Memory CD4 T cells induce selective expression of IL-27 in CD8+ dendritic cells and regulate homeostatic naive T cell proliferation

Abstract

Naive T cells undergo robust proliferation in lymphopenic conditions, whereas they remain quiescent in steady-state conditions. However, a mechanism by which naive T cells are kept from proliferating under steady-state conditions remains unclear. In this study, we report that memory CD4 T cells are able to limit naive T cell proliferation within lymphopenic hosts by modulating stimulatory functions of dendritic cells (DC). The inhibition was mediated by IL-27, which was primarily expressed in CD8(+) DC subsets as the result of memory CD4 T cell-DC interaction. IL-27 appeared to be the major mediator of inhibition, as naive T cells deficient in IL-27R were resistant to memory CD4 T cell-mediated inhibition. Finally, IL-27-mediated regulation of T cell proliferation was also observed in steady-state conditions as well as during Ag-mediated immune responses. We propose a new model for maintaining peripheral T cell homeostasis via memory CD4 T cells and CD8(+) DC-derived IL-27 in vivo.

Figure 1. Preexisting memory phenotype CD4 T cells inhibit naïve T cell proliferation

A. Groups of Rag−/− mice were transferred with 3 × 10⁶ Thy1.2 CD4 T cells (1′ CD4). Subsequently, the recipients were transferred with CFSE labeled FACS sorted Thy1.1 naïve 2′ T cells (1 × 10⁶, containing both CD4 and CD8) 4 weeks post the 1′ CD4 transfer. Histograms shown are CFSE profiles of Thy1.1-gated T cells (2′ CD4 or 2′ CD8) examined 7 days post transfer. B. CFSE labeled Thy1.1 naïve T cells were transferred into Rag−/− mice. CFSE profiles were determined 7 days post transfer. The experiments were repeated three times and similar results were observed. The average ± SD of the proportion of T cells that fully diluted CFSE is indicated. LN, peripheral LN; mLN, mesenteric LN. Similar results were found in the spleen (not shown).

Figure 2. Memory CD4 T cells suppress naïve T cell proliferation through interacting the same APC

A and B. Lethally irradiated Rag−/− mice were transferred with BM cells from TCRβ−/− (A) or 1:1 mixture of TCRβ−/− and TCRβ−/− MHC II−/− mice (B). After 6 weeks of BM reconstitution, MHC I and MHC II expression of CD11c+ splenic DCs was examined by FACS analysis. The dot plots show the expression of K^b and I-A^b of CD11c+ splenic DC. The reconstituted mice were transferred with 3 × 10⁶ Thy1.2 CD4 T cells (1′ CD4). CFSE labeled Thy1.1 naïve CD8 T cell transfer was performed 4 weeks post the 1′ CD4 T cell transfer. CFSE profiles of the Thy1.1 CD8 T cells were examined 7 days post transfer (A and B, right histogram). CD8 T cells were also transferred into reconstituted recipients without the 1′ CD4 transfer (A and B, left histogram). Histograms shown are CFSE profiles of Thy1.1-gated CD8 T cells. C. CFSE labeled naïve Thy1.1 CD8 T cells were transferred into TCRβ−/− recipients and their proliferation was examined 7 days post transfer. Data are representative of 4 individually tested mice from two independent experiments. The average ± SD of the proportion of T cells that fully diluted CFSE is indicated. D and E. Irradiated B6 × B10.A Rag−/− F1 mice were reconstituted with BM cells from B6 × B10.A Rag−/− F1 (D) or with 1:1 mixture of BM cells from B6 Rag−/− and B10.A Rag−/− mice (E). After 6 weeks of BM transfer, reconstitution was confirmed by FACS analysis. Contour plots show expected I-A^b and I-E^k expression of CD11c+ splenic DCs (D and E). BM chimeras generated above were transferred with 3 × 10⁶ 1′ I-A^b restricted (and I-E^k/I-A^k tolerant) Thy1.1 CD4 T cells (please see Supplementary Figure S4). Four weeks post the 1′ transfer, the recipients were transferred with CFSE labeled Ly5.1 I-A^k/I-E^k restricted (I-A^b tolerant) FACS sorted naïve CD4 T cells. 250γg anti-NK1.1 mAb was injected at days -1, 3 and 6 of the second transfer. CFSE profiles were examined 7 days post transfer (right histogram). Naïve CD4 T cells were also transferred into reconstituted recipients without the 1′ CD4 transfer (left histogram). Histograms shown are CFSE profiles of Ly5.1-gated CD4 T cells. Data are representative of 4-5 individually tested mice from two independent experiments. The average ± SD of the proportion of T cells that fully diluted CFSE is indicated. (F) Total numbers of 1′ memory CD4 T cells from the groups described in the Figure 2A and 2B were counted by FACS analysis. (G) Total numbers of 1′ memory CD4 T cells from the groups described in the Figure 2D and 2E were counted. Data shown are the average ± SD from two independent experiments.

Figure 3. Continuous presence of memory CD4 T cells is necessary to inhibit naïve T cell proliferation

Rag−/− mice that received 1′ Thy1.2 CD4 T cells and 2′ CFSE labeled Thy1.1 naïve T cells as described above were injected with 250μg anti-Thy1.2 Ab at the time of 2′ naïve T cell injection. Shown are the CFSE profiles of donor T cells in the LN tissues examined 7 days post the 2′ transfer. Similar results were observed in the spleen tissues. The number shown in the histograms indicates the proportion of donor T cells that fully diluted the CFSE. Similar results were observed from two independent experiments.

Figure 4. IL-27 induced by CD4 T cells inhibits naïve T cell proliferation

A. Groups of Thy1.1 TCRβ−/− mice were transferred with 3 × 10⁶ Thy1.1 CD4 T cells (1′ CD4) and with 1 × 10⁶ CFSE labeled Thy1.2 naïve CD4 T cells isolated from wild type or IL-27R−/− mice at 4 weeks post 1′ transfer. CFSE profiles of Thy1.2-gated CD4 T cells were examined 7 days post transfer. Histograms shown are representative of 4 individually tested mice from two independent experiments. The average ± SD of the proportion of T cells that fully diluted CFSE is indicated. LN, peripheral LN; mLN, mesenteric LN. Similar results were found in the spleen (not shown). B. CD11c+ DC were FACS sorted from spleen cells of wild type, Rag−/−, and Rag−/− mice that received 1′ CD4 T cells (4 weeks earlier). Expression of Ebi3 and p28 was measured by quantitative PCR analysis. The expression was normalized to endogenous control GAPDH. Data shown are the mean ± SD of two independent experiments. *, p<0.01; **, p<0.05.

Figure 5. IL-27 is primarily expressed by DC

(A) Kinetics of IL-27 expression following CD4 T cell transfer. Groups of TCRβ−/− mice were transferred with CD4 T cells. IL-27 expression in CD11c+ splenic DC was weekly measured by qPCR. Wild type B6 and TCRβ−/− mice are included as positive and negative control, respectively. (B) IL-27 subunits expression by DCs and B cells. CD11c+ DCs and CD19+ B cells were sorted from splenocytes of TCRβ−/− mice that received CD4 T cells 4 weeks earlier. Expression of IL-27 subunits was subsequently examined. Experiments were repeated three times as shown.

Figure 6. IL-27 is primarily expressed by CD8+ DC subsets

A. Splenic DC from wild type B6 mice were sorted into different subsets based on the expression of CD8. Expression of IL-27 subsets was determined by real time PCR analysis. The experiments were repeated twice with similar results. B. Bone marrow derived DCs were cocultured with OVA-specific OT-II CD4 T cells with or without OVA peptide antigen. After 48 hours of culture, DCs were sorted from the culture into CD8+ and CD8− CD11c+ cells, and the expression of IL-27 subunits was determined by qPCR analysis. The experiments were repeated twice with similar results. C. CD8+ and CD8− splenic DC subsets were purified from Rag−/− mice that received CD4 T cells 7 days earlier by cell sorting and IL-27 expression was examined by qPCR. Expression of Ebi3 and p28 was normalized to endogenous control GAPDH. The data shown is average ± SD from two independent experiments. *, p<0.01; **, p<0.05.

Figure 7. IFNγ induces IL-27 expression in CD8+ DC

A. OT-II T cells were cocultured with BM derived DC in the presence or absence of neutralizing anti-IFNγ Ab. CD8+ and CD8− DC were sorted from the culture 48 hours later. IL-27 expression was examined by qPCR. Coculture without peptide was included as a control. The data shown is average ± SD from two independent experiments. B. Rag−/− mice were transferred with wild type or IFNγ−/− CD4 T cells. Four weeks later, CFSE labeled Thy1.1 naïve CD4 T cells were transferred into the recipients. CFSE profiles of newly transferred cells were determined 7 days post transfer. The data shown is average ± SD from two independent experiments. C. CD4 T cells were transferred into Rag−/− or Rag−/− IFNγR−/− recipients. CD8+ and CD8− splenic DC were sorted 10 days post transfer and IL-27 expression was examined by qPCR. *, p<0.01; **, p<0.05.

Figure 8. IL-27 regulates T cell proliferation in steady-state conditions as well as during Ag-mediated immune responses

A. Groups of wild type (filled circle) and IL-27R−/− (open circle) mice were injected with BrdU. Mice were sacrificed 24 hours later and BrdU incorporation by CD44^high memory phenotype T cells was examined. Each symbol represents individually tested mice. B and C. Groups of irradiated Rag−/− mice were reconstituted with 1:1 mixture of Ly5.1 WT plus Ly5.2 IL-27R−/− (or Ly5.2 WT) BM cells. B. Following the reconstitution, mice were bled every two weeks, and the relative proportion of Ly5.1/Ly5.2 T cells was examined. C. In vivo proliferation of reconstituted CD4 T cells was examined by BrdU incorporation experiments as described above. D. 5 × 10⁵ WT or IL-27R−/− Ly5.1 OT-II CD4 T cells were transferred into B6 recipients. The recipients were subsequently immunized s.c. with 50γg OVA protein plus 10μg LPS. Ly5.1 T cell expansion within the draining lymph node was determined by FACS analysis 7 days post immunization. Each symbol represents individually tested mouse. *, p<0.01; **, p<0.05; ***, p<0.001.