Thymic DCs derived IL-27 regulates the final maturation of CD4+ SP thymocytes

IL-27, as a pleiotropic cytokine, promotes the differentiation of naïve T cells to Th1, while suppressing Th2 and Th17 differentiation in the periphery. However, the role of IL-27 in the thymocyte development remains unknown. Here we showed that IL-27 was highly expressed in thymic plasmacytoid dendritic cells (pDCs) while its receptor expression was mainly detected in CD4+ single-positive (SP) thymocytes. Deletion of the p28 subunit in DCs resulted in a reduction of the most mature Qa-2+ subsets of CD4+ SP T cells. This defect was rescued by intrathymic administration of exogenous IL-27. In vitro differentiation assay further demonstrated that IL-27 alone was able to drive the maturation of the newly generated 6C10+CD69+CD4+ SP cells into Qa-2+ cells. Collectively, this study has revealed an important role of thymic DCs-derived IL-27 in the regulation of the phenotypic maturation of CD4+ SP thymocytes.

For the isolation of CD4 + naive T cells, cells from lymph nodes (including axillary, mesenteric, and inguinal lymph nodes) were stained for CD4, CD8, CD44 and CD62L. Sorting was performed to obtain CD4 + CD8 − CD44 lo CD62L hi cells.
RNA was isolated from freshly isolated cells or cells cultured under different conditions using Trizol reagent. cDNA was synthesized using reverse transcription kit (Progema). For quantitative PCR, iQ SYBR Green Supermix (Bio-Rad Laboratories) was used according to the manufacturer's instructions. The amplification was performed on an iCycler (Bio-Rad Laboratories). GAPDH was used as an internal control. The primers used in the study were listed in supplementary Table 1.
In vitro cell culture. 1 × 10 6 SP1 cells were cultured in RPMI1640 with 10%FCS in 24-well plates in the presence or absence of 1 × 10 5 thymic DCs. IL-27 (2 ng/ml), IL-30 (IL-27p28) (10 ng/ml) and/or anti-IL-27p28 (1 μ g/ml) were included in the culture if necessary. At various time points, the culture was harvested and analyzed for the generation of Qa-2 + cells by flow cytometry.

IL-27 intrathymic injection.
Intrathymic injection was performed as described 10 . In brief, 20 μ l IL-27 (8 μ g/ml) was injected intrathymically using a Hamilton syringe. The control mice were injected with 20 μ l PBS. After 48 hours post injection, the recipient mice were sacrificed and analyzed for thymocyte development.

Results
The thymic expression of IL-27 and its receptors. To investigate the potential role of IL-27 in thymocyte development, we first examined the expression of IL-27 and its receptors in various thymic cell populations. The thymic DCs are composed of three major subsets, including CD11c int CD45RA + plamacytoid Total thymocyte, CD11c int CD45RA + pDC, CD11c hi CD8α + CD172α − and CD11c hi CD8α − CD172α + cDC, CD45 − Epcam hi Ly51 + cTEC, CD45 -Epcam hi Ly51 -mTEC and CD11b + F4/80 + macrophage were isolated from C57BL/6J mice. The expression of p28 and EBI3 mRNA was evaluated by quantitative PCR. nd, no detectable. (B) IL-27 protein expression in pDC, CD8α + and CD172α + cDC was analyzed by flow cytometry following intracellular staining with anti-IL-27. Representative histograms are shown on the left and the mean fluorescence intensity (MFI) on the right. (C) Purified DN, DP, CD8 SP and the four subsets of CD4 SP (SP1-4) thymocytes were analyzed for IL-27RA and gp130 mRNA expression using quantitative PCR. The lymph node (LN) naïve T cell was used as a positive control. All the experiments were repeated for more than three times and the data are shown as mean ± s.d. *p < 0.05; * * p < 0.01.
Peripheral naïve T cells, NK and NKT cells are good responder of IL-27 13,29 . To identify the target cells of IL-27 in the thymus, we analyzed expression of IL-27RA and gp130 in thymocytes. Compared with DN, DP and CD8 SP cells, each subset of CD4 SP thymocytes had an increased expression of IL-27RA and gp130 mRNA, approaching approximately half of the level in naive T cells (Fig. 1C). Therefore, CD4 SP thymocytes may represent the major IL-27-responding population in the thymus.
Reduction of CD69 − Qa−2 + SP4 thymocytes in CD11c-cre p28 flox/flox mice. Given the highly restricted expression of IL-27 in thymic DCs, we sought to determine the impact of DC-specific deletion of p28 on thymocyte development using CD11c-cre p28 flox/flox mice previously reported by Zhang et al. 28 . PCR analysis revealed the effective disruption of the gene encoding p28 in the genome ( Fig. 2A), which was accompanied by a sharp reduction in p28 mRNA and IL-27 protein expression in purified thymic DCs (Fig. 2B). In contrast, the expression of EBI3, another composing subunit of IL-27, was not affected (Fig. 2B). In consideration of the possibility that ligand deficiency may alter the expression of cognate receptors, we also examined IL-27RA and gp130 and Ebi3 mRNA levels were determined by quantitative PCR in total thymic DCs form CD11c-cre p28 flox/flox and WT mice. The IL-27 protein expression was measured by flow cytometry. (C) IL-27RA and gp130 mRNA expression was compared in CD4 SP thymocytes form CD11c-cre p28 flox/flox and WT mice. The experiments shown in B and C were repeated for more than three times and the data are presented as mean ± s.d. * * p < 0.01. expression in CD4 SP thymocytes. Comparable mRNA levels of IL-27RA and gp130 were detected in CD11c-cre p28 flox/flox mice and the WT littermates (Fig. 2C).
Thymocyte development in the CD11c-cre p28 flox/flox mice was subsequently analyzed by flow cytometry. The KO and WT mice showed similar patterns of CD4/CD8 staining for total thymocytes and CD44/CD25 staining for DN thymocytes (Fig. 3A) and contained equal numbers of DN, DP, CD4 SP and CD8 SP cells (data not shown). In view of the high expression of IL-27R in CD4 SP thymocytes, a detailed analysis was performed for this population. We had previously defined a four-stage developmental program for CD4 SP thymocytes on the basis of cell surface expression of 6C10, CD69 and Qa-2. The SP4 stage, which was featured by Qa-2 expression, represents the most mature one 4 . Although there was no obvious difference in the total number of CD4 SP thymocytes, p28-deficiency resulted in a significant decrease of CD69 − Qa-2 + SP4 cells (Fig. 3B-D), indicating that IL-27 was involved in the final maturation of CD4 SP thymocytes.
Enhanced production of Qa-2 + cells by IL-27 in cultures of SP1 thymocytes. It has been previously shown that purified SP1 cells, when co-cultured with thymic DCs or TECs, were able to undergo phenotypic maturation, leading to the generation of Qa-2 + cells 9,30 . This system provided an opportunity to directly examine the impact of IL-27 on the maturation of CD4 SP thymcoytes in vitro. The profile and purity of freshly isolated CD4 + SP1 cells was shown in Fig. 4A. Notably, IL-27 alone induced a substantial number of Qa-2 + cells. Such an effect was abolished with the addition of anti-IL-27 (Fig. 4B), confirming the specificity of IL-27 action. Similar numbers of live cells were recovered from the cultures with or without IL-27 (Fig. 4B), ruling out the possibility that IL-27 preferentially expanded contaminating Qa-2 + cells in the culture. As the p28 subunit is reported to be able to act independently as IL-30, we also test the biological activity of IL-30 in similar cultures. As shown in Fig. 4B, IL-30 failed to support the production of Qa-2 + cells.
Thymic DCs were next isolated from WT and p28-deficient mice and compared for their capacity to support SP1 maturation. Compared to the co-culture with WT DCs, the one with p28-deficient DCs gave rise to a significantly reduced number of Qa-2 + SP4 cells. More intriguingly, the majority of thymocytes in the culture acquired Qa-2 expression with the addition of exogenous IL-27, regardless of the origin of the thymic DCs (Fig. 4C). These in vitro data support an important role of IL-27 in the regulation of SP thymocyte maturation.  Restored Qa-2 + CD4 SP population in p28-deficient mice following intrathymic injection of IL-27. We further tested whether administration of exogenous IL-27 was capable of rescue the maturation defect in CD11c-cre p28 flox/flox thymi. Mice received intrathymic injection of recombinant IL-27 and their thymocyte development was evaluated 48 hours after injection. As shown in Fig. 5, treatment with IL-27 resulted in a drastic increase of the Qa-2 + SP4 subset in CD11c-cre p28 flox/flox mice, raising it to a level comparable to that in WT mice. On the other hand, neither the total CD4 SP thymocyte nor the SP4 subset was significantly altered in WT mice following IL-27 administration. These results support that loss of IL-27 is a direct cause of the defective maturation of CD4 SP thymocytes in p28-deficient mice.

Qa-2 expression was positively co-related with IFIT1, IFIT3 and IRF7. In peripheral T cells, IL-27
conducts signals through interaction with its cognate receptor IL-27RA and gp130 and the subsequent activation of the JAK-STAT pathway 31,32 . Both IL-27 receptor subsets IL-27RA and gp130 were abundant in CD4 SP thymocytes (Fig. 1C). We speculate that the JAK-STAT pathway may also be involved in IL-27-induced maturation of CD4 SP cells. Indeed, the tyrosine phosphorylation of STAT1 and STAT3 were found to be significantly elevated when freshly isolated CD4 + SP thymocytes were stimulated with IL-27. As expected, p28 deficiency did not altered the response of CD4 + SP thymocytes to IL-27 (Fig. 6A).
Our previous transcriptome analyses of CD4 SP thymocytes have demonstrated that there is a sharp increase of multiple gene transcripts at the SP4 stage, many of which are related to the JAK-STAT signaling cascade, such as IFIT1, IFIT3, IRF7 and IRF8 33 (The microarray data have been deposited in NCBI's Gene Expression Omnibus. http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE30083). Notably, IL-27 stimulation induced a drastic increase of their transcription in SP3 cells (Fig. 6B). On the other hand, SP3 cells from p28-deficient mice had a much lower expression of IFIT1, IFIT3, and IRF7 than the wild type counterparts (Fig. 6C). Therefore, these genes may be directly regulated by IL-27 signaling.

Discussion
The present study revealed a previously undescribed function of IL-27 in the maturation of CD4 SP thymocytes. IL-27 was found to be highly expressed in thymic DCs, while its receptor was mainly detected in CD4 SP thymocytes. When added into cultures, IL-27 was able to drive the phenotypic maturation of newly generated CD4 SP thymocytes into Qa-2 + cells. On the contrary, mice with DC-specific deletion of IL-27p28 exhibited a significant reduction of the most mature, Qa-2-expressing CD4 SP thymocytes. Notably, such a defect was rescued following intrathymic injection of exogenous IL-27. These data indicate that thymic DC-derived IL-27 plays an important role in the late stage developmental of thymocytes.
The newly generated SP thymocytes migrate into the thymic medulla, where they undergo phenotypic and functional maturation before being exported to the periphery 9 . Our previous studies have shown that the maturation of CD4 SP cells follows a program composed of multiple intermediate stages 10 . While the progression from SP1 to SP3 seems to be a cell-autonomous process, the transition from SP3 to SP4 apparently requires signals  provided by thymic stromal cells 9,30 . The molecular nature of the signal, however, remains elusive. The present study provides evidence that IL-27 is involved in the regulation of this transition. An intriguing yet unresolved issue is whether IL-27 drives the full maturation or simply allows the acquisition of Qa-2 expression. Despite the reduced population size, a substantial number of SP4 cells were generated in CD11c-cre p28 flox/flox mice. When tested for their functionality, these cells displayed similar proliferative responses and even increased cytokine production in comparison to the wild type cells (data not shown). Therefore, signals other than IL-27 also contribute to the final maturation of CD4 SP thymocytes. Consistent with the observation in vivo, the thymic DC isolated from CD11c-cre p28 flox/flox mice was able to support the generation of Qa-2 + cells from SP1 cells in culture, although less efficient than the wild type DC. As a matter of fact, such a property is not even restricted to dendritic cells. Thymic epithelial cells were also capable of driving the maturation of SP1 cells in culture 9 .
During CD4 SP maturation, three important events were occurred in the medulla, including the differentiation of regulatory T cells, negative selection and thymic emigration. Our results suggest that IL-27 is involved in the late stage thymocytes phenotypic maturation. Therefore, we try to figure out the impact of IL-27 on these critical events. However, using the RIP-mOVA and OT-II model, we observed no change in the deletion of Vα 2 + Vβ 5 + cells in wild type and p28-deficient mice (supplementary Fig. S1). Meanwhile, we had examined Foxp3 expression in SP thymocytes, no obvious difference was observed for the Treg population in wild type and CD11c-cre p28 flox/flox mice (supplementary Fig. S2). Indeed, our ongoing work shows that p28 deficiency does appear to render the mice more susceptible to autoimmune disorders, especially when crossed with Aire-deficient mice (data not shown). Data currently available, however, argues against the possibility of defective negative selection or impaired Treg differentiation. Using the Rag2p-GFP model, we found a decrease in GFP mid/hi cells among CD4 + T cells in lymph nodes and spleen in CD11c-cre p28 flox/flox mice (supplementary Fig. S3). In adult mice, more that 80% recent thymic emigrants were Qa-2 + cells 30 . Therefore, the thymic output is indeed affected by the retarded maturation of Qa-2 + SP4 cells in the thymus.
Should read "(B) IL-27 protein expression in pDC, CD8α + and CD172α + cDC was analyzed by flow cytometry following intracellular staining with anti-IL-27". Figure 6B and 6C were inverted. The correct Figure 6 appears below as Figure 1.