Ly9 (CD229) Cell-Surface Receptor is Crucial for the Development of Spontaneous Autoantibody Production to Nuclear Antigens

The Signaling Lymphocyte Activation Molecule Family (SLAMF) genes, which encode cell-surface receptors that modulate innate and adaptive immune responses, lay within a genomic region of human and mouse chromosome 1 that confers a predisposition for the development of systemic lupus erythematosus (SLE). Herein, we demonstrate that the SLAMF member Ly9 arises as a novel receptor contributing to the reinforcement of tolerance. Specifically, Ly9-deficient mice spontaneously developed features of systemic autoimmunity such as the production of anti-nuclear antibodies (ANA), -dsDNA, and -nucleosome autoantibodies, independently of genetic background [(B6.129) or (BALB/c.129)]. In aged (10- to 12-month-old) Ly9−/− mice key cell subsets implicated in autoimmunity were expanded, e.g., T follicular helper (Tfh) as well as germinal center (GC) B cells. More importantly, in vitro functional experiments showed that Ly9 acts as an inhibitory receptor of IFN-γ producing CD4+ T cells. Taken together, our findings reveal that the Ly9 receptor triggers cell intrinsic safeguarding mechanisms to prevent a breach of tolerance, emerging as a new non-redundant inhibitory cell-surface receptor capable of disabling autoantibody responses.


INTRODUCTION
Systemic lupus erythematosus (SLE) is a multisystem autoimmune disease characterized by major immunological stigmata, the production of antibodies against own cellular nuclear components. The mechanisms underlying SLE are complex and include genetic, epigenetic, environmental, hormonal, and immunoregulatory factors. Furthermore, multiple pathogenic pathways likely contribute to end-organ damage in this heterogeneous disease (1)(2)(3). Elucidation of these pathways, as well as the identification of new molecular disease markers and targets that account for the pathogenesis of lupus, are crucial to improve currently available therapeutic approaches.
Genetic predisposition is a central factor to the development of SLE (4,5). Both Genome Wide Association studies (GWAS) and the in depth analyses of genetically altered and congenic mice have provided insights into the genetic control of lupus. For instance, mouse strains bearing either a NZW-or 129-derived genomic segment that is embedded in chromosome 1 of B6 mice, develop a lupus-like disease due to the epistatic interactions between the 129-derived genes and B6 genes (6). Analysis of congenic NZW × B6 mice identified the Sle1 region of chromosome 1 and its subregions Sle1a, -b, and -c, as key elements involved in breaching B-and T-cell tolerance to chromatin, an essential early step leading to full-blown onset of the disease. Remarkably, multiple gene-wide linkage scans in SLE families have also identified a syngenic lupus susceptibility locus (region 1q23) in human chromosome 1 (7,8).
Precise mapping of the Sle1 region identified a gene interval of 0.9 Mb, termed Sle1b, as the most potent segment involved in the generation of autoantibodies. Interestingly, this segment includes seven signaling lymphocyte activation molecule family (SLAMF) cell-surface receptor genes (9). The proteins encoded by these genes are the best suited candidates for controlling signaling pathways in Sle1 tolerance due to their extensive polymorphic nature and isoform diversity, coupled with their ability to modulate innate and adaptive immunity (10,11). The SLAMF receptors (SLAMF1-9) are involved in the functional regulation of several immune cell types, including helper and cytotoxic T lymphocytes, NK cells, and macrophages (12)(13)(14)(15). These receptors mediate adhesion and regulate cognate T-cell-B-cell interactions, which elicit signal transduction by recruiting SLAM-associated protein (SAP). Importantly, SAP deficiency selectively impairs the ability of CD4 + T cells to stably interact with cognate B cells, avoiding differentiation toward T follicular helper cells (Tfh), and the generation of T-dependent B-cell responses (16)(17)(18). Consistent with this role, the absence of SAP mitigates autoimmunity in various mouse models of lupus by disrupting the generation of germinal centers (GC) (19,20). Increasing evidence suggests that a defective response to self-antigens in the periphery can constitute a relevant pathogenic event. More specifically, autoantibodies detected in lupus patients and lupus-prone mice had suffered isotype switching and somatic hypermutation, which facilitated the binding to self-antigens with high affinity, all indicating the involvement of GC pathways in this disease (21,22). This, therefore, provides a rational for determining not only the exact role played by SLAMF receptors as key drivers of Tfh and GC formation, but also their potential as appealing therapeutic targets for autoantibody-mediated diseases.
While recent reports have identified Slamf6 (Ly108) receptor and its isoforms and Slamf2 (2B4) as contributing to the role played by Sle1b in tolerance (23)(24)(25), the involvement of other SLAMF members cannot be excluded. Of particular interest is the Ly9 (CD229, Slamf3) molecule, since a comparative analysis between B6 and the autoimmune congenic strain B6.Sle1b revealed significant differences in its isoforms usage, as well as in the extent of polymorphisms and expression levels (9), with the evidence indicating the possible participation of Ly9 in B6.Sle1b autoimmunity. Briefly, Ly9 expression is restricted to hematopoietic cells, including B and T lymphocytes (26). As has been shown in other SLAMF members, Ly9 functions as a homophilic adhesion receptor and its cytoplasmic tail contains two copies of the conserved tyrosine-based switch motif (ITSM), which is a docking site for the adapter molecules SAP and EAT-2 (27,28). Ly9-deficient mice with a mixed 129 × B6 background exhibited no major T-cell developmental abnormalities and only very mild defects in T-cell responses (29). Recent findings demonstrate the role of Ly9 as a unique inhibitory cell-surface receptor regulating the size of the thymic innate CD8 + T-cell pool and the development of invariant Natural Killer T (iNKT) cells (30). Nonetheless, the functional role of Ly9 in lupus pathogenesis remains unknown. Here, we use Ly9-deficient mice, which were generated with 129-derived ES cells and then backcrossed onto B6 or BALB/c backgrounds, in order to determine the role of the Ly9 receptor in autoantibody development.

ANTI-NUCLEAR ANTIBODIES ANALYSIS
Anti-nuclear antibodies (ANA) titers were determined by indirect immunofluorescence using permeabilized Hep-2 cells. Serum samples were progressively diluted and incubated for 1 h at room temperature on Hep-2 cells followed by Texas Red-conjugated anti-mouse IgG (Jackson Laboratory, Bar Harbor). After washing, the nucleus was stained with 4 ,6-diamidino-2-phenylindole (DAPI). Analysis was performed by fluorescence detection using a Nikon Eclipse fluorescent microscope (Nikon, Tokyo).

ANTI-DOUBLE-STRANDED DNA AND ANTI-CHROMATIN DETECTION
ELISA assays were performed to quantify levels of anti-doublestranded DNA (anti-dsDNA) and anti-chromatin antibodies in sera of mice. For anti-dsDNA detection, an ELISA was carried out using heat-denatured calf thymus DNA (Sigma Chemical Co., St Louis, MO, USA). dsDNA was coated onto 96-well plates (Corning Costar, Corning, NY, USA) at 10 µg/ml. Purified antibody anti-dsDNA (Clone HpS22, Immunotools, Friesoythe, Germany), used as standard, was serially diluted. Standards and test serums (dilution 1:100) were incubated on plates for 1 h at room temperature. After extensive washing, autoantibodies were detected using a HRP-conjugated anti-mouse IgG (Sigma-Aldrich) and developed with OPD substrate (Sigma-Aldrich).
Anti-chromatin autoantibodies were detected using nucleosome antigen (Arotec Diagnostics Limited, Wellington, New Zealand). The nucleosome antigen was coated on 96-well plates at 3 µg/ml. Serums were diluted 1:100 and incubated for 1 h at room temperature. Autoantibodies against nucleosome were detected using a HRP-conjugated anti-mouse IgG and developed with substrate. All samples were handled simultaneously under the same experimental conditions and results are expressed as OD values.

IgG ISOTYPE DETECTION
Basal serum IgG isotypes were determined by ELISA using purified goat anti-mouse IgG (Sigma-Aldrich) coated 96-well plates. 1:100 diluted mouse serums were incubated for 1 h at room temperature. After extensive washing, IgG isotypes were detected using biotin-conjugated anti-mouse IgG 1 , IgG 2a , IgG 2b , and IgG 3 (Jackson Laboratory). All samples were handled simultaneously under the same experimental conditions and results are expressed as OD values.

KIDNEY HISTOLOGY AND URINE ASSAYS
Kidneys were fixed with 4% formalin (PBS), dehydrated, and embedded in paraffin. All sections were counterstained with Gill's hematoxylin (Panreac, Spain), dehydrated with graded alcohol and xylene, and mounted with DPX (VWR International, Radnor, PA, USA). To evaluate IgG-immunocomplex deposits on kidneys, snap-frozen spleens in OCT media (Sakura Finetek Europe B.V., The Netherlands) were cryosectioned (5 µm), blocked with 6% FCS in PBS, and stained with alexa 488-conjugated anti-mouse IgG (Life Technologies Corporation, Invitrogen, Paisley, UK). Images were analyzed using a Nikon Optiphot-2 microscope (Nikon) and acquired with a high-definition color camera (Nikon).
Freshly voided urine samples were tested for proteinuria using Albustix (Siemens Healthcare Diagnostics Inc. Tarrytown, NY, USA).

MEASUREMENT OF CYTOKINES
The amounts of IFN-γ in sera or in cell-culture supernatants were evaluated with commercially available ELISA kits (R&D Systems). Intracellular analysis of cytokines produced by CD4 + and CD8 + T cells, and iNKT cells was carried out by FACS analysis.

IN VITRO PROLIFERATION AND VIABILITY ASSAYS
Splenocytes were resuspended in PBS with 5% FCS (10 6 cells/mL), and stained with CFSE (Invitrogen) (1 µL CFSE for each 10 6 cells), during 5 min at room temperature. Then cells were washed with cold PBS with 5% FCS and with cold PBS. After the second wash, lymphocytes were resuspended in RPMI media (10% FCS, and 2.5 µM β-mercaptoethanol). Cells were stimulated with 10 µg/mL F(ab ) 2 anti-mouse IgM (Jackson ImmunoResearch), or complete media as a negative control.
In order to measure viability, lymphocytes were stained with fluorochrome-labeled antibodies against surface antigens using standard methods. Cells were washed twice in azide-free and serum/protein-free PBS, and stained with Live/Dead Fixable far Red (Invitrogen) following manufacturer's protocol. Then lymphocytes were stained with eFluor 450-conjugated Annexin V eBioscience during 15 min at room temperature, and washed with RPMI media. Proliferation and viability were assessed 72 h later.

STATISTICS
Mann-Whitney tests were used to calculate p values for all numeric data.

Ly9-DEFICIENT MICE DEVELOP SPONTANEOUS AUTOIMMUNITY
To assess the role of Ly9 in humoral autoimmune responses, Ly9deficient mice were generated by mutated 129-derived ES cells (29) and backcrossed with B6 (C57BL/6) or BALB/c mice for 12 generations. First, we analyzed the presence of ANA in sera of 3, 6, 9, and 12 months old Ly9 −/− (B6.129) mice by immunofluorescence staining of Hep-2 cells. As shown in Figures 1A,B, an ageassociated increase in ANA titers was detected in Ly9 −/− (B6.129) mice that was significantly higher than those of their age-matched Ly9 +/+ (wild-type; wt) counterparts. The differences in ANA serum levels between wt and Ly9 −/− mice were statistically significant as early as 3 months of age. At 12 months of age, 80% of the serum samples from Ly9 −/− (B6.129) showed ANA titers ≥1:4096. In contrast, 100% of the serum titers from their wt counterparts measured ≤1:512. Further analysis demonstrated that by age 12 months Ly9 −/− (B6.129) mice displayed significant increases in their circulating levels of anti-double-stranded (ds) DNA and anti-nucleosome IgG antibodies in comparison to their wt counterparts (Figures 1C,D).
In order to further dissect the role of Ly9 in the humoral autoimmune response, excluding any effects stemming from www.frontiersin.org epistatic interactions, we determined autoantibodies in the serum of Ly9 −/− (BALB/c.129) mice. Once again, an age-associated increase in ANA titers was also detected in Ly9 −/− (BALB/c.129) mice that was significantly higher than those of their agematched wt counterparts. As was observed in the B6 background mice, the differences in ANA serum levels between wt and Ly9 −/− (BALB/c.129) mice were statistically significant as early as 3 months of age, reaching its highest level at age 12 months when 91.6% of Ly9 −/− (BALB/c.129) mice presented ANA titers ≥1:256 compared with 72.7% of their wt counterparts whose titers measured ≤1:64 (Figures 2A,B). Notably, ANA titers in Ly9 −/− (B6.129) mice were always higher than those of Ly9 −/− (BALB/c.129) at the evaluated time points, most likely due to the additional effect of the epistatic interactions, which induced spontaneous loss of immune tolerance to nuclear antigens (25). Comparable levels of anti-dsDNA and anti-nucleosome were detected on Ly9-deficient mice from both genetic backgrounds (Figures 2C,D and  Figures 1C,D), although Ly9 −/− (B6.129) mice displayed much higher ANA titers. This likely reflects the presence of preferential nuclear antigen recognition based on genetic background, which has been also observed in other lupus models (31).
Ly9-deficient mice exhibited a significant increase in IgG2b ( Figure 3A) as well as higher ratios of IgG2a, 2b, and IgG3 isotypes vs. IgG1 than wt mice. Due to the high titers of autoantibodies observed on aged Ly9-deficient mice sera, we further investigated the presence of alterations in these animals' renal physiology and functionality. Twelve-month-old Ly9 −/− (BALB/c.129) mice did not exhibit proteinuria (Figure 3B) or differences in their glomeruli morphology compared to their wt counterparts. Although we observed a mild increase in IgG-immunocomplex deposits in Ly9-deficient mice (Figure 3C), this initial trait prior to any sign of glomerulonephritis did not trigger a renal pathology.
In summary, our findings reveal that the absence of the Ly9 receptor per se initiates the progressive development of autoantibodies, independently of any epistatic interactions.
A deeper analysis of cell subsets by flow cytometry demonstrated altered B-and T-cell peripheral homeostasis in Ly9-deficient mice. We first inspected the B-cell signature of the disease in 12-month-old Ly9 −/− (BALB/c.129) mice (Table 1). Notably, the most significant difference was found in GC B cells, with a percentage of 0.64 ± 0.19 in wt and 1.28 ± 0.30 in Ly9deficient mice (p = 0.0004), which was also evident due the striking expansion of GC cell numbers ( Figure 4C and Table 1). Although a major percentage of GC was observed in Ly9-deficient mice, only a slight decrease in the percentage of Follicular B cells was detected: 88.39 (Ly9 +/+ ) and 8.70 ± 2.97 (Ly9 −/− ) ( Table 1). Thus, the previously mentioned B-cell subsets, most of them involved in autoimmune diseases (34,35), showed a near twofold percentage increase in Ly9 −/− (BALB/c.129) mice compared to their control counterparts.
No altered proportion was found in the studied peritoneal Bcell subsets from the 12-month-old mice (data not shown). Our data also revealed a slightly increased frequency in the occurrence of the most immature B-cell lineages, multipotent progenitor (MMP) and Pro-B cells in the bone marrow of Ly9-deficient mice (data not shown), which is a variation that has been observed in other lupus-prone mice (36).
In examining the T-cell signature of Ly9 deficiency-mediated autoimmunity, the most remarkable difference we observed was in Tfh cells, which showed a threefold percentage increase and cell number expansion (Figure 4D). Importantly, excessive Tfh-cell numbers have been linked to a positive-selection defect in GC, which would account for such differences in the autoantibody generation (21). As previously mentioned, Ly9-deficient mice also exhibited higher ratios of IgG2a,b and IgG3 isotypes vs. IgG1, suggesting that the peripheral tolerance checkpoint that controls GC and Tfh cells becomes altered by the absence of Ly9 molecule ( Figure 3A). In the Ly9-deficient spleen, we found a slight increase in the ratios occurring between CD4 + and CD8 + T cells. We also observed an increase in effector CD4 + and CD8 + T-cell subsets, although it was not significant. Interestingly, the iNKT cell pools in these animals were also enlarged: 3.53 ± 1.23% wt and 7.67 ± 2.80% Ly9 −/− mice ( Table 1).
We conclude that Ly9 gene ablation in a BALB/c background results in the disturbance of B and T cell subsets involved in autoimmunity, with major differences occurring in both Tfh cells and GC B cells.

Ly9 RECEPTOR MODULATES IFN-γ SECRETION BY CD4 + T CELLS
To determine the lymphocyte subsets that facilitate the ignition of SLE-related pathology in the absence of Ly9, we searched for any altered peripheral cell populations in 8-to 12-week-old Ly9 −/− (BALB/c.129) mice. Slight differences were observed in the T-splenic compartment with a small increase in percentage of CD4 + T and iNKT. Notably, most of the alterations were displayed by B-cell subsets, with Transitional T1 B cells presenting the major difference ( Table 2). A deeper examination of T1 and MZ B cells including IgM as a cell marker (T1:CD23 − CD21 − IgM + , MZ: CD23 − CD21 + IgM + ) (37) demonstrates that Ly9-deficient mice possessed approximately a threefold increase of T1 subset and also an expanded MZ B subset ( Figure S1 in Supplementary Material). Based on these results, we conclude that Ly9-deficiency alters the development of B-cell subsets which may be involved in the generation of autoantibodies.
Abnormalities in BCR signaling could shape the splenic B-cell populations and predispose to autoimmune disease. Therefore, we assessed the proliferation and survival of splenic B cell from Ly9 −/− mice after IgM stimulation. The proliferation and apoptotic responses of Ly9-deficient B cells were similar to those observed in the wt B cells, with the exception of a slight decrease in percentage of late apoptotic cells in the Ly9-deficient mice ( Figure 5).
Since earlier studies have demonstrate the critical implication of SLAMF receptors in Th1/Th2/Th17 polarization (14,26,38), we hypothesized that Ly9-deficient splenocytes in 8-to 12-weekold mice would foster an alteration in T-cell cytokine production prior to disease onset, thereby enabling autoantibody production at older ages. First, we asked whether the ablation of murine Ly9 could modulate IL-17 secretion by CD4 + T cells under Th17 polarizing conditions, since earlier reports have implicated the human Ly9 receptor in IL-17 T-cell secretion (38,39). No significant differences were detected in the percentage of IL-17 secreting CD4 + T cells (Figure 6A). On the other hand, our group has previously shown that monoclonal antibodies against Ly9 negatively regulate TCR signaling, thereby inhibiting ERK phosphorylation and IFN-γ secretion (26,40). Herein, we investigated the role of Ly9 in IFN-γ modulation by activating splenic T cells with anti-CD3 and anti-CD28. A significant increase in IFN-γ secretion, as well as an increase in the expression of CD40L in CD4 + T cells was detected in the absence of Ly9 compared with wt cells (Figures 6B,C). A similar result was obtained when evaluating the expression of the CD25 activation marker (data not shown). In addition, PMA/ionomycin activation of splenic cells revealed that Ly9-deficient mice foster CD4 + T, CD8 + T, and iNKT cells capable of secreting major quantities of IFN-γ prior to the development of autoimmunity ( Figure 6D). Furthermore, isolated Ly9-deficient CD4 + T cells consistently showed an increased IFN-γ production after PMA/ionomycin activation as compared with wt mice (Figure 6E). In contrast, no significant difference in percentage of IL-4 producing CD4 + T cells was observed (data not shown). In accordance with these results, 12-month-old Ly9 −/− (BALB/c.129) mice exhibited an increased percentage of IFN-γ producing CD4 + T cells, which correlated with the high ANA titers detected in serum ( Figure 6F). Nevertheless, we could not detect IFN-γ in the serum of these mice (data not shown). Thus, in the absence of the Ly9 molecule occurs an increase in peripheral T and iNKT IFN-γ secretion, a key cytokine in the pathogenesis of SLE which has been previously shown to trigger the accumulation of pathogenic Tfh and GC B cells (41).

DISCUSSION
Despite extensive research, the mechanisms by which susceptibility and effector genes initiate and promote autoantibody production and tissue damage remain poorly defined. In the present study, we have shown that Ly9 deficiency results in a spontaneous loss of tolerance, reflected principally in the development of autoantibodies, a process which is thought to underlie the initiation of SLE syndrome.
We first found that the absence of the Ly9 gene in mice with a B6 and BALB/c background resulted in the development of autoantibodies. The loss of tolerance to nuclear antigens detected in Ly9 −/− (B6.129) mice was reminiscent of those previously observed in B6.Sle1b and B6.129Chr1b congenic mouse strains (6,42). We found that Ly9 −/− (B6.129) mice developed autoantibodies as early as 12 weeks of age. In fact, by 9 months of age all mice were autoantibody positive, a pattern similar to that found in SLAMF receptor Slamf1-deficient (B6.129) mice (25). Most SLAMF receptor-deficient mice have been generated by altering those SLAMF genes located in chromosome 1 via the homologous recombination of 129-derived embryogenic stem cells (25,29,43,44). Once the resultant mice are backcrossed onto B6 mice [e.g., Slamf1 −/− (B6.129)], the deficient mice are affected by the epistatic interaction that occurs between the 129 gene segment and the B6 genome. Thus, epistatic interactions between 129-derived and B6 genes in Ly9 −/− (B6.129) mice greatly contribute to the autoantibody response. Exceptionally, the disruption of the Ly108 gene in congenic mice [Ly108 −/− (B6.129)] mitigates the generation of autoantibodies, indicating that SLAMF members carry out opposing functions (e.g., Ly108 vs. Ly9).

Parental Subset Marker
Ly9 We next assessed the role of Ly9 in autoimmune disease without the confounding influence of mixed haplotypes by utilizing Ly9 −/− (BALB/c.129) mice. Although autoantibody titers were lower compared to Ly9 −/− (B6.129) mice, this strain clearly developed an autoimmune response based on the significant increases of ANA, anti-dsDNA, and anti-chromatin antibodies compared to their wt counterparts. By contrast, other SLAMF receptor-deficiencies embedded in the BALB/c genome, such as Slamf1 −/− (BALB/c.129) and Slamf2 −/− (BALB/c.129), do not develop any autoimmune response (25), underscoring the role played by Ly9 as a negative regulator in the pathogenesis of lupus. Consequently, among the various SLAMF receptors, both Ly9 and 2B4 rise as unique factors contributing to the reinforcement of tolerance (24).
As a consequence of breaching tolerance, a plethora of disorders can develop a posteriori; e.g., T and B cells are reportedly involved in the amplification and perpetuation of the autoimmune response, resulting in inflammation and cytokine dysregulation (45). This proved to be the case in our 12-month-old Ly9-deficient mice in which various B and T cells subsets underwent alterations. Notably, the most noteworthy differences in older mice were found in Tfh and GC B cells, two cell types which have been shown to preferentially express the Ly9 receptor (46).
In order to begin to understand why Ly9-deficient mice developed spontaneous autoimmunity, we search for any abnormal peripheral B and T cell development prior to full autoantibody disorder on 8-to 12-week-old Ly9-deficient mice. We observed alterations in the B-cell splenic compartment, with the most prominent expansion displayed by Ly9-deficient transitional T1 B cells. Interestingly, SLE patients present increased numbers of T1 cells, although their role in lupus is still ill defined (35). In addition, 8-to 12-week-old Ly9-deficient mice displayed an increase in MZ B cells. Even though many autoreactive antibodies appear to be the product of GC reactions, major evidence begins to reveal that MZ B cells play a key role in homeostasis and tolerance. Notably, the MZ B-cell expansion has been directly implicated in lupus pathogenesis in some murine models (47-50), but not others (51,52). Autoimmunity mediated by B cell is usually linked to a Bcell hyperresponse which change the B-cell splenic composition. Thus, an increased BCR signal, as shown by mice deficiencies in SHP-1, FcγRII, CD22, Cbl-b, or overexpressing CD19, leads to a B-cell hyperresponsiveness upon BCR stimulation, change B-cell subsets normal architecture, and culminates in a systemic autoimmune disease (53). Here, we show that there's similar proliferation and survival of Ly9-deficient B cells as compared with wt cells, although we can not exclude the abnormal function of particular B-cell subsets in the Ly9-deficient mice which could be the subject of future investigations.
Next, we investigated how the absence of Ly9 could possibly contribute to a cytokine imbalance prior to disease onset. Others have shown that the engagement of human naive CD4 + T lymphocytes with an anti-human Ly9 monoclonal antibody Frontiers in Immunology | Inflammation under Th17 polarizing conditions results in an increase in IL-17 (38,39). No alterations in IL-17 producing T lymphocytes were observed once Ly9-deficient mice splenic cells were activated under Th17 polarizing conditions. Previous reports showed a diminishing IL-4 production and no altered IFN-γ secretion by Ly9-deficient CD4 T cells (29). In contrast, in the absence of Ly9, we observed an increase in IFN-γ producing T cells with no significant alteration in IL-4 secretion. Importantly, IFN-γ has long been associated with lupus (54). This apparent contradiction could be explained by the influence of epistatic interactions as well as the mice background, since previous studies were performed in Ly9 −/− with a mixed background (B6 × 129) mice. Recent reports have demonstrated that the overproduction of IFN-γ induces an aberrant accumulation of Tfh and GC cells (41).We also found that these subsets underwent expansion in Ly9-deficient mice, when the influence of confounding epistatic interactions was absent. These observations suggest that the Ly9 molecule may play an inhibitory role in the expansion of these subsets. In addition, Ly9-deficient mice showed a skewed isotype switching toward IgG2a/b, an isotype induced by a Th1 response that requires T cell-stimulated B lymphocytes. Although IgG2a antibodies are the most pathogenic class of immunoglobulin (55, www.frontiersin.org 56), we did not observe any major kidney pathology. However, a mild increase in IgG-immunocomplex deposits was detected in Ly9-deficient mice, a stage that often precedes full-blown kidney pathology. Thus, an Ly9 mutation confers a predisposition to autoantibody generation, but one that may require additional factors (e.g., a viral infection or additional gene alterations) before a full lupus disease pathogenesis becomes evident. In addition, we cannot exclude the influence of Ly9-deficient iNKT cells, a population that underwent expansion in 8-to 12-week-old Ly9 −/− mice (30), on the observed autoimmune response. Although iNKT cells generally play a tolerogenic role in autoimmunity, some models, and depending on the mouse strain and treatment protocol used, iNKT cell activation exacerbates rather than protects against autoimmunity (57), e.g., the stimulation of iNKT cells triggers a Th1 immune response in adult NZB/W mice, resulting in the exacerbation of systemic autoimmunity (58). Thus, iNKT disease protection is associated with enhanced Th2 and/or reduced Th1 responses against targeted antigens. A deeper study of Ly9deficient iNKT cell will be needed to determine the specific role of this cell-type in the observed autoimmune phenotype.
In summary, the data presented in this study sheds light on the inhibitory function of the Ly9 cell-surface receptor, suggesting that this molecule is involved in the maintenance of peripheral cell tolerance by serving as a negative regulator of the immune response. Further investigation will be required to elucidate the precise mechanism by which Ly9 confers protection from autoimmunity. Nonetheless, this study demonstrates that Ly9 may not only prove to be a valuable target for the treatment of autoimmunity, but may also help build a more rational road map toward understanding the molecular causes of SLE syndrome.

ACKNOWLEDGMENTS
Jose de Salort performed most of the experiments and analyzed the data; Marta Cuenca performed the B-cell functional experiments; Pablo Engel and Xavier Romero designed the experiments and supervised the study; Jose de Salort, Pablo Engel, and Xavier Romero wrote the manuscript; Cox Terhorst provided input into the conceptual development of the experiments and edited the manuscript. We thank Adriana Lázaro for technical assistance. We also thank the NIH Tetramer Core Facility for providing R-PE labeled murine CD1d tetramer pre-loaded with PBS57. This work was supported by the Ministerio de Educación y Ciencia through Grants SAF 2009-07071/SAF 2012-39536 (to Pablo Engel) and by a grant from the NIH (2P01AI65687-06A1 to Pablo Engel and Cox Terhorst). Jose de Salort was supported by Ayuda Personal Investigador en Formación (APIF), a grant from the Universitat de Barcelona. Marta Cuenca was supported by Ministerio de Educación, Cultura y Deporte (AP2010-1754). Xavier Romero was supported by Beatriu de Pinos (2010 BP-B).