2B4 Is Dispensable for T-Dependent B Cell Immune Responses, but Its Deficiency Leads to Enhanced T-Independent Responses Due to an Increase in Peritoneal Cavity B1b Cells

The signaling lymphocyte activation molecule (SLAM) family plays important roles in adaptive immune responses. Herein, we evaluated whether the SLAM family member 2B4 (CD244) plays a role in immune cell development, homeostasis and antibody responses. We found that the splenic cellularity in Cd244 -/- mice was significantly reduced due to a reduction in both CD4 T cells and follicular (Fo) B cells; whereas, the number of peritoneal cavity B cells was increased. These findings led us to examine whether 2B4 modulates B cell immune responses. When we examined T-dependent B cell responses, while there was no difference in the kinetics or magnitude of the antigen-specific IgM and IgG1 antibody response there was a reduction in bone marrow (BM) memory, but not plasma cells in Cd244 -/- mice. When we evaluated T-independent immune responses, we found that antigen-specific IgM and IgG3 were elevated in the serum following immunization. These data indicate that 2B4 dampens T-independent B cell responses due to a reduction in peritoneal cavity B cells, but has minimal impact on T-dependent B cell responses.


Introduction
2B4 is a member of the signaling lymphocyte activation molecule (SLAM)-related receptor family and is also known as SLAMF4 and CD244 [1]. All members of the SLAM family share a similar structure, including an extracellular domain, a transmembrane region, and a tyrosine rich cytoplasmic region [1]. Unlike most SLAM family members, 2B4 does not bind via hemophilic interactions, but binds to CD48, which is broadly expressed by hematopoietic cells and functions as an adhesion and co-stimulatory receptor for both B and T cells [2]. By means of their immunoreceptor tyrosine-based switch motifs (ITSM) in the cytoplasmic domain, SLAM family receptors signal by interacting with members of the SLAM-associated protein (SAP) (SH2D1A) family of adaptors [1]. The SAP adaptors couple SLAM proteins to biochemical signaling pathways mediating the various biological functions of the SLAM family [1,3].
2B4 expression by B cells has been best studied in humans where its expression by all B cell subsets was reported to be very low to absent as compared to other SLAM family members [4]. However, upon transformation with Epstein-Barr virus, 2B4 expression was induced with up to 79% of blasts staining positive [5]. 2B4 expression was also upregulated by pokeweed mitogen with 5-38% of B cell blasts positive [5]. Interactions between CD48 and 2B4 can lead to signaling through both receptors [2,6]. CD48 signaling in B cells leads to homotypic adhesion, proliferation and/or differentiation, release of inflammatory effector molecules and isotype class switching [2,7,8]. In addition, all of these processes are also elicited in T cells via CD48 ligation with the addition of promoting their activation and/or cytotoxicity [2]. 2B4 signaling requires SAP or EWS-activated transcript 2 (EAT-2; also called SH2D1B) [6,[9][10][11]. In CD8 T cells and NK cells 2B4 has been reported to exert both positive and negative regulation [9][10][11]. A specific role for 2B4 in B cells has not been reported.
Here we investigated the role of 2B4 in B cells and found that Cd244 -/mice have a significant reduction in splenic cellularity that was due to a reduction in CD4 T and follicular (Fo) B cells. We also found that peritoneal cavity B cells were increased in Cd244 -/mice due to a significant increase in B1b and B2, but not B1a cells. When we examined 2B4 expression, we found that B cell subsets expressed no to very low levels of 2B4. Following a T-dependent immune response, there was no difference in the kinetics and the magnitude of the antigenspecific IgM and IgG 1 response between WT and Cd244 -/mice. However, late in the response there was a significant decrease in the number of bone marrow (BM) memory B cells in Cd244 -/mice. Following immunization with a T-independent antigen, Cd244 -/mice exhibited a significant increase in antigen-specific IgM production on day 14 and isotype-class switched IgG 3 on days seven and 14. These data indicate that even though a global deficiency in 2B4 is associated with reduced numbers of Fo and BM memory B cells it has minimal impact on Tdependent B cell responses. In contrast, the increase in peritoneal cavity B cells in Cd244 -/mice is directly correlated to an increase in the T-independent immune response.

Ethics statement
All animal protocols used were approved by the Medical College of Wisconsin's Institutional Animal Care and Use Committee. We monitored immunized animals for adverse health issues and used appropriate methods of euthanasia including isoflurane or CO 2 followed by cervical dislocation.

Mice and reagents
C57BL/6 (WT) mice were purchased from The Jackson Laboratories (Bar Harbor, ME).

Cell isolation and flow cytometry
Single cell suspensions were prepared from spleens and 1 x 10 6 cells were stained with specific antibodies. Peritoneal cavity B cells were isolated as described [13]. For 2B4 expression, anti-CD244 staining was followed by anti-IgG 2a -biotin and then Strepavidin-PE-Cy5.5 to amplify the signal. Samples were acquired with a LSR II flow cytometer (Becton Dickinson, San Jose, CA) and data were analyzed using FlowJo software (Tree Star, Ashland, OR).

Statistical analysis
WT and Cd244 -/groups were compared using a two-tailed unpaired t-test. A p value < 0.05 was considered significant.

Cd244 -/mice have reduced numbers of CD4 T cells and Fo B cells
In our mouse colony, we found that Cd244 -/mice had a significant reduction in splenic cellularity ( Fig 1A). This finding was not reported in the original study describing the Cd244 -/mice [12]. In this same study, it was reported that a global deficiency in 2B4 did not alter the frequency (percentage) of CD3 + , CD4 + , CD8 + , CD11b + , CD19 + nor NK1.1 + cells in the spleen [12]. However, the percentage does not reflect changes in absolute cell number. Thus, we repeated the splenic phenotyping and calculated absolute numbers of cells to determine the specific immune cell populations that are reduced in Cd244 -/mice in our colony. We found no difference in the absolute number of CD11b + myeloid cells (Fig 1B), CD11c + dendritic cells (Fig 1C) or NK1.1 + NK cells (Fig 1D) between WT and Cd244 -/mice. In contrast, there was a significant reduction in αβ T cells in the spleen (Fig 1E). Further analysis revealed that the reduction was due to lower numbers of CD4 + T cells (Fig 1F). In addition, while a trend for lower numbers of CD8 + T cells (p = 0.1) (Fig 1G) and CD4 + Foxp3 + T regulatory cells (p = 0.11) (Fig 1H) was observed in Cd244 -/mice, it did not reach significance. When we examined the percentage of the above immune cell populations in the spleen, the non-significant increase in CD11b + myeloid cells (p = 0.06) was reflected in a significant increase in percentage of total CD11b + cells ( Fig 1I) and percentage of the CD11b + dendritic ( Fig 1J) and NK cells (Fig 1K) subsets. Interestingly, even though significance was observed for absolute number there was no difference in the percentage of αβ T cells (Fig 1L) or CD4 T cells (Fig 1M). In addition, there was no difference in the percentage of CD8 T cells (Fig 1N) or in Treg (Fig 1O). It is not clear why we observed an increase in the percentage of myeloid cells that was not seen in the original report describing Cd244 -/mice, but likely reflects differences in composition of the microbiota in the two animal colonies. Nevertheless, our findings that significant differences in percentages is not reflected by the absolute cell number highlights why only analyzing changes in the percentages of specific cell subsets can be misleading.
We determined whether the absence of 2B4 altered B cell differentiation in the spleen using a gating strategy to identify the Fo and MZ splenic mature B cells populations as well as their transitional (T) 1, T2 and T3 precursors (Fig 2A) [16]. We also analyzed the T2-MZ precursor (P), which is the immediate precursor to the MZ lineage (Fig 2A) [17]. When we quantitated total B cells, we found that the reduction in splenic cellularity in Cd244 -/mice was also due to a significant reduction in the number of B220 + B cells (Fig 2B). By quantitating the mature splenic B cell subsets we found that Fo (Fig 2C), but not MZ (Fig 2D), B cell numbers were significantly reduced. We next asked whether the reduction in Fo B cells was due to altered differentiation into the Fo lineage and found no difference in the transitional (T) 1 (Fig 2E) or T2 (Fig 2F) precursor populations between WT and Cd244 -/mice. We also found no difference in the T3 (Fig 2G) or T2-MZP (Fig 2F) subsets. These data indicate that the homeostasis of Fo B cells is uniquely altered by a global deletion in 2B4.
Collectively, our data demonstrate that neither development nor homeostasis of the myeloid lineage are regulated by 2B4. The data also demonstrate that even though 2B4 is highly expressed by NK cells [12], it also does not regulate their development or homeostasis. To our surprise, the reduced cellularity in the spleen of Cd244 -/mice was due to a reduction in both CD4 + T cells and Fo B cells. These data were unexpected because neither cell population is known to highly express 2B4 [4,18,19]. Thus their reduction in the Cd244 -/mouse is not likely to be cell intrinsic. Since 2B4 is the ligand for CD48, which is expressed by lymphocytes, it is more likely that a 2B4-expressing cell is interacting with and promoting CD4 and Fo B cell proliferation or survival via CD48. In this regard, ligation of CD48 on human B cells has been shown to provide an accessory signal for CD40 signaling including increasing B cell proliferation [20]. In addition, CD48 was also shown to augment both IgG and IgM production [20,21]. For T cells, CD3 signaling in combination with CD48 ligation enhanced the phosphorylation level of the TCR z chain [22] and was found to enhance T cell proliferation in response to antigen and anti-CD3 [23,24]. While the CD48 + cell involved in maintenance of CD4 and Fo B cell numbers is not known, it is possible that NK or iNK cells, which also express 2B4, play a role [2,[25][26][27].

Splenic Fo B cells lack 2B4 expression, while MZ B cells express very low levels
We next determined whether splenic B cell subsets express 2B4 in the steady state by flow cytomtery. Because the lack of reported 2B4 expression by B cells could be due to a low . The absolute number of B cells (B220 + ) (B) and that of the Fo (B220 + IgM int CD21 int CD23 + CD93 -) (C), MZ (B220 + IgM hi CD21 hi CD23 -) (D), T1 (B220 + IgM hi CD21 lo CD23 -CD93 + ) (E), T2 (B220 + IgM hi CD21 lo CD23 + CD93 + ) (F), T3 (B220 + IgM int CD21 int CD23 + CD93 + ) (G) and T2-MZP (B220 + IgM hi CD21 hi CD23 + ) (H) B cell subsets in the spleen of WT and Cd244 -/mice was determined by flow cytometry. Data are shown as the mean ± SE of four mice. *p<0.05; **p<0.01. expression level, we used a three step staining procedure to amplify the 2B4 signal. In addition, we performed the same staining in Cd244 -/mice to confirm specificity. As a positive control, we analyzed 2B4 expression on NK cells, which expressed 2B4 at a high level (Fig 3). 2B4 cell surface expression by flow cytometry was not evident on either the total B220 + or Fo subset (Fig 3). However, we found that MZ B cells express 2B4 at a very low level (Fig 3).

2B4 expression is dispensable for T-independent immune responses
It was previously shown that SAP plays an essential role in T cell:B cell cognate interactions in the germinal center [28], indicating that SLAM molecules are involved in antibody production to T-dependent antigens. While a role for SLAM (CD150), was implicated in germinal center reactions [28], it is possible that other SLAM family members also play a regulatory role. To determine whether 2B4 regulates T-dependent antibody production, we immunized with NP-CGG and quantitated serum antigen-specific IgM and IgG 1 on days 0, 7, 10, 14, 28 and 42. While there was a trend for WT B cells to produce higher levels of both IgM (Fig 4A) and IgG 1 (Fig 4B), there was no significant difference as compared to Cd244 -/mice at any of the time points examined. When we quantitated memory B cells, we found no alteration in their number in the spleen on day 7 (data not shown) or on days 11 and 42 following immunization (Fig 4C). In contrast, BM memory cells were reduced on day 42 in Cd244 -/mice (Fig 4C). Consistent with similar levels of antigen-specific Ig, we found no difference in the number of plasma cells or germinal center B cells between WT and Cd244 -/mice at the timepoints examined in the spleen and BM (Fig 4C). Thus even though Fo B cells are reduced in Cd244 -/mice this only led to a minimal impact on the magnitude of the T-dependent antibody response. In addition, we did not observe 2B4 expression by splenic memory or germinal center B cells using flow cytometry on day 11 post-NP-CGG immunization (Fig 4D). The reduction in BM memory B cells is an interesting finding that could reflect a role for 2B4 in their migration from the spleen. In addition, 2B4 could be important in memory B cell maintenance specifically in the BM niche.

Cd244 -/mice have increased numbers of peritoneal B1b cells and Tindependent B cell responses
Because T-independent immune responses were not altered due to a deficiency in 2B4, we next asked whether its loss impacted T-independent B cell responses. Because both MZ and peritoneal cavity B cells respond to T-independent antigens [17,29], we first used a gating strategy to distinguish peritoneal cavity B cell subsets (Fig 5A) [14]. We found that total peritoneal cavity B cells (B220 + ) were increased in Cd244 -/mice as compared to WT (Fig 5B). This was due to a significant increase in both the conventional B2 and the B1 subsets (Fig 5B). When B1 cells were further differentiated, we found that while the B1b subset was significantly increased in Cd244 -/mice the reciprocal was true for B1a cells, which were significantly decreased (Fig 5B).
Given the specific role for B1b cells in T-independent immune responses [30,31], we next asked whether their increase in Cd244 -/mice would augment the B cell response to the T-independent antigen NP-ficoll. In T-independent immune responses, ligation of the BCR with a repeating antigen such as NP-ficoll induces BCR signaling that is sufficient to drive the production of antigen-specific Ig without the requirement for T cell help in the germinal center [32]. Following immunization with NP-ficoll, we measured serum NP-specific Ig on days 0, 7 and 14. As expected, levels of NP-specific IgM peaked on day 7 in WT mice that had begun to wane on day 14 (Fig 3A). In Cd244 -/mice, levels of NP-specific IgM were sustained on day 14 resulting in a significant increase in their serum levels ( Fig 6A). Because T-independent immune responses lead to isotype class switching to IgG 3 , we also measured NP-specific IgG 3 and found that it was significantly elevated on both day 7 and 14 in Cd244 -/mice (Fig 3B). Given that very little isotype-class switched IgG 3 is generated following immunization with T-independent antigens the O.D. values are low. Even though the error bars are relatively small for the IgG 3 data (Fig 6B), caution needs to be used to avoid overinterpretation of the data.
These cumulative data indicate that 2B4 expression by B cells is not modulating their ability to respond to T-independent antigens and does not regulate isotype class switching. However, 2B4 may play a role in restraining IgM and IgG 3 production. Given that 2B4 can both positively and negatively regulate NK cell function depending upon whether SAP or EAT-2 are utilized, respectively, [3,10], it is possible that 2B4 could also negatively regulate B cell effector functions. Given the low level of 2B4 expression by B cells, it is more likely that it regulates T- The T-dependent B cell response is unaltered in Cd244 -/mice. WT (closed symbol) and Cd244 -/-(open symbol) mice were immunized (i.p.) with the T-dependent antigen NP-CGG (30 μg) that was alum precipitated. Blood was collected on days 0, 7, 10, 14, 28 and 42, and NP-specific IgM (A) and IgG 1 (B) titers were determined by ELISA. The O.D. value for NP-specific IgM at the 1:800 dilution and NP-specific IgG 1 at the 1:20,000 dilution are shown. Data shown are the mean ± SE of 8 mice for days 0 and 7; 7 mice for days 10, 14 and 28; and 4 mice for day 41. (C) The absolute number of memory (B220 + IgD -NP(24) + CD38 + ), plasma cells (B220 + IgD -IgM -IgG + NP(24) + CD138 + ) and germinal center (B220 + GL-7 + ) B cells were determined in the spleen on days 11 (top panels) and 42 (middle panels) and in the BM on day 42 by flow cytometry. Data are shown as the mean ± SE of four mice. (D) Expression of 2B4 was evaluated on memory and germinal center B cell from WT and Cd244 -/mice 10 days after immunization. Histograms display overlaid staining profiles from the biotin negative control (shaded), WT mice (thin line) and Cd244 -/mice (thick line). Data shown are from one representative mouse of three. *p<0.05. independent B cell responses in a cell extrinsic manner. However, we favor the hypothesis that the increase in antigen-specific Ig in Cd244 -/mice following immunization with a T-independent antigen is due to the increase in B1b cells, which are known to specifically be involved in such B cell immune responses [30,31]. Whether 2B4 alters the development or homeostasis of B1b cells is not known.
In summary, we carefully phenotyped Cd244 -/mice for immune cell populations in the spleen and discovered that the reduced cellularity was due to a reduction in both CD4 T cells and Fo B cells. In contrast, increased numbers of peritoneal cavity B cells were present in Cd244 -/mice due to an increase in B2 and B1b cells. While 2B4 is an important regulator of NK cell function, when globally deficient, our study shows that it has minimal impact on Tdependent B cell immune responses. In contrast, the increase in B1b cells in Cd244 -/mice correlates with increased production of antigen-specific Ig in response to a T-independent antigen. The data also indicate that while both NK and iNK cells have been implicated in antibody production and isotype class switching [25,26], this function is not likely mediated by 2B4:CD48 interactions. The absolute number of B cells (B220 + ) and that of the B2 (B220 + CD23 + ), B1 (B220 + CD23 -), B1b (B220 + CD23 -CD11b + CD5 -) and B1a (B220 + CD23 -CD11b + CD5 + ) subsets in the peritoneal cavity of WT and Cd244 -/mice was determined by flow cytometry. Data are shown as the mean ± SE of four mice. *p<0.05. doi:10.1371/journal.pone.0137314.g005