISA with NP-OVA (10 μg/ml) coated maxisorb plates (Nunc). A sample of pooled sera was used as standard and included on every plate. Total Ig titers from naive mice were determined by ELISA on polysorb plates (Nunc) coated with isotype-specific Abs (Southern Biotechnology Associates, Birmingham, AL). Monoclonal Ig isotype Abs served as standards (Southern Biotechnology Associates). For the detection of ds-DNA and ss-DNA or RNA specific autoantibodies maxisorb plates were coated with poly-l-lysine (20 μg/ml) and calf-thymus-DNA (20 μg/ml) or bakers yeast RNA (15 μg/ml), respectively, were used. For ssDNA Abs, the coated DNA was denaturated by 20 min incubation at 95°C. To measure the levels of rheumatoid factor, rabbit IgG (10 μg/ml) coated polysorb plates (Nunc) were used. A pool of NZB/W sera served as standard defining arbitrary units, except when determining Abs of the IgG3 or IgG2ab isotypes. In this case serum pools from MRL/lpr or FcR2b−/−YAA+ mice, respectively, were used. For all ELISA essays samples were applied as serial dilutions and their concentration quantified against the standard curve. Isotype-specific Abs coupled to alkaline phosphatase (Southern Biotechnology Associates) or HRP (Jackson ImmunoResearch Laboratories) were used for detection. Biotin labeled Ig haplotype specific Abs (BD Bioscience) and HRP coupled-streptavidin (GE Healthcare, Buckinghamshire, U.K.) were used to detect IgG2aa, IgG2ab Abs.

TLR stimulation and proliferation assays

B1a cells from the peritoneal cavity were isolated by magnetic activated cell sorting using anti-CD90 (Miltenyi Biotec, Auburn, CA) and CD23-PE (eBioscience)/anti-PE (Miltenyi Biotec) Abs coupled to magnetic beads for negative selection, followed by positive selection with anti-CD5 beads. Purity of 90% was achieved, as validated by FACS analysis. Splenic follicular B cells (CD23hi, CD21int, CD5−) were isolated by FACS sorting. Cells (5 × 105/ml) were stimulated with TLR ligands: LPS(Calbiochem, San Diego, CA), R848 (InvivoGen, San Diego, CA), CpG (oligodeoxynucleotide 1826, InvivoGen) for 48 h. Proliferation was measured by incorporation of [3H]thymidine (1 μCi/sample) during the last 8 h of culture.

Calcium mobilization assays

Cells from peritoneal cavity and spleen were loaded with Indo-1 as described (22) (Molecular Probes, Eugene, OR) and stained extracellularly. Baseline fluorescence was measured for 1 min using an LSR II (Becton Dickinson, San Jose, CA), subsequently cells were stimulated by IgM crosslinking (B7.6 clone, our hybridoma) and calcium influx was measured for additional 3 min. The loading efficiency was determined by separate ionomycin stimulation.

Staining of Hep2 slides and histopathology

Antinuclear Abs were detected by incubating a 1:100 serum dilution on Hep2 slides (Euroimmun, Dassow, Germany) according to manufacturer’s instructions. Captured Abs were detected with FITC-labeled anti-IgG (Jackson ImmunoResearch Laboratories).

For histological analysis one kidney was snap-frozen, the other kidney was formalin fixed and sectioned in 2-mm thick slices perpendicular to the longitudinal axis. The slices were then embedded in paraffin and 2-μm thick sections were cut and stained with H&E, periodic acid Schiff for changes of the basement membranes and protein deposition, and Sirius-red for fibrous tissue. Thereafter, renal morphology was investigated by light microscopy (various magnifications) with the investigator (K.A.) being blinded to the animal group. Ten randomly sampled view fields per animal were scored for glomerular (hypercellularity, thickening of glomerular basement membrane, apoptotic cells, matrix expansion), tubulointerstitial (tubular atrophy, interstitial fibrosis, inflammation), and vascular changes (thickening of the vascular wall, vasculitis) using the following scoring system: 0, no changes; +, mild changes; ++. moderate changes; and +++, severe changes (23). Immunohistochemistry for IgG in paraformaldehyde-fixed kidney sections was performed using horse anti-mouse IgG (H+L) peroxidase-labeled Abs and AEC staining (Vector Laboratories, Burlingame, CA).

Software and statistical analysis

Flowcytometric data were analyzed using CellQuest Pro (BD Biosciences) software and FlowJo (TreeStar, Ashland, OR). Statistical analyses were performed using Prism software. The Mann-Whitney U test was used to evaluate the statistical significance of differences in mean Ig levels determined by ELISA, for other data unpaired t test was used. Statistical data are presented as mean ± SD unless stated otherwise.

Results

Expansion of the B1 cell population and preactivated B2 cells in Siglec-G × CD22 double-deficient mice

To investigate whether Siglec-G and CD22 have partly redundant functions, we have crossed Siglec-G–deficient (on BALB/c background) and CD22-deficient mice (on C57BL/6 background) to obtain Siglec-G × CD22 double-deficient mice. The Siglecg and Cd22 genes are linked on chromosome 7 with a 13-Mbp distance. We obtained the crossover between the two loci with the expected frequency. We first analyzed the B cell development in the bone marrow of Siglec-G × CD22 double-deficient mice and found no overt abnormalities in the cell numbers of pro-B cells, pre-B cells, immature, and transitional B cells (Table I). Consistent with earlier reports (7) we saw a reduction of mature recirculating B cells in the bone marrow of Cd22−/− mice and also in the bone marrow of the Siglec-G × CD22 double-deficient mice, but this difference was statistically not significant (p = 0.06). In the peritoneal cavity, we detected a large expansion of the B1a cell population and slightly higher absolute numbers of B1b cells in the Siglec-G × CD22 double-deficient mice and in the Siglec-G–deficient mice consistent with our previous report on the Siglec-G–deficient mice (16) (Fig. 1). The 10-fold increase in B1a cells in the peritoneal cavity fluid from Siglec-G × CD22 double-deficient mice is reflected by an average 6-fold increase in total cell numbers in this compartment. We also found a significant expansion of B1 cells (CD5loB220+) in the spleen, LNs, blood, and the bone marrow of Siglec-G × CD22 double-deficient mice (Fig. 1, Table I). The expansion of B1 cells in the spleen is also seen in Siglecg−/− mice but is significantly higher in Siglec-G × CD22 double-deficient mice. The expansion of B1 cells in the LN and the bone marrow is exclusive for the Siglec-G × CD22 double-deficient mice (Fig. 1, Table I), whereas B1 cells are infrequent in these compartments of normal mice (18). Of note, some of the B1 cells of Siglec-G × CD22 double-deficient mice failed to downregulate B220 expression that is a typical property of B1 cells (Fig. 1). The B1 cell expansion seems to happen at the expense of conventional B cells, because lower conventional B2 cell numbers (CD5− B220+ cells) were found in the spleen, the LNs, and the blood of Siglec-G × CD22 double-deficient mice (Fig. 1, Table I). Siglec-G × CD22 double-deficient B2 cells showed a partially activated phenotype with a higher expression of CD5 and B7.2 (CD86) (Fig. 2B), whereas the levels of the activation markers CD69 and MHC class II were not affected (not shown). Generally, activation markers on B1 cells were not changed between control, single-deficient or double-deficient mice (not shown).

View this table:
Table I. Absolute cell numbers of lymphocyte populations
FIGURE 1.
FIGURE 1.

Expansion of B1a cell population in Siglecg−/− and Siglecg−/−Cd22−/− mice. The top panel shows a representative flow cytometry analysis of CD5 and B220 expression for at least eight independent experiments. Numbers represent percentages of lymphocyte populations B1cells (CD5+B220+), B2 cells (CD5−B220+), T cells (CD5+B220−) in blood, peritoneal cavity, spleen, and LNs. The lower panel shows summarized absolute cell numbers displayed as means + SD. *p < 0.05; **p < 0.005; ***p < 0.001.

FIGURE 2.
FIGURE 2.

Expansion of the MZ B cell population and upregulation of activation markers in Siglecg−/−Cd22−/− mice. A, The top panel shows representative flow cytometry analysis of CD21 and CD23 expression for at least 13 independent experiments. Numbers represent percentages of lymphocyte populations-follicular B cells (CD21loCD23hi) and MZ B cells (CD21hiCD23neg-lo). The lower panel shows summarized absolute cell numbers displayed as means + SD. B, Diagrams of B7.2 expression (top) and CD5 expression (bottom) on B2 cells from different compartments as given as mean fluorescence intensity. Diagrams summarize data from at least 5 and 10 independent experiments, respectively. B2 cells were gated for analysis as shown in Fig. 1. *p < 0.05; **p < 0.005. PC, peritoneal cavity; spl, spleen.

Although CD22-deficiency on a C57BL/6 background results in a reduction of the MZ B cells (13), we could not observe this in the Cd22−/− mice analyzed in this study, which could be due to the mixed BALB/c × C57BL/6 genetic background. In contrast, Siglec-G × CD22 double-deficient mice showed an enlarged MZ B cell population (of ∼2-fold), as identified by CD21hi and CD23lo-neg cells, compared with control mice (Fig. 2A). These results have been also confirmed by staining for the cell surface markers C1d and B220 (not shown). No obvious changes in the relative proportions of transitional T1, T2, or mature B cells were detected.

Increased calcium signaling and serum IgM levels in Siglec-G × CD22 double-deficient mice

Siglec-G acts as inhibitory coreceptor of BCR-induced calcium mobilization in B1a cells (16) and CD22 plays the same role on normal B cells (7–10). To test how the combined deficiency in both negative regulators, Siglec-G and CD22, influences BCR-induced Ca2+ signaling, we performed calcium mobilization assays to study the calcium response in Siglec-G × CD22 double-deficient B cells after anti-IgM stimulation. In both B cell populations investigated, B1a cells from the peritoneal cavity and B2 cells from the spleen, we found an increased calcium response in the Siglec-G × CD22 double-deficient mice compared with the control mice (Fig. 3). This is in contrast to Siglecg−/−mice that only showed increased calcium mobilization in peritoneal B1 cells, whereas Cd22−/− mice showed increased calcium mobilization both in B1 cells, as well as in splenic B2 cells (Fig. 3). Siglec-G × CD22 double-deficient mice also showed elevated calcium mobilization in splenic B1 cells and in the MZ B cell population (not shown).

FIGURE 3.
FIGURE 3.

Increased calcium mobilization in Siglecg−/−Cd22−/− mice. Calcium mobilization in splenic follicular B cells (CD21loCD23hi) and in peritoneal B1a cells (CD5+B220+), which were stimulated with anti-IgM (clone B7.6) at indicated time (arrow). Concentrations of anti-IgM are depicted in histograms. Results are plotted as medians of bound/unbound Indo-1 over time, smoothed by the Gaussian algorithm. One of five experiments with similar results is shown.

B1a cells are thought to be the most prominent producers of natural Abs that are mainly of the IgM isotype (18). Therefore, increased serum IgM levels are often found in mice with increased B1a cell populations. As expected, Siglecg−/− and Siglec-G × CD22 double-deficient mice showed 5- to 8-fold increased serum IgM levels (to 430 μg/ml and 640 μg/ml on average, respectively) (Fig. 4A). With age the IgM production increased even further and by 24-wk Siglec-G × CD22 double-deficient mice expressed significantly higher IgM levels than Siglecg−/− mice (data not shown). Other Ig isotypes were produced at comparable levels with the exception of IgA, which was found in lower concentrations in the sera from Siglec-G × CD22 double-deficient mice compared with the control mice (Fig. 4A). This finding was consistent with higher numbers of IgM-producing cells and lower numbers of IgA-producing cells in the bone marrow and the spleen, as detected by ELISPOT (data not shown).

FIGURE 4.
FIGURE 4.

Higher IgM serum levels, impaired immune responses of Siglecg−/−Cd22−/− mice and increased TLR-induced proliferation of their B cells. A, Serum Ig concentrations determined by ELISA. Each symbol represents a single mouse. B, Immune response to TNP-Ficoll immunization. TNP specific Ig levels were determined by ELISA C, Immune response to NP-OVA immunization. Mice were immunized with NP-OVA at d 0 and d 14. NP-specific IgG1 and IgG2b responses were determined by ELISA. D, Proliferation of B cells after stimulation with TLR ligands Purified B1a cells (CD23−,CD5+) from the peritoneal cavity and follicular B cells (CD23+,CD21lo,CD5−) from the spleen were in vitro stimulated with LPS (TLR4 ligand), R848 (TLR7 ligand), CpG (TLR9 ligand) at indicated concentrations. Proliferation was measured by [3H]thymidine uptake, data are presented as counts per minute (cpm). In the experiments A, B, and C, Cd22−/− mice were either Siglecg+/+ or Siglecg+/− (the majority was +/−). Siglecg−/− mice were either Cd22+/+ or Cd22+/− (the majority was +/−). In the experiments D, only Siglecg−/− Cd22+/+ and Siglecg+/+Cd22−/− mice were used. *p < 0.05; **p < 0.005; ***p < 0.001.

Siglec-G × CD22 double-deficient mice show impaired thymus-independent and thymus-dependent type II immune responses, but increased proliferation in response to TLR ligands

To determine the ability of Siglec-G × CD22 double-deficient mice to respond to thymus-independent type II or thymus-dependent Ags, groups of adult age-matched mice were immunized with TNP-Ficoll or NP-OVA. Prior to immunization all three groups of mutant mice displayed elevated amounts of TNP-binding IgM. The specific IgM levels in Siglec-G × CD22 double-deficient mice prior to immunization were comparable to those in control mice 5 d after immunization with TNP-Ficoll (Fig. 4B). Five and 8 d after TNP-Ficoll immunization the levels of TNP-specific Abs increased slightly in the mutant mice but the magnitude of the increase was substantially lower than in the control mice. The maximal levels of TNP-specific IgM reached by Siglec-G × CD22 double-deficient mice were also significantly lower than in the Siglecg−/− or the Cd22−/− mice (Fig. 4B). The IgG3 response to TNP-Ficoll challenge was severely impaired in Siglec-G × CD22 double-deficient mice (Fig. 4B), indicating that Siglec-G × CD22 double-deficient mice are not able to respond correctly to this Ag. Siglecg−/− or Cd22−/− mice showed a markedly higher IgG3 response to TNP-Ficoll immunization than Siglec-G × CD22 double-deficient mice, but the response was slower when compared with the control mice.

The primary response of Siglec-G × CD22 double-deficient mice to the thymus-dependent Ag NP-OVA was delayed and weaker than that of the other three groups of mice (Fig. 4C). Compared with the response in control mice, Siglecg−/− or Cd22−/− mice also showed a slightly lower primary response. The same result of a weaker primary response of Siglec-G × CD22 double-deficient mice after NP-OVA immunization was also found for other NP-specific Igs, namely, IgM, IgG3, IgA (data not shown). In contrast to the primary response, all three types of mutant mice showed normal secondary responses to NP-OVA (Fig. 4C). Because the Siglecg−/− and Cd22−/− mice used for TNP-ficoll or NP-OVA immunizations were mainly of the Siglecg−/− Cd22+/− and Cd22−/− Siglecg+/− genotypes, respectively, we cannot exclude an influence of the third mutant allele on the immune responses.

To determine whether Siglec-G and CD22 play a role in the regulation of TLR-mediated B cell responses, peritoneal B1a cells and follicular B cells were isolated and stimulated in vitro with LPS (TLR4 ligand), R848 (TLR7 ligand), or CpG (TLR9 ligand) at several different concentrations. Both B cell populations of the three mutant mouse lines showed a higher proliferative response to all three TLR ligands, compared with the corresponding wild-type B cell population, as shown by higher [3H]thymidine incorporation (Fig. 4D). Similar results were obtained when proliferation was measured by CFSE dilution (not shown).

Aged Siglec-G × CD22 double-deficient mice spontaneously develop autoantibodies and signs of glomerulonephritis

Several mouse strains that display signs of systemic autoimmunity show higher numbers of B1a cells and a phenotype of preactivated B cells (19, 24). Therefore, we measured autoantibodies in 24-, 36- or 48-wk-old mice. We first measured the relative concentrations of different autoantibodies of the IgG isotype. Already at the age of ∼24 wk, 50% of tested Siglec-G × CD22 double-deficient mice showed considerable IgG anti-dsDNA titers and 40% mice were positive for anti-ssDNA Abs of the IgG isotype, whereas hardly any positive mice were found in the control or single-deficient mice (Fig. 5A, 5B). At later time points, some positive mice with anti-dsDNA or -ssDNA Abs were also found in the two groups of either Siglecg−/− mice, or in mice with three mutant alleles, namely, Siglecg−/−Cd22+/− or Siglecg+/− Cd22−/− (Fig. 5A, 5B). This suggests that the gene product of only one Siglecg or Cd22 allele might not be enough to secure tolerance in B cells.

FIGURE 5.
FIGURE 5.

Increased levels of IgG autoantibodies in Siglecg−/−Cd22−/− mice. Anti-dsDNA autoantobodies (A) and anti-ssDNA autoantibodes (B) and anti-RNA autoantibodies (in C) were determined by ELISA (upper panels). Each symbol represents an individual mouse. Diagrams showing the penetrance in the lower panels summarize percentage of positive mice. Sera containing at least 10% autoantibodies of the NZB/W serum mix used as standard were considered positive. Groups of 24, 36, and 48 wk old mice did not necessary contain the same animals. The group of control mice included Siglecg+/+Cd22+/+ and Siglecg+/+Cd22+/− mice. The group of Cd22−/− mice (CD22) included only Siglecg+/+Cd22−/− and the group of Siglecg−/− mice (SigG) included only Siglecg−/−Cd22+/+ mice. The group of animals with three mutated alleles (3Mut), included Siglecg−/−Cd22+/− (black squares) and Siglecg+/−Cd22−/− mice (gray squares). The group DKO are Siglecg−/−Cd22−/− mice. *p < 0.05; **p < 0.005; ***p < 0.001.

Also, RNA binding Abs could be detected in the sera from Siglec-G × CD22 double-deficient mice, and with lower frequency in the other four groups of mice. From the age of 36 wk, the levels of RNA binding Abs were significantly higher in Siglec-G × CD22 double-deficient mice, compared with the control mice (Fig. 5C). Because increased production of autoantibodies of the IgG2a isotype is considered to accelerate the progression of autoimmune pathologies (23, 25), we analyzed the isotype distribution of dsDNA and RNA binding Abs by measuring Ag-specific IgG1, IgG2a, and IgG3. Although anti-dsDNA or anti-RNA IgG2a Abs were never detected in control mice, single-deficient mice, mice with three mutant alleles or Siglec-G × CD22 double-deficient mice all showed a roughly similar distribution of IgG1 and IgG2a autoantibodies (Supplemental Fig. 1). The majority of these mice produced IgG1 as well as IgG2a autoantibodies. IgG3-positive sera were found exclusively in the group of Siglec-G × CD22 double-deficient mice.

The presence of antinuclear IgG autoantibodies was tested by staining slides of Hep-2 cells with serum from 24- to 48-wk-old mice. Weakly positive sera were found in all groups of mutant mice. However, sera that showed a bright staining of the nucleus were predominantly from Siglec-G × CD22 double-deficient mice and at later time points also from mice with three mutant alleles (Fig. 6A), but generally hardly ever from other groups of mice.

FIGURE 6.
FIGURE 6.

Increased levels of antinuclear IgG Abs and similar levels of IgM autoantibodies. A, Antinuclear Abs were determined by incubating Hep2 slides with serum diluted 1:100 and captured Abs detected by fluorescence-labeled anti-IgG. B. IgM bound to erythrocytes in aged mice (over 1 y) was measured by FACS. Values obtained from 7 mo old NZB mice are displayed as a comparison. C, Rheumatoid factor IgM measured by ELISA (top). Groups of 24, 36, and 48 wk old mice did not necessary contain the same animals. The group of control mice included Siglecg+/+Cd22+/+ and Siglecg+/+Cd22+/− mice. The group of Cd22−/− mice is shown as CD22, Siglecg−/− mice are shown as SigG. Siglecg−/−Cd22−/− double-ko mice are shown as DKO. The lower panel shows the ratio of total IgM in serum to rheumatoid factor. *p < 0.05; **p < 0.005; ***p < 0.001.

To test for autoantibodies of the IgM isotype, antierythrocyte and rheumatoid factor IgM levels were determined. Erythrocyte-bound IgM levels were slightly but significantly elevated in Siglecg−/− and Siglec-G × CD22 double-deficient mice (Fig. 6B). Erythrocyte-bound Abs that can cause autohemolytic anemia are elevated in NZB mice to higher levels than that found in Siglec-G mutant mice (Fig. 6B). In addition, Siglecg−/− and Siglec-G × CD22 double-deficient mice show higher titers of rheumatoid factor of the IgM isotype at all three time points measured (24, 36, and 48 wk). This probably reflects the elevated serum IgM titers and IgM polyreactivity in these mice, because the ratio of total IgM to rheumatoid factor IgM is comparable within all four groups of mice (Fig. 6C).

Renal pathology revealed moderate glomerular, but no tubulointerstitial damage in Siglec-G × CD22 double-deficient mice and in mice with three mutant alleles, as well as in lower penetrance in Cd22−/− mice, whereas there were no such changes in wild-type littermates (Fig. 7). Signs of vasculitis were not detected at all. Immunohistochemistry of kidney sections using an IgG Ab revealed marked membranous and subendothelial depositions of IgG and occasional intracapillary thrombi within the glomeruli of Siglec-G × CD22 double-deficient mice compared with wild-type littermates (Fig. 7). Despite this glomerular damage, proteinuria was only occasionally observed in Siglec-G × CD22 double-deficient mice and no general shorter survival of the animals was detected until the mice were sacrificed for pathological examination.

FIGURE 7.
FIGURE 7.

Glomerular damage and IgG immune-complexes in kidneys of CD22 × Siglec-G double-deficient mice. A, Example of kidney sections of control mice and CD22 × Siglec-G double-deficient mice, shown in two different magnifications (×20 and ×40) with periodic acid Schiff or anti-IgG staining. B, Quantification of kidney pathology analysis. Mice, which showed moderate glomerular damage, are summarized here. The group of control mice, included Siglecg+/+Cd22+/+ and Siglecg+/+Cd22+/− mice. 3 Mut: mice with three mutant alleles (Siglecg−/− Cd22+/− or Siglecg+/− Cd22−/−).

Discussion

Siglec-G and CD22 are known to play important roles on B1a cells and B2 cells, respectively. In this study we showed that Siglec-G and CD22 also have partly redundant functions because Siglec-G × CD22 double-deficient mice exhibit phenotypic features neither seen in Siglec-G nor in CD22 single-deficient mice. The population of B1 cells is strongly expanded in Siglec-G × CD22 double-deficient mice, leading for example to 10-fold higher B1a cell numbers in the peritoneal cavity. The B1 cell expansion was clearly more profound than in Siglec-G single-deficient mice and was especially seen in the spleen, the LNs, and in the bone marrow. Furthermore, the population of MZ B cells is expanded in Siglec-G × CD22 double-deficient mice, whereas normal B2 lymphocytes are decreased in absolute numbers. B cells of Siglec-G × CD22 double-deficient mice show an activated phenotype and all types of Siglec-G × CD22 double-deficient B cells respond with elevated BCR-induced calcium responses. Corresponding to increased B1 cell numbers, Siglec-G × CD22 double-deficient mice have increased IgM serum levels, whereas other Ig isotypes are relatively normal. Both thymus-independent type II, as well as primary thymus-dependent immune responses are defective. Finally, aged Siglec-G × CD22 double-deficient show spontaneous development of anti-DNA and antinuclear autoantibodies.

The origin and development of B1a cells is still controversial (18). Fetal liver seems to be better suited to produce B1a cells than adult BM, because fetal-liver stem cells readily give rise to B1a cells when injected into irradiated mice, whereas cells from adult BM preferentially, but not exclusively reconstitute the compartment of B2 cells (26). Also, a B1 cell-specified progenitor has been identified (27), which adds support to the theory of two distinct lineages for B1 and B2 cells. On the other hand, an important factor, which can modulate the frequency of generated B1a cells, seems to be the specificity of the BCR and the BCR-mediated signaling. There is considerable evidence that B cells are selected into the compartment of B1a cells by their Ag specificity as several transgenic mice with B cells expressing Ig genes typical for B1a cells result in overproduction of B1a cells (28–30). Because B1a cells often produce Igs with a low affinity for self-Ags, the binding of Ag might be necessary for their positive selection. This was shown for Thy-1 specific B1a cells, which are not selected into the pool of mature B1a cells in Thy-1 deficient mice (31). Also, the ablation of different inhibitory molecules like Siglec-G (15, 16) or SHP-1 (20) increases the number of B1a cells supporting the role of the BCR signal in expansion or survival of this population. In vitro experiments showed, that the engagement of the BCR by anti-IgM combined with the addition of IL-6 leads to upregulation of CD5 and downregulation of CD23 resembling the B1a cell phenotype (32).

Therefore, it might be possible that B cell precursors otherwise destined to become B2 cells are differentially selected into the pool of B1 cells in Siglec-G × CD22 double-deficient mice due to enhanced signaling in these cells. This would result in an altered BCR repertoire of B1 cells in double-deficient mice. In support of this, we obtained evidence that the BCR repertoire of Siglecg−/− B1 cells is changed in comparison with control B1 cells (unpublished results). In addition, the deficiency of the two inhibitory receptors Siglec-G and CD22 could enhance the survival and/or expansion of B1a cells, perhaps on the expense of B2 cells.

In addition to increased numbers of B1 cells in the spleen of Siglec-G × CD22 double-deficient mice, there was a 2-fold higher absolute number of MZ B cells, whereas the population of follicular B cells was diminished. The phenotype of increased MZ B cells is the opposite phenotype to CD22-deficient mice, which have a smaller MZ B cell population, at least on pure C57BL/6 background (13). The double-knockout (KO) phenotype can therefore not easily be explained by just a change in the BCR signaling strength. Nevertheless, it is notable that both innate B cell populations, B1 cells and MZ B cells are expanded in absolute numbers in Siglec-G × CD22 double-deficient mice. Again, a preferred selection into the repertoire of MZ B cells or a preferential survival maybe the cause for this relative shift in splenic populations. MZ B cells seem to be a critical population of B cells in the spleen that are generally less affected than follicular B cells in cases when there is a compromised B cell population such as in IgH or L chain transgenic mice (14). The mechanisms responsible for the reduction of follicular B cells and increase of MZ B cells in Siglec-G × CD22 double-deficient mice needs to be studied in future experiments.

The absence of CD22 and Siglec-G results in preactivation of B2 cells, which show an upregulation of CD86 (B7.2) and CD5. CD86 is a classical activation marker on B cells, which is upregulated in response to BCR crosslinking. Also CD5 can be upregulated on strong crosslinking of the BCR on normal B cells (32). CD5 upregulation has also been described in the literature as being typical for anergic cells (33). However, the expression of CD5 does not seem to be a hallmark of anergy in this case, because Siglec-G × CD22 double-deficient mice show increased calcium signaling after anti-IgM stimulation in all populations investigated and develop autoimmunity, which is the opposite of an anergic phenotype. In addition, Siglec-G × CD22 double-deficient B2 cells had decreased levels of CD23 (not shown) that is also typical for activated B cells. This phenotype points to a preactivation of double-deficient B cells in vivo, probably resulting from the loss of two inhibitory coreceptors. The increased calcium signaling after BCR crosslinking in all B cell populations seems to be a combined phenotype of the Siglec-G–deficiency (affecting B1 cell calcium signaling) and the CD22-deficiency (affecting mainly B2 cell calcium signaling). In this study the anti-IgM–induced calcium response is also clearly higher in Cd22−/− B1 cells, in contrast to previous results (34). This may be due to the mixed genetic background of the mice.

Siglec-G × CD22 double-deficient mice had clearly elevated levels of serum IgM. The serum levels were even higher than in Siglec-G–deficient mice. The increased IgM levels correlate with the increase of B1 cells that are largely responsible for the natural Abs of the IgM isotype (18). Siglec-G × CD22 double-deficient mice showed a diminished Ig response to TNP-Ficoll or NP-OVA immunization. All three mutant mice showed elevated levels of NP- or TNP-specific IgM prior to immunization and the levels increased only slightly after challenge with TNP-Ficoll or NP-OVA. The IgG response of Siglec-G × CD22 double-deficient mice to both thymus-dependent and thymus-independent type II Ag was delayed and weaker compared with the other groups of mice. The reduced ability of Siglec-G × CD22 double-deficient mice to respond to thymus-independent type II and thymus-dependent Ag might be a result of an unnaturally strong BCR signal. Calcium signaling is known to be involved in B cell activation, but also plays a role in the signal transduction leading to apoptosis (35), therefore strong engagement of the BCR by Ag not regulated by Siglec-G nor by CD22 might result in deletion rather than in activation. Alternatively, the high preimmune TNP- or NP-specific IgM may capture the Ag and prevent efficient binding to the Ag receptor. Also Siglecg−/− and Cd22−/− mice showed slightly delayed IgG responses to NP-OVA stimulation compared with the control group, whereas a roughly normal immune response was reported for the original Siglec-G– and CD22-deficient mice (7, 9, 10, 16). The different ability of Siglecg−/− and Cd22−/− mice to respond to this Ag in our experiments might be explained by the fact that the majority of animals used were deficient in three alleles—they were Siglecg−/−Cd22+/− or Siglecg+/− Cd22−/−.

In contrast to the impaired immune response to immunization with thymus-dependent or thymus-independent type II Ags, B cells of all three mutant mice showed an increased in vitro response to TLR ligands (TLR4, TLR7, and TLR9). A higher proliferative response to LPS was already reported for the Siglec-G and CD22 single-deficient mice (7, 16) but here we show for the first time that enhanced proliferation is also seen in response to TLR 7 and TLR9 ligands. Although we cannot exclude that this hyperproliferative response is due to altered TLR expression in the mutant mice, it might be possible that Siglec-G and CD22 are involved in the regulation of TLR-induced signals. This would be interesting to investigate in future studies.

Increased BCR-mediated signaling due to a loss of negative regulators is often associated with autoimmunity. Although Cd22-deficiency is a risk factor for the development of autoimmunity in the mouse (6, 36) it does not seem to be sufficient to cause autoimmunity without any additional modifiers. Autoantibody generation was observed in Cd22−/− mice on a C57BL/6 × 129 background (11), but not on a C57BL/6 background (7). Also, on a genetic background of autoimmune-prone mouse strains Cd22 defects can contribute to the severity of the disease. A polymorphism in the Cd22 gene has been found with one allele, the Cd22a, encoding for a defective CD22 protein, being present in several autoimmune-prone mouse strains (37). Siglec-G deficiency on a BALB/c background results in slightly elevated titers of antierythrocyte IgM and rheumatoid factor, but no high affinity IgG autoantibodies were detected (16). This was an interesting finding, because also autoimmune strains such as NZB or NZB/W show a large B1 cell expansion and B1 cells were reported to be essential for autoantibody production (38).

In this report we showed, that the absence of two negative regulators of the BCR signal, Siglec-G and CD22, results in spontaneous development of isotype switched autoantibodies in aged mice. The appearance of anti-dsDNA autoantibodies and anti-ssDNA autoantibodies was first noticed at the age of ∼24 wk. After 48 wk also some animals in the group of Siglecg−/− mice or mice with three mutant alleles were tested positive for autoantibodies. Although RNA binding Abs were more common in all groups of mice, there was a significant increase in anti-RNA Ab titers in 36 wk and 48 wk old Siglec-G × CD22 double-deficient mice. Furthermore, ANA IgG autoantibodies were detected from the age of 36 wk in both the Siglecg−/− × Cd22−/− mice and mice with three mutant alleles. Also, both aged Siglec-G × CD22 double-deficient mice, as well as aged mice with three mutant alleles developed an Ig-immune-complex dependent glomerulonephritis with moderate severity. This shows that the negative regulation by both CD22 and Siglec-G is critical for the maintenance of B cell tolerance. It also suggests that there is a gene dosage effect with 3 mutant alleles being sufficient to break tolerance.

The increased TLR responses in Siglec-G × CD22 double-deficient mice could potentially contribute to the autoimmune phenotype. Several studies propose the model that DNA and DNA-associated autoantigens or RNA and RNA-associated autoantigens can stimulate autoreactive B cells by dual BCR and TLR signaling (39, 40). Also, gene duplication of TLR7 was shown to promote autoimmunity (41, 42) and therefore it will be interesting to study by genetic models whether the autoimmune manifestations seen in Siglec-G × CD22 double-deficient mice are dependent on TLR signals.

SLE-like systemic autoimmune diseases in mice on a mixed C57BL/6 × 129 background have been observed rather frequently. Chromosome intervals from the 129 mouse that contribute to systemic autoimmunity in mixed C57BL/6 × 129 background have been identified, and one of these intervals, called Sle3 lies on chromosome 7 and overlaps with the Siglec gene cluster (43, 44). It is therefore conceivable that autoimmune phenotypes are caused by interacting loci between 129 and C57BL/6 mice, without involvement of the targeted genes. We think we can exclude this phenomenon in our study for the following reasons: Siglec-G × CD22 double-deficient mice were generated in BALB/c or C57BL/6 ES cells, respectively, that is, they do not contain any 129 genetic background. Also, the mice analyzed here show a clear gene-dosage effect, with mice carrying 4 mutant alleles (Cd22−/− Siglecg−/−) being affected with higher penetrance than mice with 3 mutant alleles, and those affected higher than mice with two mutant alleles. Because all mice were derived from littermates or from the same generation of similar matings, a phenotype not related to the mutated Cd22 and Siglecg genes is unlikely. Furthermore, the kallikrein gene cluster within the Sle3 locus has recently been shown to be responsible for functional downmodulation of Ab-induced glomerulonephritis in BALB/c and C57BL/6 mice, but not in 129 mice (45). This kallikrein gene cluster is very close to the Siglec gene cluster on chromosome 7. This finding suggests that genetic changes in the protective kallikrein gene cluster are responsible for the lupus-susceptible Sle3 locus in 129 mice, at least in the Ab-induced nephritis model.

B cell abnormalities observed in Siglec-G × CD22 double-deficient mice resemble those seen in mice with a B cell-specific deletion of SHP-1 (20). These mice also show an enlargement of the B1a cell population in several organs, increased serum IgM levels, hyper-responsiveness of B1a cells to anti-IgM stimulation, diminished immune responses, spontaneous production of autoantibodies and systemic autoimmune disease. SHP-1 binds to several Siglecs containing ITIM sequences, including CD22 and human Siglec-10 (17), the human ortholog to Siglec-G, and is essential for the inhibitory function of CD22 (46) and of many other Siglecs (3). The phenotype of SHP-1 deficient mice could thus to a great part be explained by the loss of the inhibitory functions of Siglec-G and CD22. Also the Scr-like kinase Lyn, which phosphorylates CD22 ITIM motifs (47, 48) and probably also other ITIM-carrying Siglecs, is involved in this inhibitory pathway, as Lyn-deficient mice develop spontaneous autoimmunity (21). In parallel with the current study, heterozygosity for each Cd22, Lyn, and Shp1 genes contributed in a quantitative trait to B cell hyper-responsiveness (49). Also, the CD22-Lyn-SHP-1 pathway was recently shown to control B cell signaling and B cell tolerance, as B cells mature (50). In summary, Cd22 and Siglecg prevent spontaneous IgG autoantibody production by controlling the BCR signaling threshold with partly redundant functions in B lymphocytes.

Acknowledgments

We acknowledge the expert technical help of K. Ebert, A. Urbat, the laboratory of Dr. A. Gessner, S. Söllner, and M. Reutelshöfer.

Disclosures The authors have no financial conflicts of interest.

Footnotes

  • This work was supported by the Deutsche Forschungsgemeinschaft (FOR 832, SFB 643, SFB 423) and the Interdisciplinary Center for Clinical Research Erlangen (A31).

  • The online version of this article contains supplemental material.

  • Abbreviations used in this paper:

    BM
    bone marrow
    KO
    knockout
    LN
    lymph node
    MZ
    marginal zone
    PC
    peritoneal cavity
    SLE
    systemic lupus erythematosus
    spl
    spleen.

  • Received August 18, 2009.
  • Accepted January 29, 2010.

References

  1. ↵
    1. Tedder T. F.,
    2. M. Inaoki,
    3. S. Sato
    . 1997. The CD19-CD21 complex regulates signal transduction thresholds governing humoral immunity and autoimmunity. Immunity 6: 107–118.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Nitschke L.
    2005. The role of CD22 and other inhibitory co-receptors in B-cell activation. Curr. Opin. Immunol. 17: 290–297.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Crocker P. R.,
    2. J. C. Paulson,
    3. A. Varki
    . 2007. Siglecs and their roles in the immune system. Nat. Rev. Immunol. 7: 255–266.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Engel P.,
    2. Y. Nojima,
    3. D. Rothstein,
    4. L. J. Zhou,
    5. G. L. Wilson,
    6. J. H. Kehrl,
    7. T. F. Tedder
    . 1993. The same epitope on CD22 of B lymphocytes mediates the adhesion of erythrocytes, T and B lymphocytes, neutrophils, and monocytes. J. Immunol. 150: 4719–4732.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Hanasaki K.,
    2. A. Varki,
    3. I. Stamenkovic,
    4. M. P. Bevilacqua
    . 1994. Cytokine-induced beta-galactoside alpha-2,6-sialyltransferase in human endothelial cells mediates alpha 2,6-sialylation of adhesion molecules and CD22 ligands. J. Biol. Chem. 269: 10637–10643.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Nitschke L.
    2009. CD22 and Siglec-G: B-cell inhibitory receptors with distinct functions. Immunol. Rev. 230: 120–148.
    OpenUrl
  7. ↵
    1. Nitschke L.,
    2. R. Carsetti,
    3. B. Ocker,
    4. G. Köhler,
    5. M. C. Lamers
    . 1997. CD22 is a negative regulator of B-cell receptor signalling. Curr. Biol. 7: 133–143.
    OpenUrlCrossRefPubMed
    1. O’Keefe T. L.,
    2. G. T. Williams,
    3. S. L. Davies,
    4. M. S. Neuberger
    . 1996. Hyperresponsive B cells in CD22-deficient mice. Science 274: 798–801.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Otipoby K. L.,
    2. K. B. Andersson,
    3. K. E. Draves,
    4. S. J. Klaus,
    5. A. G. Farr,
    6. J. D. Kerner,
    7. R. M. Perlmutter,
    8. C. L. Law,
    9. E. A. Clark
    . 1996. CD22 regulates thymus-independent responses and the lifespan of B cells. Nature 384: 634–637.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Sato S.,
    2. A. S. Miller,
    3. M. Inaoki,
    4. C. B. Bock,
    5. P. J. Jansen,
    6. M. L. Tang,
    7. T. F. Tedder
    . 1996. CD22 is both a positive and negative regulator of B lymphocyte antigen receptor signal transduction: altered signaling in CD22-deficient mice. Immunity 5: 551–562.
    OpenUrlCrossRefPubMed
  11. ↵
    1. O’Keefe T. L.,
    2. G. T. Williams,
    3. F. D. Batista,
    4. M. S. Neuberger
    . 1999. Deficiency in CD22, a B cell-specific inhibitory receptor, is sufficient to predispose to development of high affinity autoantibodies. J. Exp. Med. 189: 1307–1313.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Nitschke L.,
    2. H. Floyd,
    3. D. J. Ferguson,
    4. P. R. Crocker
    . 1999. Identification of CD22 ligands on bone marrow sinusoidal endothelium implicated in CD22-dependent homing of recirculating B cells. J. Exp. Med. 189: 1513–1518.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Samardzic T.,
    2. D. Marinkovic,
    3. C. P. Danzer,
    4. J. Gerlach,
    5. L. Nitschke,
    6. T. Wirth
    . 2002. Reduction of marginal zone B cells in CD22-deficient mice. Eur. J. Immunol. 32: 561–567.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Martin F.,
    2. J. F. Kearney
    . 2002. Marginal-zone B cells. Nat. Rev. Immunol. 2: 323–335.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Ding C.,
    2. Y. Liu,
    3. Y. Wang,
    4. B. K. Park,
    5. C. Y. Wang,
    6. P. Zheng,
    7. Y. Liu
    . 2007. Siglecg limits the size of B1a B cell lineage by down-regulating NFkappaB activation. PLoS One 2: e997.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Hoffmann A.,
    2. S. Kerr,
    3. J. Jellusova,
    4. J. Zhang,
    5. F. Weisel,
    6. U. Wellmann,
    7. T. H. Winkler,
    8. B. Kneitz,
    9. P. R. Crocker,
    10. L. Nitschke
    . 2007. Siglec-G is a B1 cell-inhibitory receptor that controls expansion and calcium signaling of the B1 cell population. Nat. Immunol. 8: 695–704.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Whitney G.,
    2. S. Wang,
    3. H. Chang,
    4. K. Y. Cheng,
    5. P. Lu,
    6. X. D. Zhou,
    7. W. P. Yang,
    8. M. McKinnon,
    9. M. Longphre
    . 2001. A new siglec family member, siglec-10, is expressed in cells of the immune system and has signaling properties similar to CD33. Eur. J. Biochem. 268: 6083–6096.
    OpenUrlPubMed
  18. ↵
    1. Berland R.,
    2. H. H. Wortis
    . 2002. Origins and functions of B-1 cells with notes on the role of CD5. Annu. Rev. Immunol. 20: 253–300.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Hayakawa K.,
    2. R. R. Hardy,
    3. D. R. Parks,
    4. L. A. Herzenberg
    . 1983. The “Ly-1 B” cell subpopulation in normal immunodefective, and autoimmune mice. J. Exp. Med. 157: 202–218.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Pao L. I.,
    2. K. P. Lam,
    3. J. M. Henderson,
    4. J. L. Kutok,
    5. M. Alimzhanov,
    6. L. Nitschke,
    7. M. L. Thomas,
    8. B. G. Neel,
    9. K. Rajewsky
    . 2007. B cell-specific deletion of protein-tyrosine phosphatase Shp1 promotes B-1a cell development and causes systemic autoimmunity. Immunity 27: 35–48.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Hibbs M. L.,
    2. D. M. Tarlinton,
    3. J. Armes,
    4. D. Grail,
    5. G. Hodgson,
    6. R. Maglitto,
    7. S. A. Stacker,
    8. A. R. Dunn
    . 1995. Multiple defects in the immune system of Lyn-deficient mice, culminating in autoimmune disease. Cell 83: 301–311.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Stork B.,
    2. M. Engelke,
    3. J. Frey,
    4. V. Horejsí,
    5. A. Hamm-Baarke,
    6. B. Schraven,
    7. T. Kurosaki,
    8. J. Wienands
    . 2004. Grb2 and the non-T cell activation linker NTAL constitute a Ca(2+)-regulating signal circuit in B lymphocytes. Immunity 21: 681–691.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Wellmann U.,
    2. M. Letz,
    3. A. Schneider,
    4. K. Amann,
    5. T. H. Winkler
    . 2001. An Ig mu-heavy chain transgene inhibits systemic lupus erythematosus immunopathology in autoimmune (NZB x NZW)F1 mice. Int. Immunol. 13: 1461–1469.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Wither J. E.,
    2. V. Roy,
    3. L. A. Brennan
    . 2000. Activated B cells express increased levels of costimulatory molecules in young autoimmune NZB and (NZB x NZW)F(1) mice. Clin. Immunol. 94: 51–63.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Takahashi S.,
    2. L. Fossati,
    3. M. Iwamoto,
    4. R. Merino,
    5. R. Motta,
    6. T. Kobayakawa,
    7. S. Izui
    . 1996. Imbalance towards Th1 predominance is associated with acceleration of lupus-like autoimmune syndrome in MRL mice. J. Clin. Invest. 97: 1597–1604.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Herzenberg L. A.
    2000. B-1 cells: the lineage question revisited. Immunol. Rev. 175: 9–22.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Montecino-Rodriguez E.,
    2. H. Leathers,
    3. K. Dorshkind
    . 2006. Identification of a B-1 B cell-specified progenitor. Nat. Immunol. 7: 293–301.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Chumley M. J.,
    2. J. M. Dal Porto,
    3. S. Kawaguchi,
    4. J. C. Cambier,
    5. D. Nemazee,
    6. R. R. Hardy
    . 2000. A VH11V kappa 9 B cell antigen receptor drives generation of CD5+ B cells both in vivo and in vitro. J. Immunol. 164: 4586–4593.
    OpenUrlAbstract/FREE Full Text
    1. Clarke S. H.,
    2. L. W. Arnold
    . 1998. B-1 cell development: evidence for an uncommitted immunoglobulin (Ig)M+ B cell precursor in B-1 cell differentiation. J. Exp. Med. 187: 1325–1334.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Lam K. P.,
    2. K. Rajewsky
    . 1999. B cell antigen receptor specificity and surface density together determine B-1 versus B-2 cell development. J. Exp. Med. 190: 471–477.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Hayakawa K.,
    2. M. Asano,
    3. S. A. Shinton,
    4. M. Gui,
    5. D. Allman,
    6. C. L. Stewart,
    7. J. Silver,
    8. R. R. Hardy
    . 1999. Positive selection of natural autoreactive B cells. Science 285: 113–116.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Cong Y. Z.,
    2. E. Rabin,
    3. H. H. Wortis
    . 1991. Treatment of murine CD5- B cells with anti-Ig, but not LPS, induces surface CD5: two B-cell activation pathways. Int. Immunol. 3: 467–476.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Hippen K. L.,
    2. L. E. Tze,
    3. T. W. Behrens
    . 2000. CD5 maintains tolerance in anergic B cells. J. Exp. Med. 191: 883–890.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Lajaunias F.,
    2. L. Nitschke,
    3. T. Moll,
    4. E. Martinez-Soria,
    5. I. Semac,
    6. Y. Chicheportiche,
    7. R. M. Parkhouse,
    8. S. Izui
    . 2002. Differentially regulated expression and function of CD22 in activated B-1 and B-2 lymphocytes. J. Immunol. 168: 6078–6083.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Orrenius S.,
    2. B. Zhivotovsky,
    3. P. Nicotera
    . 2003. Regulation of cell death: the calcium-apoptosis link. Nat. Rev. Mol. Cell Biol. 4: 552–565.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Walker J. A.,
    2. K. G. Smith
    . 2008. CD22: an inhibitory enigma. Immunology 123: 314–325.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Nitschke L.,
    2. F. Lajaunias,
    3. T. Moll,
    4. L. Ho,
    5. E. Martinez-Soria,
    6. S. Kikuchi,
    7. M. L. Santiago-Raber,
    8. C. Dix,
    9. R. M. Parkhouse,
    10. S. Izui
    . 2006. Expression of aberrant forms of CD22 on B lymphocytes in Cd22a lupus-prone mice affects ligand binding. Int. Immunol. 18: 59–68.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Murakami M.,
    2. H. Yoshioka,
    3. T. Shirai,
    4. T. Tsubata,
    5. T. Honjo
    . 1995. Prevention of autoimmune symptoms in autoimmune-prone mice by elimination of B-1 cells. Int. Immunol. 7: 877–882.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    1. Lau C. M.,
    2. C. Broughton,
    3. A. S. Tabor,
    4. S. Akira,
    5. R. A. Flavell,
    6. M. J. Mamula,
    7. S. R. Christensen,
    8. M. J. Shlomchik,
    9. G. A. Viglianti,
    10. I. R. Rifkin,
    11. A. Marshak-Rothstein
    . 2005. RNA-associated autoantigens activate B cells by combined B cell antigen receptor/Toll-like receptor 7 engagement. J. Exp. Med. 202: 1171–1177.
    OpenUrlAbstract/FREE Full Text
  40. ↵
    1. Viglianti G. A.,
    2. C. M. Lau,
    3. T. M. Hanley,
    4. B. A. Miko,
    5. M. J. Shlomchik,
    6. A. Marshak-Rothstein
    . 2003. Activation of autoreactive B cells by CpG dsDNA. Immunity 19: 837–847.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Deane J. A.,
    2. P. Pisitkun,
    3. R. S. Barrett,
    4. L. Feigenbaum,
    5. T. Town,
    6. J. M. Ward,
    7. R. A. Flavell,
    8. S. Bolland
    . 2007. Control of toll-like receptor 7 expression is essential to restrict autoimmunity and dendritic cell proliferation. Immunity 27: 801–810.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Pisitkun P.,
    2. J. A. Deane,
    3. M. J. Difilippantonio,
    4. T. Tarasenko,
    5. A. B. Satterthwaite,
    6. S. Bolland
    . 2006. Autoreactive B cell responses to RNA-related antigens due to TLR7 gene duplication. Science 312: 1669–1672.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Heidari Y.,
    2. A. E. Bygrave,
    3. R. J. Rigby,
    4. K. L. Rose,
    5. M. J. Walport,
    6. H. T. Cook,
    7. T. J. Vyse,
    8. M. Botto
    . 2006. Identification of chromosome intervals from 129 and C57BL/6 mouse strains linked to the development of systemic lupus erythematosus. Genes Immun. 7: 592–599.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Liu K.,
    2. Q. Z. Li,
    3. Y. Yu,
    4. C. Liang,
    5. S. Subramanian,
    6. Z. Zeng,
    7. H. W. Wang,
    8. C. Xie,
    9. X. J. Zhou,
    10. C. Mohan,
    11. E. K. Wakeland
    . 2007. Sle3 and Sle5 can independently couple with Sle1 to mediate severe lupus nephritis. Genes Immun. 8: 634–645.
    OpenUrlCrossRefPubMed
  45. ↵
    1. Liu K.,
    2. Q. Z. Li,
    3. A. M. Delgado-Vega,
    4. A. K. Abelson,
    5. E. Sánchez,
    6. J. A. Kelly,
    7. L. Li,
    8. Y. Liu,
    9. J. Zhou,
    10. M. Yan,
    11. et al.
    2009. Kallikrein genes are associated with lupus and glomerular basement membrane-specific antibody-induced nephritis in mice and humans. J. Clin. Invest. 119: 911–923.
    OpenUrlCrossRefPubMed
  46. ↵
    1. Chen J.,
    2. P. A. McLean,
    3. B. G. Neel,
    4. G. Okunade,
    5. G. E. Shull,
    6. H. H. Wortis
    . 2004. CD22 attenuates calcium signaling by potentiating plasma membrane calcium-ATPase activity. Nat. Immunol. 5: 651–657.
    OpenUrlCrossRefPubMed
  47. ↵
    1. Chan V. W.,
    2. C. A. Lowell,
    3. A. L. DeFranco
    . 1998. Defective negative regulation of antigen receptor signaling in Lyn-deficient B lymphocytes. Curr. Biol. 8: 545–553.
    OpenUrlCrossRefPubMed
  48. ↵
    1. Smith K. G. C.,
    2. D. M. Tarlinton,
    3. G. M. Doody,
    4. M. L. Hibbs,
    5. D. T. Fearon
    . 1998. Inhibition of the B cell by CD22: a requirement for Lyn. J. Exp. Med. 187: 807–811.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    1. Cornall R. J.,
    2. J. G. Cyster,
    3. M. L. Hibbs,
    4. A. R. Dunn,
    5. K. L. Otipoby,
    6. E. A. Clark,
    7. C. C. Goodnow
    . 1998. Polygenic autoimmune traits: Lyn, CD22, and SHP-1 are limiting elements of a biochemical pathway regulating BCR signaling and selection. Immunity 8: 497–508.
    OpenUrlCrossRefPubMed
  50. ↵
    1. Gross A. J.,
    2. J. R. Lyandres,
    3. A. K. Panigrahi,
    4. E. T. Prak,
    5. A. L. DeFranco
    . 2009. Developmental acquisition of the Lyn-CD22-SHP-1 inhibitory pathway promotes B cell tolerance. J. Immunol. 182: 5382–5392.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section