Cd1d regulates B cell development but not B cell accumulation and IL10 production in mice with pathologic CD5(+) B cell expansion

Abstract

Background: CD1d is a widely expressed lipid antigen presenting molecule required for CD1d-restricted invariant natural killer T (iNKT) cell development. Elevated CD1d expression is detected in CD5(+) IL10-producing B cells, called B10 B cells, and is correlated with poorer prognosis in chronic lymphocytic leukemia (CLL), a CD5(+) B cell malignancy with B10-like functional properties. Whether CD1d expression regulates CD5(+) B cell accumulation, IL10 competence, and antibody production in naïve mice with pathologic CD5(+) B cell expansion remains untested.

Results: Using three different transgenic mouse models of benign or leukemic CD5(+) B cell expansion, we found that CD1d was differentially expressed on CD5(+) B cells between the three models, but loss of CD1d expression had no effect on CD5(+) B cell abundance or inducible IL10 expression in any of the models. Interestingly, in the CLL-prone Eμ-TCL1 model, loss of CD1d expression suppressed spontaneous IgG (but not IgM) production, whereas in the dnRAG1xEμ-TCL1 (DTG) model of accelerated CLL, loss of CD1d expression was associated with elevated numbers of splenic CD4(+) and CD8(+) T cells and an inverted CD4(+):CD8(+) T cell ratio. Unexpectedly, before leukemia onset, all three transgenic CD1d-deficient mouse strains had fewer splenic transitional B cells than their CD1d-proficient counterparts.

Conclusions: The results show that CD1d expression and iNKT cells are dispensable for the development, accumulation, or IL10 competence of CD5(+) B cells in mice prone to benign or leukemic CLL-like B cell expansion, but reveal a novel role for iNKT cells in supporting B cell progression through the transitional stage of development in these animals. These results suggest CD1d-directed therapies to target CLL could be evaded by downregulating CD1d expression with little effect on continued leukemic CD5(+) B cell survival. The data also imply that iNKT cells help restrain pro-leukemic CD8(+) T cell expansion in CLL, potentially explaining a reported correlation in human CLL between disease progression, the loss of NKT cells, and a paradoxical increase in CD8(+) T cells.

Fig. 1

CD5⁺ B cells in dnRAG1 and DTG mice express high levels of CD1d. a-b Cohorts of 12 week-old (a) and 36 week-old (b) mice were analyzed for CD1d expression on CD19⁺ and CD5⁻ or CD5⁺ B cells from WT, dnRAG1, Eμ-TCL1 and DTG mice. (Left Panel) Overlay of CD1d expression on CD19⁺CD5⁻ (WT) or CD19⁺CD5⁺ (WT, dnRAG1, Eμ-TCL1 and DTG) B cells from representative animals. (Right panel) Mean fluorescence intensity of CD1d staining in the gated populations in the left panel (n = 4–6 animals/genotype). Statistically significant differences (p < 0.05) were only detected in comparisons between dnRAG1 mice and the other mouse genotypes (a, vs WT mice; b, vs Eμ-TCL1 mice; c, vs DTG mice)

Fig. 2

CD5⁺ B cells in 12 week-old dnRAG1 and DTG mice support inducible IL10 expression in vitro, but IL10 competence is CD1d-independent. a-b Splenocytes from 12 week-old wild-type, dnRAG1, Eμ-TCL1, and DTG mice on a Cd1d ^+/+ (a) or Cd1d ^del/del (b) background were incubated with monensin alone or LPS + PIM for 4 h. After permeabilization, the cells were stained with IL-10-specific or isotype-control monoclonal antibody (Ab) as indicated. Gated CD19⁺ B cells were analyzed for CD5 and IL10 expression. The percentage of cells in each quadrant is shown for representative animals. c The mean percentage of total CD5⁺ B cells staining positive for IL10 for n = 4 mice/genotype is plotted in bar graph format. Error bars represent the SEM. Statistically significant (p < 0.05) differences between groups of WT and transgenic animals with the same Cd1d genotype are indicated (a, vs dnRAG1; b, vs Eμ-TCL1; c, vs DTG). No significant differences between Cd1d ^+/+ and Cd1d ^del/del animals with the same transgenes were detected

Fig. 3

Eμ-TCL1 mice at 36 weeks of age show increased inducible IL10 expression in vitro compared to 12 weeks, but IL10 competence remains CD1d-independent. (a-c) Splenocytes from 36 week-old wild-type, dnRAG1, Eμ-TCL1, and DTG mice on a Cd1d ^+/+ or Cd1d ^del/del background were analyzed as Fig. 2. No significant differences between Cd1d ^+/+ and Cd1d ^del/del animals with the same transgenes were detected

Fig. 4

Progression through the transitional stages of B cell development is impaired in Cd1d ^del/del mice. a Gated splenic CD19⁺CD5⁻ B cells from 12 week-old or 36 week-old WT, dnRAG1, Eμ-TCL1 and DTG mice on a Cd1d ^+/+ (+/+) or Cd1d ^del/del (del/del) background were initially segregated into transitional (trans.) and mature B cell subsets based on differential expression of AA4.1 and B220. Transitional (AA4.1⁺B220^hi) and mature (AA4.1⁻B220^hi) B cell subsets were further separated into T1-T3, marginal zone (MZ) and follicular mature (FM) populations based on differential expression of IgM, CD21, and CD23 using the gating strategy found at the bottom of the figure. The percentage of cells within the identified gates is shown for representative animals. Data collected from n = 4–6 animals/genotype were used to calculate the absolute number of cells in each population, and are summarized in Additional file 2: Table S1 and Additional file 4: Table S2. b Values for the mean absolute number of total transitional B cells (B220^hiAA4.1⁺CD5⁻) and for each transitional B cell subset (T1-T3) are plotted in bar graph format for the 12 week time point in (a). Error bars represent the standard error of the mean. Statistically significant differences (p < 0.05) between groups of Cd1d ^+/+ or Cd1d ^del/del mice with the same transgenic background are identified by an asterisk

Fig. 5

Loss of CD1d expression in 36 week-old DTG mice is associated with expansion of splenic T cells, particularly CD8⁺ T cells. a Gated splenic CD3⁺B220⁻ lymphocytes were analyzed for CD4 and CD8 expression. The percentage of cells within the identified gates or quadrants is shown for representative animals. Note that in DTG mice, the B220⁻ gate does not exclude all CD5⁺ B cells, which are highly abundant and express variably low levels of B220 in these animals. Data collected from n = 4–6 animals/genotype were used to calculate the absolute number of cells in each population, and are summarized in Additional file 4: Table S2. b Values for the mean absolute number of CD4⁺ and CD8⁺ T cells determined from (a) are plotted in bar graph format. Error bars represent the standard error of the mean. Statistically significant differences (p < 0.05) between groups of Cd1d ^+/+ or Cd1d ^del/del mice with the same transgenic background are identified by an asterisk

Fig. 6

Loss of CD1d expression does not significantly affect spontaneous antibody production in dnRAG1 and DTG mice. Serum IgM and IgG was measured from 12 (a) and 36 (b) week-old WT, dnRAG1, Eμ-TCL1, and DTG mice on a Cd1d ^+/+ or Cd1d ^del/del background (closed circles or open triangles, respectively). Individual values and means (identified by horizontal bar) are shown for each group; statistically significant differences between groups are indicated