IL-21 limits peripheral lymphocyte numbers through T cell homeostatic mechanisms

Abstract

Background: IL-21, a member of the common gamma-chain utilizing family of cytokines, participates in immune and inflammatory processes. In addition, the cytokine has been linked to autoimmunity in humans and rodents.

Methodology/principal findings: To investigate the mechanism whereby IL-21 affects the immune system, we investigated its role in T cell homeostasis and autoimmunity in both non-autoimmune C57BL/6 and autoimmune NOD mice. Our data indicate that IL-21R knockout C57BL/6 and NOD mice show increased size of their lymphocyte population and decreased homeostatic proliferation. In addition, our experimental results demonstrate that IL-21 inhibits T cell survival. These data suggest that IL-21 acts to limit the size of the T cell pool. Furthermore, our data suggest IL-21 may contribute to the development of autoimmunity.

Conclusions/significance: Taken together, our results suggest that IL-21 plays a global role in regulating T cell homeostasis, promoting the continuous adaptation of the T cell lymphoid space.

Figure 1. Increased cell numbers in IL-21R-deficient mice.

(A) Total splenocyte numbers in 6 to 8-week old C57BL/6 mice compared to age-matched NOD mice (n = 14 mice/group, p = 0.03). (B) Total cell numbers (n = 14 mice/group, p = 0.008), absolute CD4+ T cell numbers (n = 13 mice/group, p = 0.0002) and absolute CD8+ T cell numbers (n = 14 mice/group, p = 0.003) for the splenocytes of NOD mice, at 6 to 8-weeks of age, compared to those from age-matched IL-21R KO NOD mice. (C) Total cell numbers (n = 13 mice/group, p = 0.01), absolute CD4+ T cell numbers (n = 12 mice/group, p = 0.02), and absolute CD8+ T cell numbers (n = 19 mice/group, p = 0.04) from the spleens of C57BL/6 mice, average age of 8 weeks, and IL-21R KO C57BL/6 mice, average age of 9 weeks. (D) Absolute B cell numbers (n = 6 mice/group, p = 0.01) for mice described in C. Data shown are from three or four independent experiments. Results are presented as mean absolute numbers ±SEM.

Figure 2. Reduced T cell proliferation in IL-21R-deficient mice.

Representative histogram plots for individual NOD (thin line) and IL-21R KO NOD (thick line) mice showing flow cytometric analysis of BrdU incorporation in CD4+CD44^hi and CD8+CD44^hi populations after 5 days of BrdU treatment. The data, obtained from analyzing the spleens of 7 to 9-week-old NOD and age-matched IL-21R KO NOD mice, is presented as mean percentages of BrdU+ cells ±SEM from one experiment with a total of n = 4 mice/group. The experiment, using the same number and age of mice, was repeated twice with similar results. (B) Representative flow cytometric analyses of splenocytes from individual C57BL/6 (thin line) and IL-21R KO C57BL/6 (thick line) mice showing BrdU incorporation in CD4+CD44^hi and CD8+CD44^hi T cells after 5 days of BrdU treatment. Mean percentages of BrdU+ cells ±SEM are shown from one experiment with a total of n = 4 mice/group with an average age of 10–12 weeks. The experiment was repeated three times with similar results. For both A and B, the gate on the histogram plots represents isotype control staining. (C) Representative dot plot showing expression of CD44 and CD62L in the CD4+ population of individual NOD and IL-21R KO NOD mice. Pooled data ±SEM from a total of n = 8 mice/group, at 7 to 9-weeks of age, from two independent experiments is shown. Gates are based on isotype control staining.

Figure 3. Expansion of IL-21R-deficient T cells upon encounter with self-antigens.

CFSE-labeled splenocytes from 6 to 8-week-old (A) OT1 or (B) IL-21R KO OT 1 mice were adoptively transferred into wild-type Rip-OVA recipient mice, with an average age of 8-weeks, for five days. Cells were then recovered from the pancreatic lymph nodes of recipient mice and analyzed by flow cytometry for CFSE dilution to measure proliferation. A representative CFSE dilution profile of the indicated Vβ5.1+/5.2+ (donor) T cells from individual recipient mouse is shown. The data shown reflects the mean percentage of CFSE^hi and CFSE^lo T cells ± SEM for a total of n = 15 mice/group from four independent experiments, p = 0.04. Analysis of cells before transfer indicated >80% were naïve (CD8+CD44l^o, data not shown).

Figure 4. Increased T cell survival factors in IL-21R-deficient mice.

(A) Representative histograms showing total Bcl-2 expression in indicated CD8 populations for individual NOD (dashed line) and IL-21R KO NOD (solid line) mice relative to isotype control staining (gray line). Below, the graph represents Bcl-2 flow cytometric data ±SEM for one of two independent experiments using n = 4 mice/group. Similar results were found in both experiments. Pancreatic lymph nodes of NOD mice with an average age of 13-weeks and age-matched IL-21R KO NOD mice were analyzed. (B) Levels of Bcl-2 expression in the indicated splenocyte subsets for individual C57BL/6 (dashed line) and IL-21R KO C57BL/6 (solid line) mice are shown in representative histograms relative to isotype control staining (gray line). Below, graphical representations of the percentage of Bcl-2 expression ±SEM pooled from three independent experiments with a total of n = 12 mice/group. 6-8-week-old C57BL/6 mice and aged-matched IL-21R KO C57BL/6 mice were used.

Figure 5. IL-21 antagonizes the pro-survival effects of IL-7 in vitro.

Analysis of the effects of IL-21 on T cell survival in the presence or absence of IL-7. Cells from the spleen and pancreatic lymph nodes of NOD or OT1 C57BL/6 mice were cultured with IL-7 (1 ng/ml) and/or IL-21 (1 ng/ml or 10 ng/ml) as described in Materials and Methods. Graphical representation of Annexin V staining by FACS analysis for CD8+ T cell populations from (A) NOD and (B) OT1 C57BL/6 mice. Percentage of Annexin V+ cells cultured in media alone (no cytokine treatment) was set at 100 and Annexin V+ cells treated with cytokine are represented as fold change relative to media alone ±SEM. Percentage of Annexin V+ cells cultured with IL-21 by itself at 10 ng/ml is similar to cells cultured with IL-21 at 1 ng/ml (data not shown). The data shown is pooled from two independent experiments.

Figure 6. Decreased T cell effector function in IL-21R-deficient NOD mice.

Representative dot plots from individual NOD and IL-21R KO NOD mice showing frequency of (A) TNFα+/CD69+, TNFα+ /CD44+, (B) IFNγ+ /CD69+, IFNγ+/CD44+ populations in CD8+ T cells and (C) TNFα+/CD69+, TNFα+/CD44+, (D) IFNγ+/CD69+, IFNγ+/CD44+ populations in CD4+ T cells. Gates are based on isotype control staining. Graphs represent pooled flow cytometric data from pancreatic lymph nodes of 11-to-13 week old mice for two independent experiments with a total of n = 8 mice/group. Results are presented as the mean percentages ±SEM. (E) Percentages of the indicated cell populations from A–D were multiplied by the total number of cells in the pancreatic lymph nodes for each individual animal to give the absolute numbers for each indicated cell population. Data are shown as mean absolute numbers ±SEM. (F) Cumulative incidence of diabetes was monitored by measuring blood glucose levels in NOD (close circles) and IL-21R KO NOD (open squares) mice (n = 15 mice/group) at the indicated ages. (G) Paraffin sections of pancreata from NOD and IL-21R KO NOD mice stained with H&E. In total, four NOD and four IL-21R KO NOD mice were evaluated, with fifty individual islets examined per strain. The images shown in these panels were obtained from 2 different mice per strain; the NOD mice were 11 weeks of age; the IL-21R KO NOD mice were 13 weeks of age.