FcRγ controls the fas-dependent regulatory function of lymphoproliferative double negative T cells

Abstract

Patients with autoimmune lymphoproliferative syndrome (ALPS) and lymphoproliferation (LPR) mice are deficient in Fas, and accumulate large numbers of αβ-TCR(+), CD4(-), CD8(-) double negative (DN) T cells. The function of these DN T cells remains largely unknown. The common γ subunit of the activating Fc receptors, FcRγ, plays an important role in mediating innate immune responses. We have shown previously that a significant proportion of DN T cells express FcRγ, and that this molecule is required for TCR transgenic DN T cells to suppress allogeneic immune responses. Whether FcRγ plays a critical role in LPR DN T cell-mediated suppression of immune responses to auto and allo-antigens is not known. Here, we demonstrated that FcRγ(+), but not FcRγ(-) LPR DN T cells could suppress Fas(+) CD4(+) and CD8(+) T cell proliferation in vitro and attenuated CD4(+) T cell-mediated graft-versus host disease. Although FcRγ expression did not allow LPR DN T cells to inhibit the expansion of Fas-deficient cells within the LPR context, adoptive transfer of FcRγ(+), but not FcRγ(-), DN T cells inhibited lymphoproliferation in generalized lymphoproliferative disease (GLD) mice. Furthermore, FcRγ acted in a cell-intrinsic fashion to limit DN T cell accumulation by increasing the rate of apoptosis in proliferated cells. These results indicate that FcRγ can confer Fas-dependent regulatory properties on LPR DN T cells, and suggest that FcRγ may be a novel marker for functional DN Tregs.

Figure 1. FcRγ-expressing DN T cells are a distinct effector-memory subset in both B6 and LPR mice.

A. Freshly isolated B6 (n = 3, left column) and LPR (n = 4, right column) splenocytes were stained for TCRβ, CD4, CD8, NK1.1, CD16/32, CD44, and CD62L expression and examined by flow cytometry. Within the DN T cell gate (TCRβ⁺, CD4⁻, CD8⁻, NK1.1⁻), expression of CD44 (top row) and CD62L (bottom row) in the CD16⁺ (shaded) and CD16⁻ (unshaded) subsets was plotted. B. Median fluorescence intensity (MFI) of CD44 staining in CD16⁺ and CD16⁻ DN T cells for all 7 mice is shown. Unpaired t tests p = 0.006 (CD16⁺ vs. CD16⁻ B6 DN T cells) and p = 0.0009 (CD16⁺ vs. CD16⁻ LPR DN T cells). C. MFI of CD62L staining in CD16⁺ and CD16⁻ DN T cells for all 7 mice is shown. Unpaired t tests p = 0.0036 (CD16⁺ vs. CD16⁻ B6 DN T cells) and p = 0.0165 (CD16⁺ vs. CD16⁻ LPR DN T cells).

Figure 2. FcRγ deficiency results in an increased accumulation of DN T cells.

A. B6.LPR.FcRγ^+/+ and B6.LPR.FcRγ^−/− mice were given 4×10⁷ CB6F1 splenocytes intravenously. After 7 days, varying numbers of DN T cells were purified and incubated for a further 3 days with irradiated CB6F1 splenocytes (10⁵/well), after which 1 µCi ³H-thymidine was added to each culture. Thymidine uptake, reflecting live proliferated cell number, was determined by scintillation counting. Two-way ANOVA p<0.0001; **Bonferroni post tests p<0.001. B. B6.SCID mice (FcRγ^+/+) received 10⁷ B6.LPR.FcRγ^+/+ (n = 3, upright triangles) or B6.LPR.FcRγ^−/− (n = 3, inverted triangles) DN T cells. On days 1, 5, 7, 10, and 14 blood samples were obtained and peripheral blood mononuclear cells (PBMCs) were stained for TCRβ, CD4, CD8, and NK1.1 expression. The percentage of DN T cells in the PBMC compartment was then determined by flow cytometry. Two-way repeated measures ANOVA p = 0.0279 for the effect of FcRγ genotype; Bonferroni post test p<0.001 at day 14. C. At day 7 and day 14, the splenocytes of the DN T cell recipients were counted and stained for TCRβ, CD4, CD8, and NK1.1 and examined by flow cytometry. The number of splenic DN T cells in each type of recipient was then determined. Two-way ANOVA p = 0.0005 for the effect of FcRγ genotype; Bonferroni post test p<0.001 at day 14.

Figure 3. Fc receptor γ-expressing DN T cells are lost with disease progression in LPR mice and have an increased rate of apoptosis ex vivo.

A. Splenocytes of young (≤12 weeks, n = 4) and older (≥16 weeks, n = 4) female LPR mice were counted and stained for expression of CD4, CD8, NK1.1, TCRβ, and CD16 and then examined by flow cytometry. The percentage of DN T cells expressing CD16 was examined as a function of total spleen cell count; linear regression r² = 0.91, p = 0.0002. B. LPR FcRγ^+/+ (n = 5) and LPR FcRγ^−/− mice (n = 5) aged 8 weeks were fed BrdU for 6 days. Their splenocytes were then stained for expression of CD4, CD8, NK1.1, TCRβ, annexin V and CD16, fixed and stained for BrdU incorporation. They were then examined by flow cytometry. Left panels show expression of CD16 versus side light scatter in LPR FcRγ^+/+ (top) and LPR FcRγ^−/− DN T cells (bottom); CD16^hi and CD16^lo gates are indicated, and the numbers above each gate indicate the percentage of DN T cells falling into the indicated gates. Right panels show representative BrdU and annexin V staining in LPR FcRγ^−/− DN T cells (bottom) and in the CD16^lo and CD16^hi subsets of LPR FcRγ^+/+ DN T cells (middle and top panels, respectively). Numbers inside contour plots reflect the percentage of gated cells falling into each quadrant. C. The percentages of DN T cells staining with annexin V in LPR FcRγ^−/− mice and in the CD16^hi and CD16^lo subsets of LPR FcRγ^+/+ mice are presented with respect to BrdU incorporation. Two-way ANOVA p<0.0001; **Bonferroni post tests p<0.01 compared with either CD16^lo or LPR FcRγ^−/− DN T cells amongst both BrdU⁺ and BrdU⁻ DN T cells.

Figure 4. Control of lymphoproliferative disease by LPR DN T cells requires both Fas and FcRγ.

A. B6.LPR.FcRγ^−/− mice aged 4 weeks received two i.v. injections, two weeks apart, each of either B6.LPR.FcRγ^+/+ DN T cells (n = 4) or B6.LPR.FcRγ^−/− DN T cells (n = 4). Cells derived from one-two donor mice (generally 10–20×10⁶ cells per dose) were used for each injection, ensuring that an equivalent number of B6.LPR.FcRγ^−/− and B6.LPR.FcRγ^+/+ DN T cells were transferred on each occasion. After a further 4 weeks, total spleen and lymph node cell counts were determined. Total cell counts for a group of 8 week old uninjected B6.LPR.FcRγ^−/− mice (n = 9) are shown for comparison. One-way ANOVA p<0.0001; ***Bonferroni post tests p<0.001 for the comparison of untreated mice to both groups of treated mice; Bonferroni post test p = NS for the comparison of the two groups of treated animals. B. B6.GLD mice aged 6 weeks were given either B6.LPR.FcRγ^+/+ or B6.LPR.FcRγ^−/− DN T cells. Cells derived from one-two donor mice were used for each injection (between 10–20×10⁶ cells), ensuring that an equivalent number of B6.LPR.FcRγ^−/− and B6.LPR.FcRγ^+/+ DN T cells were transferred. After 3 weeks, total splenocyte counts were determined. Unpaired t test, p = 0.0027.

Figure 5. FcRγ expression by LPR DN T cells is required for their regulatory function toward Fas-expressing syngeneic T cells.

A. B6 CD4⁺ T cells (10⁴/well) were co-cultured with irradiated CB6F1 splenocytes (10⁵/well) and IL-2. Purified B6.LPR.FcRγ^+/+ or B6.LPR.FcRγ^−/− DN T cells were added in varying ratios. After 4 days, ³H-thymidine (1 µCi/well) was added. After 18 h, ³H-thymidine uptake was determined, and percent suppression was calculated. Two-way ANOVA p<0.0001; Bonferroni post-tests **p<0.01. Data are from one of three experiments with similar results. B. B6 or B6.LPR.FcRγ^+/+ CD8⁺ T cells (10⁴/well) were cultured as in C. Percent suppression was calculated. Two-way ANOVA p<0.0001 for the effect of FcRγ on suppression of B6 CD8⁺ T cells; p = 0.0043 for the effect of FcRγ on suppression of B6.LPR CD8⁺ T cells. Bonferroni post-tests **p<0.01. Data are from one of three experiments with similar results. C. Male CB6F1 mice were lethally irradiated and reconstituted with 2×10⁶ TCD BM alone (BM Only) or with 10⁶ B6 CD4⁺ T cells (BM+CD4⁺), with or without 5×10⁶ B6.LPR.FcRγ^+/+ or B6.LPR.FcRγ^−/− DN T cells. Mice losing >25% of their body weight were sacrificed. *Log rank test p = 0.01 for the comparison of mice treated with B6.LPR.FcRγ^+/+ and B6.LPR.FcRγ^−/− DN T cells. Data are derived from two independent experiments, each with 2–5 mice per group. D. Naive B6 CD8⁺ T cells were used as responders and stimulated by irradiated bm1 splenocytes. Varying numbers of DN T cells isolated from spleens of bm1 splenocyte-treated B6.FcRγ^+/+ (squares) or B6.FcRγ^−/− (triangles) mice were added to the MLR cultures as putative suppressor cells. Cell proliferation was measured by ³H-thymidine incorporation. The data are expressed as percent inhibition of proliferation as compared with the controls to which no putative suppressor cells were added. Data points are the mean +/− SD of triplicate wells and are derived from one of three independent experiments.

Figure 6. CD8⁺ T cells proliferating in response to alloantigen are selectively killed by LPR DN T cells via the Fas pathway.

CD8⁺ T cells (10⁵/well) from B6 (Fas^+/+) or B6.LPR (Fas^LPR/LPR) were labelled with CFSE and cocultured with irradiated CB6F1 splenocytes and IL-2 for 5 days without or with LPR DN T cells in the indicated ratios. After 5 days, the cultures were stained with anti-CD8-APC and 7-AAD and analyzed by flow cytometry. A. The percentage of undivided cells (CFSE^hi) was used to determine the percent suppression at each DN:CD8⁺ T cell ratio. Two-way ANOVA p<0.0001 for the effect of CD8⁺ T cell Fas expression. B. Representative histograms of 7-AAD staining gated on proliferated (CFSE-diluted) Fas^+/+ (top row) and Fas^LPR/LPR (bottom row) CD8⁺ T cells at the indicated DN:CD8⁺ ratios. Numbers inside histograms are the percentages of cells falling in the 7-AAD⁺ gate. C. The fold increase in cell death for proliferated Fas^+/+ (white bars), proliferated Fas^LPR/LPR (black bars), unproliferated Fas^+/+ (light grey bars), proliferated Fas^LPR/LPR (dark grey bars) CD8⁺ T cells is shown. Two-way ANOVA p<0.0001; Bonferrroni post-tests ***p<0.001 and *p<0.05. Data are derived from two independent experiments each with duplicate wells.