Siglec-G Deficiency Leads to More Severe Collagen-Induced Arthritis and Earlier Onset of Lupus-like Symptoms in MRL/lpr Mice

Susanne Bökers, Anne Urbat, Christoph Daniel, Kerstin Amann, Kenneth G. C. Smith, Marion Espéli and Lars Nitschke
J Immunol April 1, 2014, 192 (7) 2994-3002; DOI: https://doi.org/10.4049/jimmunol.1303367

Abstract

Siglec-G is a member of the sialic acid–binding Ig-like lectin (Siglec) family expressed on all B cells. Siglec-G–deficient mice show a large expansion of the B1 cell compartment, demonstrating the crucial role of Siglec-G as an inhibitory receptor on this cellular subset. Although Siglec-G–deficient mice did not develop spontaneous autoimmunity, mice double-deficient for Siglec-G and the related Siglec protein CD22 did show autoimmunity at an older age. In this study, we addressed the question of whether loss of Siglec G on its own affects disease severity in animal models of rheumatoid arthritis and systemic lupus erythematosus. Siglec-G–deficient mice showed moderately increased clinical severity and higher inflammation of the knee joints following collagen-induced arthritis, when compared with control mice. The Siglec-G–deficient mouse was also backcrossed to the autoimmune prone MLR/lpr background. Although both Siglec-G–deficient and control MRL/lpr mice developed a lupus-like disease, Siglec-G–deficient MRL/lpr mice showed an earlier occurrence of autoantibodies; a higher lymphoproliferation of B and T cells; and an earlier onset of disease, as shown by proteinuria and glomerular damage in the kidney. Moreover, Siglec-G–deficient female mice showed a significantly reduced survival compared with female control MRL/lpr mice. Thus, the loss of the inhibitory receptor Siglec-G led to a moderate exacerbation of disease severity and early onset in both collagen-induced arthritis and spontaneous lupus nephritis in MRL/lpr mice.

Introduction

B cells are generated in the bone marrow, where central B cell tolerance induction takes place. This tolerance induction by inactivation of autoreactive B cells in the bone marrow is achieved by different mechanisms, such as B cell depletion, receptor editing, or anergy induction (1). Surprisingly, however, many human peripheral B cells carry BCRs with some degree of autoreactivity (2). Furthermore, autoreactive Abs can be generated through somatic hypermutation during germinal center responses. This observation raises the question of how autoimmunity caused by stimulation of self-reactive BCRs can be avoided. There is good evidence that inhibitory receptors expressed on the B cell surface play an important role in downmodulating BCR signaling and in reducing B cell activation by autoantigens (3). Inhibitory receptors carry Ig-like tyrosine-based inhibitory motifs (ITIMs). These motifs are phosphorylated by the tyrosine kinase Lyn and are subsequently bound by phosphatases such as the tyrosine phosphatase SHP-1 (4), which downregulate BCR signaling.

Genetically modified mice that are deficient for ITIM-carrying inhibitory receptors, which are expressed on B cells such as FcγRIIb (5), PIR-B (6), or CD72 (7), all develop symptoms of lupus-like autoimmune disease to various degrees. The same is true for mice, which lack the downstream signaling molecules Lyn (8) or SHP-1 (4). CD22 and Siglec-G also belong to the class of inhibitory receptors that downmodulate B cell responses (9). They are members of the sialic acid–binding Ig-like lectin (Siglec) family. Neither CD22-deficient nor Siglec-G–deficient mice develop autoimmunity; however, CD22 × Siglec-G double-deficient mice develop a lupus-like disease with occurrence of the typical autoantibodies and glomerulonephritis (10). This finding indicates functional redundancy of these two Siglecs in the control of B cell tolerance. In addition, mutation in the sialic acid acetyl esterase gene, which codes for an enzyme that modifies CD22 and Siglec-G ligands, leads to autoimmunity in mice and is linked to several autoimmune diseases in humans (11, 12). Moreover, the loss of CD22 alone can contribute to autoimmunity, as was demonstrated by crossing Cd22−/− mice to the Y chromosome–linked autoimmune accelerator background (13). Whether Siglec-G deficiency on its own can also contribute to the development of autoimmunity is still unknown.

Siglec-G, the mouse ortholog of human Siglec-10, is expressed on B cells, as well as on dendritic cell (DC) populations (14). Like all Siglecs, Siglec-G binds to sialic acids, terminal carbohydrates abundantly expressed in higher vertebrates and found on many soluble and cell surface–bound glycoproteins. Siglec-G preferentially binds ligands carrying a sialic acid in α 2,3 or α-2,6 linkage attached to galactose on the glycan core. Siglec-G carries an ITIM and an ITIM-like sequence (9). For human Siglec-10, it has been demonstrated that the tyrosine-phosphorylated ITIM is bound by SHP-1 and SHP-2 phosphatases (15). Siglec-G–deficient mice showed a phenotype restricted to a subpopulation of B cells, the B1 cells (16). Siglec-G–deficient mice had a highly increased B1 cell population. Ca2+ signaling was highly elevated in Siglec-G−/− B1 cells, but not in conventional B2 cells. Siglec-G−/− B1 cells showed increased survival, both in vitro and in vivo, when compared with wild-type (WT) B1 cells (17). These data showed that Siglec-G is an inhibitory receptor that exhibits B1 cell–restricted functions in knockout (KO) mice, although Siglec-G is equally expressed on conventional B2 cells (14). Recently, we were able to show that the sialic acid–binding domain of Siglec-G is crucial in regulating its inhibitory function. Mice expressing a mutated form of Siglec-G that can no longer bind to sialic acids showed a loss of IgM–Siglec-G association, especially in B1 cells, leading to higher B1 cell Ca2+ responses. Furthermore, these mice had a phenotype similar to that of mice with no Siglec-G expression (S. Hutzler, L. Özgör, Y. Naito-Matsui, K. Kläsener, T. Winkler, M. Reth, and L. Nitschke, unpublished observations).

B1 cells of the mouse have characteristic cell surface markers and are found in the pleural and peritoneal cavities and spleen and have a self-renewing capacity (18, 19). Many studies have shown that B1 cell maintenance depends on B1 cell signaling, which explains why the loss of the inhibitory receptor Siglec-G on B1 cells leads to expansion of this population. B1 cells secrete natural IgM Abs that are important for early antibacterial responses, as they can bind common bacterial surface structures. The Ig repertoire of IgM secreted by B1 cells is skewed toward germline-encoded polyreactive, weakly autoreactive specificities (18, 19). Some studies claimed that increased B1 cells are functionally involved in autoimmunity, as autoimmune mouse strains such as NZB/NZW mice have largely increased B1 cell numbers (20). In contrast, it has been demonstrated that polyreactive IgM secreted by B1 cells is important for clearance of apoptotic cells, which would be a natural and nonpathological function of these self-reactive Abs (18).

In this study we aimed to address the question of whether loss of Siglec-G can contribute to induction and severity of autoimmunity. For this we chose collagen-induced arthritis (CIA) as a model for rheumatoid arthritis and the mouse MRL/lpr strain as a mouse model for systemic lupus erythematosus (SLE). By analyzing these two models, we demonstrated that Siglec-G expression is important for the control of disease onset and contributes moderately to disease severity.

Materials and Methods

Mice

Siglec-G–deficient BALB/c mice have been described previously (16). The Siglec-G–deficient mice were backcrossed to the MRL/MpJ-Fas/lpr (MRL/lpr) background for five generations by “speed congenics” (see details later in the article). To obtain Siglec-G−/− MRL/lpr and Siglec-G+/+ MRL/lpr littermates, only heterozygous mice were bred. All experiments were performed according to the regulations of the U.K. Home Office Scientific Procedures Act (1986) or to the German law for protection of animals, after approval by the animal welfare committee.

Marker-assisted congenics

The BALB/c donor strain containing the mutated Siglec-G gene was backcrossed into the MRL/MpJ fas/lpr recipient strain. For the backcrossing strategy of “marker-assisted congenics,” microsatellites as genetic markers were used to select individuals of each offspring with both the required chromosomal fragment and as much as possible of genetic background for the recipient strain. For backcrossing the Siglec-G deficiency on the lupus-prone MRL/MpJ fas/lpr background, one to three genetic markers per chromosome were selected and produced PCR products with ≥6 bp differences in size, which were distinguishable on a 4% agarose gel. The PCR primer pairs for 49 genetic microsatellite markers are listed in Table I. After five generations of backcrossing, mice containing all MRL-derived markers were obtained. Mice were also tested for Siglec-G deficiency (16) and for the lpr mutation in the fas gene locus with the PCR primers 5′-GTA AAT AAT TGT GCT TCG TCA G-3′, 5′-TAG AAA GGT GCA CGG GTG TG-3′, and 5′-CAA ATC TAG GCA TTA ACA GTG-3′.

Collagen-induced arthritis

Chicken collagen type II [(CII); Sigma-Aldrich] was dissolved in 10 mM acetic acid overnight to obtain a 4 mg/ml solution and was then combined with an equal volume of keyhole limpet hemocyanin (KLH) (4 mg/ml). The collagen/KLH mix was emulsified with an equal volume of CFA (5 mg/ml, Chondrex). Mice were injected with 50 μl emulsion each side intradermally into the lower back. After 21 d, the mice were boosted with a second injection.

Clinical scoring of CIA symptoms

Mice were scored for CIA symptoms three times per week (Monday, Wednesday, Friday), starting from day 21 for up to 63 d. Individual limbs were scored from 0 to 3 for the severity of clinical disease (thus, the maximum score per mouse was 12). Scores were as follows: 0: no swelling; 1: one or two toes swollen or mild swelling, erythema and redness of the paw; 2: more than two swollen toes and/or moderate swelling, erythema and redness of the paw; 3: severe swelling, erythema, joint ankylosis, stiffness, or distortion.

Histology

At the end of the CIA experiment, the mice were sacrificed, and paws and knees of the mice were removed, fixed in formalin, decalcified in 10% EDTA in H2O for ≥20 d, and paraffin embedded. Sections of 5 μm were stained with H&E and analyzed blinded for histological signs of inflammation, pannus formation, and cartilage and bone destruction. Knees of mice challenged with CIA were stained with H&E and scored blinded for signs of arthritis. Scores were as follows: 0: normal appearance; 1: mild inflammation and synovitis, mild cartilage destruction without bone erosion; 2–4: increasing degrees of inflammatory cell infiltration, synovitis and pannus formation, cartilage destruction, and bone erosion.

Kidneys of Siglec-G MRL/lpr were fixed in formalin, and sections of 5 μm were stained with H&E and with periodic scid–Schiff stain for blinded scoring for lupus nephritis.

Kidney sections were deparaffinized in Xylol two times for periods of 3 min. Slides were then rehydrated in a graded ethanol series, as described above. Periodic acid solution 0.5% was added in excess for 10 min. Slides were then left to rinse in deionized water for 5 min before being saturated with Schiff reagent for 15 min, followed by another washing step in lukewarm tap water for 5 min. Excess of hematoxylin was added to the slides for 30 s, and they were washed again in tap water for 5 min. Slides were then dehydrated in a graded ethanol series followed by Xylol. As the final step, the slides were mounted with DPX mounting medium and left to dry overnight.

Kidney periodic scid–Schiff staining was scored blinded for renal morphology. Sections were scored for glomerular (hypercellularity, thickening of glomerular basement membrane, apoptotic cells, matrix expansion, extracapillary proliferation) and tubulointerstitial (tubular atrophy, interstitial fibrosis, inflammation) changes using the following scoring system: 0: normal; 1: altered. Scores of different subclasses were added to glomerular and tubulointerstitial changes.

ELISA

Levels of anti-CII were measured by ELISA. MaxiSorp (NUNC, Thermo Fisher Scientific) 96-well ELISA plates were coated with ELISA grade type II collagen in deionized H2O (5 μg/ml, 50 μl per well) and incubated overnight at 4°C or at 37°C for a minimum of 2 h. After washing three times with PBS supplemented with 0.5% Tween 20, the plates were blocked with 1% BSA in PBS (200 μl per well) for 1 h at room temperature (RT). After another wash, sera were added in 3-fold serial dilutions in PBS supplemented with 0.5% Tween 20, starting at 1:100 for all IgG Ab subtypes and 1:200 for IgM. Pooled serum of mice, challenged with collagen, served as the standard, with a starting dilution of 1:50 (1:100 for IgM). Plates were incubated for 2 h at RT or overnight at 4°C and washed again three times before alkaline phosphatase (AP)–conjugated anti-mouse IgM, IgG1, IgG2a, IgG2b, and IgG3 Abs (1:1000, 50 μl) were used to detect the relevant Igs by incubation at 37°C for 2 h.

Levels of anti-dsDNA Abs were measured by ELISA. MaxiSorp plates were precoated with poly-L-lysine 0.01% w/v (Sigma-Aldrich) for 2 h at RT or overnight at 4°C and coated with dsDNA from calf thymus (20 μg/ml; Sigma-Aldrich) in H2O for 2 h at 37°C or overnight at 4°C. Plates were washed with PBS/0.05% Tween 20. Sera were added in 1/3 serial dilutions starting at 1/50. Pooled serum of SLE affected MRL/lpr mice, with a starting dilution of 1/150 serving as the standard. AP-conjugated anti-mouse IgM, IgG, IgG1, IgG2a, IgG2b, and IgG3 were used to detect the relevant Abs.

Levels of rheumatoid factor (RF) of the IgM isotype were measured by ELISA. MaxiSorp plates were coated with purified IgG from rabbit serum (10 μg/ml; Sigma-Aldrich) in NaHCO3, pH 8.5, for 2 h at RT, followed by blocking with PBS/0.05 Tween 20/2% FCS for 1 h at RT. After washing with PBS, sera were added in 1/3 serial dilutions, starting at 1/100. Pooled serum of SLE affected MRL/lpr mice, with a starting dilution of 1/80 serving as the standard. After washing, AP-conjugated goat anti-mouse IgM was added.

For RF IgG ELISA, MaxiSorp plates were coated with mAb 23,3 anti-mouse IgG2a (1 μg/ml) in PBS at 4°C overnight, followed by blocking with PBS/1% BSA for 2 h at 37°C. After washing with PBS, sera were added in 1/3 serial dilutions, starting at 1/200. Pooled serum of SLE affected MRL/lpr mice, with a starting dilution of 1/80 serving as the standard. After washing, a biotinylated anti-κ L chain Ab (clone: 187.1, 1 μg/ml; BD Pharmingen) was incubated for 2 h at 37°C. After another washing, AP-conjugated streptavidin (Life Technologies/BRL) was added (1/500 diluted) and incubated for 2 h at 37°C.

All ELISA plates were revealed using p-nitrophenylphosphate in diethanolamine buffer, read with 405-nm wavelength, and analyzed using Softmax Pro Software. The absorbance values were fit to a curve, using the four-parameter fit.

Antinuclear Abs

Antinuclear Abs (ANAs) were assessed using a commercial indirect fluorescent Ab assay (ANA HEp-2 slide; Cambridge Life Sciences) according to the manufacturer’s instructions. Serum from each mouse was added to a well at a dilution of 1/400 and was incubated for 30 min at RT before being washed with PBS for 10 min. Wells were incubated for 30 min with Alexa Fluor 488–conjugated rabbit anti-mouse IgG Ab (Invitrogen) at 1:500. After a second wash with PBS, slides were mounted with Mowiol (Calbiochem)

Flow cytometric analysis

After preparing a single-cell suspension of spleen and peritoneal lavage, the cells were treated with Gey’s solution for 5 min at RT. Cells were stained with combinations of anti-B220 PeCy5.5 (RA3-6B2; eBioscience) or Pacific Orange (RA3-6B2; Invitrogen), anti-IgM PE (II/4l; eBioscience), anti-CD5 PE (53-7.3; BD Biosciences), anti-CD19 FITC or biotin (ID3α, own hybridoma) or eFluor 450 (eBio1D3; eBioscience), anti-CD3 allophycocyanin (145-2C11; eBioscience), anti-CD4 PE (GK1.5; BD Bioscience) or Pacific Orange (RM4-5; Invitrogen); anti-CD8a FITC or allophycocyanin (53-6.7; eBioscience), anti-CD138 allophycocyanin or PE (281-2; BD Bioscience), and anti-GL7 biotin or Alexa Fluor 647 (GL7; eBioscience).

Biotinylated Abs were detected with streptavidin conjugated to FITC, PeCy5.5, PE (all eBioscience), or allophycocyanin (BD Biosciences). Cells were analyzed using a flow cytometer (FACSCalibur; BD Biosciences or Dako Cyan; Beckman Coulter) and FlowJo software (TreeStar).

Staining of immune complexes in the kidney

For immunohistological analysis, kidneys were snap-frozen in OCT Medium directly after being harvested. After sectioning on a cryostat, 6-μm sections were fixed with ice-cold acetone for 5 min. Kidney sections were saturated with 20% horse serum in PBS for 1 h at RT and stained with 100 μl goat anti-mouse IgG Alexa Fluor 488 1:200 (Invitrogen). After mounting with Mowiol mounting medium, slides were dried and analyzed on a fluorescence microscope. For all kidney sections, the microscope settings (exposure time, objective, illumination, and filter cube) were identical to allow comparison among samples. The mean fluorescence intensity in the sections was measured using the Image J software.

Measurement of proteinuria

Fresh mouse urine was tested for protein with reagent strips for urine analysis (Albustix) by comparing the color change of the strip with a given color code. Six different readings can be distinguished: negative, trace, 30, 100, 300, and ≥2000 mg/dL protein.

Statistical analysis

Two experimental groups of independent samples were compared using the Mann–Whitney U test. In the case of parametric samples and n > 10, a log transformation was performed in case this reduced variance in-homogeneity before the respective Student's t test was used. For comparison of three or more groups, ANOVA was used. The log-rank test was performed to compare two groups for survival and cumulative incidence. For individual time points of incidence, the χ2 test was used. Statistical tests were performed with GraphPad Prism (GraphPad software) and Statistica (StatSoft, Tulsa, OK). All values are given as mean ± SEM unless otherwise noted. Differences were considered significant at p < 0.05.

Results

Siglec-G–deficient mice show a moderately increased severity and earlier onset of CIA compared with BALB/c control mice

To examine the role of Siglec-G in rheumatoid arthritis, we used the model of inflammatory joint disease CIA. CIA is strongly dependent on the mouse MHC alleles. Mouse strains with H-2q or H-2r alleles are susceptible, whereas C57BL/6 (H-2b) and BALB/c (H-2d) are resistant strains (21). Because Siglec-G–deficient mice were available only on BALB/c background, we used a described modified immunization protocol (optimized for BALB/c) consisting of CII in a higher concentration of CFA, mixed with KLH included in the collagen inoculum (22). Siglec-G–deficient mice on the BALB/c background and WT BALB/c mice were challenged with an emulsion of CII and KLH in CFA. After 21 d, animals were boosted with the same mixture and monitored for clinical arthritis signs over a period of ≥64 d. Combined data of three independent experiments showed a moderately higher clinical arthritis score in Siglec-G–deficient mice compared with WT control mice (p = 0.017) (Fig. 1A).

FIGURE 1.
FIGURE 1.

Loss of Siglec-G leads to earlier onset and more severe CIA in BALB/c mice. (A) Clinical arthritis score for Siglec-G–deficient (n = 27) compared with WT BALB/c (n = 25) mice. Mice were immunized on day 0 and day 21. Data represent three independent CIA experiments combined; p = 0.017. The p values for clinical arthritis score were calculated using two-way ANOVA [ANOVA F (1.16) = 6.1]. (B) Cumulative arthritis incidence for Siglec-G–deficient (n = 27) and WT BALB/c (n = 25) mice (score > 1). Data represent three independent CIA experiments combined. Significance for individual time points was tested with the χ2 test. *p < 0.05, **p < 0.01. (C) Histological arthritis score of H&E-stained knee joints in Siglec-G–deficient (n = 18) and WT BALB/c (n = 17) mice. Original magnification ×40. Data represent two independent experiments combined. (D) Flow cytometric endpoint analysis of CII-challenged Siglec-G–deficient (n = 27) and WT BALB/c (n = 25) mice. Data represent three independent experiments combined. Left panel, Total cells in the spleen; right panel, total cell count of plasmablasts in the spleen. (E) Sera from Siglec-G–deficient and WT BALB/c mice were tested for anti-chicken CII Abs of IgM and IgG1 isotype. (F) Sera from Siglec-G–deficient (n = 27) and WT BALB/c (n = 25) mice were tested for anti-mouse CII Abs of IgM, IgG1, and IgG2a isotypes. Data represent three independent experiments combined.

The cumulative incidence of arthritis in Siglec-G–deficient and WT BALB/c mice showed an earlier onset of disease (Fig. 1B). After 64 d, mice were sacrificed, and knee joints of CIA-challenged Siglec-G−/− and WT BALB/c mice were processed, stained for H&E, and scored blinded for histological signs of arthritis, such as cartilage destruction, bone erosion, and cell infiltration (severity score: 0–4). As examples, one joint of a WT mouse (score 1) and one joint of a Siglec-G–deficient mouse (score 2) are shown (Fig. 1C). Scoring each individual knee joint of 17 WT and 18 Siglec-G−/− mice revealed more severe joint destruction in Siglec-G–deficient mice than in WT BALB/c mice; p = 0.045 (Fig. 1C).

At the end point of the experiment (day 64), the splenic cell populations of CII-challenged Siglec-G–deficient and WT BALB/c mice were analyzed by flow cytometry (Fig. 1D). No difference was seen in the total cell count of spleen cells between Siglec-G–deficient and WT BALB/c mice (Fig. 1D, left). A significant increase of plasmablasts (B220lo, CD138+) was seen in Siglec-G–deficient BALB/c mice challenged with CII, compared with their WT littermates (Fig. 1D, right), but cell populations like T cells and B1, B2, and germinal center B cells in the spleen did not show differences (data not shown).

CII-challenged Siglec-G–deficient and WT BALB/c mice were tested for anti-chicken CII Abs of the IgM, IgG1 (Fig. 1E), IgG2a, and IgG2b isotype (data not shown). ELISAs were performed with sera from the last time points, when animals were sacrificed. Levels of anti-CII IgM Abs were significantly increased in Siglec-G–deficient mice compared with WT BALB/c mice, whereas levels of the IgG subclasses were not significantly different. In addition, anti-mouse collagen type II Abs were determined by ELISA. Anti-mouse collagen IgM Abs were significantly increased, and IgG1 and IgG2a showed a higher tendency in Siglec-G−/− mice (Fig. 1F). Taken together, our data suggest that Siglec-G contributes moderately to the control of CIA onset and severity.

Siglec-G–deficient MRL/lpr mice show increased autoantibody titers

The MRL/lpr mouse strain is a widely used mouse model for the autoimmune disease SLE. MRL/lpr mice develop autoantibodies, such as anti-dsDNA or ANA, and lupus nephritis and die of the consequences of this autoimmune disease (23). To examine the role of Siglec-G in a lupus-prone mouse model, the mutated Siglec-G null allele was backcrossed from BALB/c into the MRL/lpr background by the speed congenics technique (Table I) (24). The severity and onset of disease in the MRL/lpr strain, as in human SLE, show a sex bias toward females. We thus analyzed male and female MRL/lpr mice in separate groups. We indeed observed a slightly more severe disease phenotype in female than in male MRL/lpr mice. We first analyzed the appearance of autoantibodies. Anti-dsDNA autoantibodies were analyzed by ELISA. Anti-dsDNA autoantibodies of the IgM and IgG3, but not other isotypes, were increased in Siglec-G–deficient versus control MRL/lpr mice, affecting in each case one sex only (Fig. 2A).

View this table:
Table I. Primer pairs for microsatellites used for speed congenic backcrossing Siglec-G−/−BALB/c mice into the MRL/MpJ background
FIGURE 2.
FIGURE 2.

Elevated anti-dsDNA, RF IgM, and ANA levels in Siglec-G–deficient MRL/lpr mice. (A) Anti-dsDNA Abs (IgM, IgG2a, IgG2b, and IgG3 isotype) in 18- to 25-wk-old mice were measured by ELISA. (B) Semiquantitative analysis to measure ANAs. Upper panel, Indirect fluorescence autoantibody assay using HEp-2 slide representing examples of 8-wk female Siglec-G−/− MRL/lpr, Siglec-G+/+ MRL/lpr, and Siglec-G−/− BALB/c mice (negative control). Lower panel, Fluorescence intensity in 8-wk and in 18- to 25-wk-old mice. Original magnification ×10. (C) Left panel, Elevated RF IgM level in 18- to 25-wk-old female Siglec-G–deficient MRL/lpr mice; right panel, RF IgG level in 18- to 25-wk-old Siglec-G–deficient and WT MRL/lpr mice.

ANAs were examined using an indirect immune fluorescence Ab assay with HEp-2 slides. In Fig. 2B, representative slides can be seen, showing samples of 8-wk-old female Siglec-G−/− (left), WT MRL/lpr (middle), and Siglec-G−/− BALB/c (right) mice. In 8-wk-old female MRL/lpr mice, a striking increase in ANAs was seen in the Siglec-G–deficient animals, when compared with controls (Fig. 2B). Male Siglec-G–deficient MRL/lpr mice at that age showed no significant difference in ANAs. In 18- to 25-wk-old mice, a tendency in the female Siglec-G KO mice for higher ANAs was found, but the difference was not significant (Fig. 2B). In Siglec-G–deficient BALB/c mice, which served as control, no ANAs were detectable. Female, but not male, Siglec-G–deficient MRL/lpr mice showed higher titers of RF of the IgM isotype at the age of 18 to 25 wk (Fig. 2C, left). Siglec-G–deficient BALB/c control titers were at the same level as for the male MRL/lpr mice. Levels of RF of the IgG isotype showed no differences between Siglec-G–deficient and WT MRL/lpr mice (Fig.2C, right).

Male Siglec-G–deficient MRL/lpr mice show increased B and T cell numbers

MRL/lpr mice have a lymphoproliferative disease with increased lymphocyte populations and alterations in cell populations owing to the fas defect of the lpr mutation. Lymphocyte numbers in the spleen of Siglec-G–deficient MRL/lpr mice were compared with control MRL/lpr mice by flow cytometric analysis. We focused on cell populations, which are known to contribute to the development of SLE. In male Siglec-G−/− MRL/lpr mice, the total cell count of B cells (B220+, CD19+) (Fig. 3A), germinal center B cells (B220+,CD19+, Gl7+) (Fig. 3B), and plasmablasts (B220lo, CD138+) (Fig. 3C) was significantly increased, when compared with control MRL/lpr mice. Notably, no differences were seen in the female MRL/lpr mice for these cell populations. When B1 cells were analyzed in spleen and peritoneal cavity, it was noted that Siglec-G−/− MRL/lpr mice did not show the large increase of B1 cell numbers, which is typical in Siglec-G−/− mice on a BALB/c background (16) (Fig. 3D, 3E).

FIGURE 3.
FIGURE 3.

Increased levels of cell counts for B and T cell populations in the spleen of 12-wk-old male Siglec-G–deficient MRL/lpr mice. Flow cytometric analysis. (A) B cells (B220+, CD19+). (B) Germinal center B cells (B220+, CD19+, GL7+). (C) Plasmablasts (B220lo, CD138+). (D) B1 cells, spleen (CD5+ IgM+). (E) B1a cells, peritoneal cavity (CD5+ IgM+). (F) CD4+ T cells (B220−, CD3+, CD4+). (G) CD8+ T cells (B220−, CD3+, CD8+). (H and I) Unusual T cells (B220+, CD3+, CD4+, or CD8+).

The T cell subsets CD4+ and CD8+ T cells (Fig. 3F, 3G) were also increased in male Siglec-G−/− MRL/lpr mice compared with control MRL/lpr mice, whereas for the unusual T cell populations (B220+ CD4+ and B220+ CD8+ T cells), which are typical for MRL/lpr mice, no significant increase for male Siglec-G−/− MRL/lpr mice compared with WT MRL/lpr mice was seen (Fig. 3H, 3I). A generally higher lymphoproliferation for B and T cell subsets was observed for female than for male mice, which affected both WT and Siglec-G−/− genotypes. The differences between Siglec-G−/− and WT mice were more evident in the less affected male mice (Fig. 3).

Male Siglec-G−/− MRL/lpr mice show an earlier onset of glomerular damage in the kidney

To follow the development of kidney disease, protein in the urine was measured each week in MRL/lpr mice. At 19 wk, >50% of the mice had developed significant proteinuria; therefore, 19-wk-old mice were sacrificed, and kidney histological studies were performed to detect pathological alterations in the kidney morphology. Histopathological evaluation of kidneys from female MRL/lpr mice showed renal injury typical for experimental lupus nephritis, including glomerular hypertrophy, intracapillary thrombus formation, abnormal glomerular basement membrane, mesangial cell proliferation, and matrix expansion independent of the Siglec-G genotype (Fig. 4 C, 4D). In addition, apoptotic cells and extracapillary proliferation were also frequently observed (not shown). Similar glomerular changes were observed in male Siglec-G−/− MRL/lpr mice (Fig. 4B). In contrast, male Siglec-G+/+ MRL/lpr mice showed significantly less glomerular injury, indicating a role for Siglec-G in the pathogenesis of lupus nephritis (Fig. 4A, 4E). We observed only minor tubulointerstitial alterations in renal biopsies, showing no significant differences between all investigated groups (Fig. 4F). Immune complex deposition containing IgG was detected in the kidney of both control and Siglec-G–deficient MRL/lpr mice and occurred to a similar extent (Fig. 5)

FIGURE 4.
FIGURE 4.

More severe glomerular alteration in male Siglec-G–deficient MRL/lpr mice. Representative photomicrographs of periodic acid–Schiff–stained renal biopsy specimens for 19-wk old male and female MRL/lpr of Siglec-G+/+ and Siglec-G −/− genotype are shown (A–D). Glomerular alterations were scored for typical changes observed in lupus nephritis (E). Tubulointerstitial changes, including acute tubular necrosis, atrophy, fibrosis, and inflammation, were monitored for all mice (F). n = 10 for (E) and (F) per mouse group and sex.

FIGURE 5.
FIGURE 5.

Similar immune complex IgG deposition detected in the kidney of all MRL/lpr mice. Kidney sections were stained with anti-IgG Abs. Examples of stainings of Siglec-G+/+ and Siglec-G−/− MRL/lpr mice, as well as WT BALB/c and Siglec-G−/− BALB/c mice, are shown on the left. Glomerular depositions were quantified with the ImageJ program, and fluorescence intensity is given on the right. Original magnification ×20.

Earlier onset of proteinuria and lower survival of female Siglec-G−/− MRL/lpr mice

Immune complex deposition in the glomeruli supports an inflammatory response that subsequently leads to glomerular nephritis. The amount of protein found in the urine gives a rough idea of the extent of glomerular nephritis. During the observation period of 250 d, proteinuria was measured once a week with urine dipsticks and scored in the range of 0–4, corresponding from “not detectable” to “protein >2000 mg/dl.” The cumulative proteinuria score increased in both genotypes over time. Analysis revealed no significant difference between the two groups, neither in male nor in female individuals (Fig. 6A, 6B). However, the severe onset of proteinuria (≥300 mg/dl) was earlier in female Siglec-G–deficient MRL/lpr mice compared with female control MRL/lpr mice (Fig. 6C). MRL/lpr mice are known to eventually die of lupus nephritis. Kaplan–Meier survival curves showed a similar incidence of death in control MRL/lpr male and female mice, whereas female Siglec-G–deficient MRL/lpr showed a significantly lower survival (Fig. 6D). In addition, when survival curves of both genders were combined, significantly lower survival of Siglec-G−/− MRL/lpr compared with WT MRL/lpr mice was observed (Fig. 6E).

FIGURE 6.
FIGURE 6.

Earlier onset of proteinuria and shorter survival of female Siglec-G–deficient MRL/lpr mice. (A) Cumulative proteinuria score of male (left panel, n = 10–30) and (B) female (right panel, n = 10–33) Siglec-G–deficient and WT MRL/lpr mice. (C) Onset (in days) of severe proteinuria (≥300 mg/dl) in Siglec-G–deficient mice compared with WT MRL/lpr mice of both sexes. (D) Survival of Siglec-G–deficient MRL/lpr compared with WT MRL/lpr mice of both sexes, separately. Female (KO, n = 16; WT, n = 28) and male (KO, n = 26; WT, n = 38). (E) Survival of Siglec-G–deficient MRL/lpr (n = 42) compared with WT MRL/lpr mice (n = 66) of both sexes combined. Siglec-G–deficient and WT MRL/lpr mice were observed over a period of 250 d. Log-rank test for KO compared with WT MRL/lpr mice (p = 0.03 for female mice, p = 0.04 for all mice).

Discussion

This study shows that Siglec-G deficiency affects the severity of CIA. The clinical arthritis score of paws and arthritis in inflamed knee joints was modestly elevated in Siglec-G–deficient mice. In addition, in the MRL/lpr mouse lupus model, we observed an earlier onset of some types of autoantibodies, as well as increased numbers of activated B cells and T cells, in Siglec-G–deficient mice. Male Siglec-G−/− MRL/lpr mice showed an earlier onset of glomerular damage in the kidney, and female Siglec-G−/− MRL/lpr mice showed an impaired survival when compared with control MRL/lpr mice. These data reveal that in both models the loss of Siglec-G leads to moderately increased severity of the studied autoimmune diseases.

In CIA we observed increased plasmablast numbers and increased collagen-specific IgM, but only modestly increased IgG1 Abs in Siglec-G–deficient mice. B cells and Abs play an important role in CIA, as B-cell–deficient mice are resistant to CIA and the disease can be induced by transferred collagen-specific Abs. It is likely that a general overstimulation of Siglec-G–deficient B cells is responsible for the moderately higher disease severity observed in Siglec-G−/− mice. Collagen-specific Abs were measured only at the endpoint, when mice were sacrificed. It is possible that stronger differences of collagen-specific IgG Abs were present during earlier time points of disease progression. Because Siglec-G is also expressed on DCs (14), the contribution of enhanced DC activity to disease progression cannot be excluded (see below). In the MRL/lpr model, we observed an earlier development of some autoantibody types in a gender-specific way. Female MRL/lpr mice showed an earlier occurrence of ANA, for example. MRL/lpr mice are also characterized by lymphoproliferation, which was detected at earlier time points in female than in male mice. In this case, we observed that male Siglec-G–deficient mice generally showed a larger increase of activated B cell or T cell numbers, compared with male control MRL/lpr mice (Fig. 3). These increased cell numbers in Siglec-G–deficient mice reached the level of control female MRL/lpr mice. The same was true for glomerular damage of glomeruli in the kidney. At 19 wk of age, control female MRL/lpr mice were more severely affected than control male mice, whereas both male and female Siglec-G–deficient mice reached the same disease score as female controls (Fig. 4). Finally, an earlier onset of proteinuria was observed in female Siglec-G−/−, compared with control MRL/lpr mice, and female Siglec-G−/− MRL/lpr mice showed an impaired survival. In humans, SLE is known to show a higher incidence in women than in men (25). It is interesting that Siglec-G–deficient mice seem to reflect this sex-specific difference in disease severity to some extent.

Overall, Siglec-G–deficient mice show only a modest degree of increased severity in both CIA and the mouse SLE model. This finding is in contrast to other inhibitory receptors. For example, mice deficient for the inhibitory IgG receptor (FcγRIIb−/− mice) show a more severe phenotype in CIA, as well as in the MRL/lpr model (26, 27). However, the severity of the disease depends on the genetic background (28). In addition, when the FcγRIIb was overexpressed on B cells of transgenic mice, a suppressed CIA or suppressed lupus disease in MRL/lpr mice was observed (24, 29). Siglec-G has a close homolog, CD22 on B cells. Siglec-G × CD22 double-deficient mice develop autoimmunity and a lupus-like disease, whereas Siglec-G−/− mice do not show this (10). Therefore, it is likely that CD22 also plays compensatory functions to Siglec-G in the CIA and in the MRL/lpr models. We would therefore expect a stronger phenotype in double-KO mice in both models. This idea will be tested in the future.

The moderately increased disease severity in both the CIA and the MRL/lpr model can be explained by enhanced B cell activities of Siglec-G–deficient mice. This activity was indicated by enhanced autoantibodies, enhanced plasmablast numbers, or enhanced germinal center B cells. However, Siglec-G is also expressed on DC subpopulations (Ref. 14; S. Hutzler et al., unpublished observations), and Siglec-G–deficient mice have been shown to exhibit increased inflammatory responses, in both a liver injury and a bacterial infection model (30, 31). Siglec-G expression on DCs was suggested to be responsible for these effects. In the MRL/lpr model, we did not observe increased DC populations but did not characterize DCs further. Thus, a contribution of more highly activated DCs of Siglec-G−/− mice to both CIA and lupus disease in MRL/lpr mice cannot be excluded.

Siglec-G–deficient mice have shown largely B1 cell–restricted phenotypes (16, 17). However, both in rheumatoid arthritis and in lupus, high-affinity IgG autoantibodies, which cannot be generated by B1 cells, are crucial. We therefore think that the contribution of Siglec-G in these diseases is a function on conventional B cells, where Siglec-G, to some extent, suppresses formation of high-affinity Abs in a T cell–dependent germinal center response. The contribution of B1 cells to these processes is unlikely. Of note, we did not observe a highly increased B1 cell population in Siglec-G–deficient MRL/lpr mice, which is typical for Siglec-G–deficient BALB/c mice. Therefore, we conclude that B1 cells play no crucial role in increasing the disease severity of MRL/lpr mice. Notably, Siglec-G can also transmit inhibitory signals on conventional B2 cells when targeted with high-affinity Siglec-G ligands (14). Thus, the Siglecg gene seems to be a genetic factor influencing the severity of both murine rheumatoid arthritis and SLE models. Because of Siglec redundancies, potential future Siglec-directed therapies should target both Siglecs, Siglec-G/Siglec-10, and CD22 on B cells simultaneously.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Dr. Lorna Jarvis for assistance in animal husbandry, Dr. Paul Lyons for discussions, Dr. Thomas Winkler for discussions and providing reagents, Dr. Michael Fischer for help with statistical analysis, and Dr. Mark Shlomchik for providing reagents.

Footnotes

  • This work was supported by the Deutsche Forschungsgemeinschaft (Grants FOR832, TRR130, and SFB643), the Women in Science Prize from the Faculty of Natural Sciences at the Friedrich-Alexander-Universität (to S.B.), the Wellcome Trust, and the National Institute for Health Research Cambridge Biomedical Research Centre.

  • Abbreviations used in this article:

    ANA
    antinuclear Ab
    AP
    alkaline phosphatase
    CIA
    collagen-induced arthritis
    CII
    chicken collagen type II
    DC
    dendritic cell
    ITIM
    Ig-like tyrosine-based inhibitory motif
    KLH
    keyhole limpet hemocyanin
    KO
    knockout
    RF
    rheumatoid factor
    RT
    room temperature
    Siglec
    sialic acid–binding Ig-like lectin
    SLE
    systemic lupus erythematosus
    WT
    wild-type.

  • Received December 17, 2013.
  • Accepted January 18, 2014.

References

  1. ↵
    1. Nemazee D.
    2000. Receptor editing in B cells. Adv. Immunol. 74: 89–126.
    OpenUrlPubMed
  2. ↵
    1. Wardemann H.,
    2. S. Yurasov,
    3. A. Schaefer,
    4. J. W. Young,
    5. E. Meffre,
    6. M. C. Nussenzweig
    . 2003. Predominant autoantibody production by early human B cell precursors. Science 301: 1374–1377.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Nitschke L.
    2005. The role of CD22 and other inhibitory co-receptors in B-cell activation. Curr. Opin. Immunol. 17: 290–297.
    OpenUrlCrossRefPubMed
  4. ↵
    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
  5. ↵
    1. Bolland S.,
    2. J. V. Ravetch
    . 2000. Spontaneous autoimmune disease in Fc(gamma)RIIB-deficient mice results from strain-specific epistasis. Immunity 13: 277–285.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Kubo T.,
    2. Y. Uchida,
    3. Y. Watanabe,
    4. M. Abe,
    5. A. Nakamura,
    6. M. Ono,
    7. S. Akira,
    8. T. Takai
    . 2009. Augmented TLR9-induced Btk activation in PIR-B-deficient B-1 cells provokes excessive autoantibody production and autoimmunity. J. Exp. Med. 206: 1971–1982.
    OpenUrlAbstract/FREE Full Text
  7. ↵
    1. Li D. H.,
    2. M. M. Winslow,
    3. T. M. Cao,
    4. A. H. Chen,
    5. C. R. Davis,
    6. E. D. Mellins,
    7. P. J. Utz,
    8. G. R. Crabtree,
    9. J. R. Parnes
    . 2008. Modulation of peripheral B cell tolerance by CD72 in a murine model. Arthritis Rheum. 58: 3192–3204.
    OpenUrlCrossRefPubMed
  8. ↵
    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
  9. ↵
    1. Jellusova J.,
    2. L. Nitschke
    . 2011. Regulation of B cell functions by the sialic acid-binding receptors siglec-G and CD22. Front. Immunol. 2: 96.
    OpenUrlPubMed
  10. ↵
    1. Jellusova J.,
    2. U. Wellmann,
    3. K. Amann,
    4. T. H. Winkler,
    5. L. Nitschke
    . 2010. CD22 × Siglec-G double-deficient mice have massively increased B1 cell numbers and develop systemic autoimmunity. J. Immunol. 184: 3618–3627.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Cariappa A.,
    2. H. Takematsu,
    3. H. Liu,
    4. S. Diaz,
    5. K. Haider,
    6. C. Boboila,
    7. G. Kalloo,
    8. M. Connole,
    9. H. N. Shi,
    10. N. Varki,
    11. et al
    . 2009. B cell antigen receptor signal strength and peripheral B cell development are regulated by a 9-O-acetyl sialic acid esterase. J. Exp. Med. 206: 125–138.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Surolia I.,
    2. S. P. Pirnie,
    3. V. Chellappa,
    4. K. N. Taylor,
    5. A. Cariappa,
    6. J. Moya,
    7. H. Liu,
    8. D. W. Bell,
    9. D. R. Driscoll,
    10. S. Diederichs,
    11. et al
    . 2010. Functionally defective germline variants of sialic acid acetylesterase in autoimmunity. Nature 466: 243–247.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Mary C.,
    2. C. Laporte,
    3. D. Parzy,
    4. M. L. Santiago,
    5. F. Stefani,
    6. F. Lajaunias,
    7. R. M. Parkhouse,
    8. T. L. O’Keefe,
    9. M. S. Neuberger,
    10. S. Izui,
    11. L. Reininger
    . 2000. Dysregulated expression of the Cd22 gene as a result of a short interspersed nucleotide element insertion in Cd22a lupus-prone mice. J. Immunol. 165: 2987–2996.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Pfrengle F.,
    2. M. S. Macauley,
    3. N. Kawasaki,
    4. J. C. Paulson
    . 2013. Copresentation of antigen and ligands of Siglec-G induces B cell tolerance independent of CD22. J. Immunol. 191: 1724–1731.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    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
  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. Jellusova J.,
    2. S. Düber,
    3. E. Gückel,
    4. C. J. Binder,
    5. S. Weiss,
    6. R. Voll,
    7. L. Nitschke
    . 2010. Siglec-G regulates B1 cell survival and selection. J. Immunol. 185: 3277–3284.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Baumgarth N.
    2011. The double life of a B-1 cell: self-reactivity selects for protective effector functions. Nat. Rev. Immunol. 11: 34–46.
    OpenUrlCrossRefPubMed
  19. ↵
    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
  20. ↵
    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
  21. ↵
    1. Campbell I. K.,
    2. J. A. Hamilton,
    3. I. P. Wicks
    . 2000. Collagen-induced arthritis in C57BL/6 (H-2b) mice: new insights into an important disease model of rheumatoid arthritis. Eur. J. Immunol. 30: 1568–1575.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Bäckström N. F.,
    2. U. I. Dahlgren
    . 2008. Induction of experimental arthritis in BALB/c mice by inclusion of a foreign protein in the collagen inoculum. Scand. J. Immunol. 67: 322–328.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Theofilopoulos A. N.,
    2. F. J. Dixon
    . 1985. Murine models of systemic lupus erythematosus. Adv. Immunol. 37: 269–390.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Brownlie R. J.,
    2. K. E. Lawlor,
    3. H. A. Niederer,
    4. A. J. Cutler,
    5. Z. Xiang,
    6. M. R. Clatworthy,
    7. R. A. Floto,
    8. D. R. Greaves,
    9. P. A. Lyons,
    10. K. G. Smith
    . 2008. Distinct cell-specific control of autoimmunity and infection by FcgammaRIIb. J. Exp. Med. 205: 883–895.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Dörner T.,
    2. A. M. Jacobi,
    3. P. E. Lipsky
    . 2009. B cells in autoimmunity. Arthritis Res. Ther. 11: 247.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Bolland S.,
    2. Y. S. Yim,
    3. K. Tus,
    4. E. K. Wakeland,
    5. J. V. Ravetch
    . 2002. Genetic modifiers of systemic lupus erythematosus in FcgammaRIIB(-/-) mice. J. Exp. Med. 195: 1167–1174.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Yuasa T.,
    2. S. Kubo,
    3. T. Yoshino,
    4. A. Ujike,
    5. K. Matsumura,
    6. M. Ono,
    7. J. V. Ravetch,
    8. T. Takai
    . 1999. Deletion of fcgamma receptor IIB renders H-2(b) mice susceptible to collagen-induced arthritis. J. Exp. Med. 189: 187–194.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Boross P.,
    2. V. L. Arandhara,
    3. J. Martin-Ramirez,
    4. M. L. Santiago-Raber,
    5. F. Carlucci,
    6. R. Flierman,
    7. J. van der Kaa,
    8. C. Breukel,
    9. J. W. Claassens,
    10. M. Camps,
    11. et al
    . 2011. The inhibiting Fc receptor for IgG, FcγRIIB, is a modifier of autoimmune susceptibility. J. Immunol. 187: 1304–1313.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. McGaha T. L.,
    2. B. Sorrentino,
    3. J. V. Ravetch
    . 2005. Restoration of tolerance in lupus by targeted inhibitory receptor expression. Science 307: 590–593.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Chen G. Y.,
    2. X. Chen,
    3. S. King,
    4. K. A. Cavassani,
    5. J. Cheng,
    6. X. Zheng,
    7. H. Cao,
    8. H. Yu,
    9. J. Qu,
    10. D. Fang,
    11. et al
    . 2011. Amelioration of sepsis by inhibiting sialidase-mediated disruption of the CD24-SiglecG interaction. Nat. Biotechnol. 29: 428–435.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Chen G. Y.,
    2. J. Tang,
    3. P. Zheng,
    4. Y. Liu
    . 2009. CD24 and Siglec-10 selectively repress tissue damage-induced immune responses. Science 323: 1722–1725.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section