Discussion
IL-10 expression by cells of the innate and adaptive immune systems reflects the importance of this cytokine in the tight regulation of the immune response, to minimize pathology during infection. IL-10 expression by Th1 cells has been reported to regulate the immune response in leishmaniasis and toxoplasmosis. However, in many situations, IL-10 is not produced by Th1 cells in response to antigenic stimulation. Our goal was to address the molecular signals that determine whether Th1 cells develop to produce IL-10 (Trinchieri, 2007). Here, we showed that Th1 cells required high-antigen-dose-induced ERK1 and ERK2 phosphorylation and IL-12-induced STAT-4 activation to produce IL-10. Our findings that ERK1 and ERK2 activation was a common pathway required for the production of IL-10 by Th1, Th2, and Th17 cell subsets, which differentiate along distinct pathways, such that IL-10 provides a highly regulated feedback loop to avoid the extremes of excessive inflammation or chronic infections and also allow a protective response to diverse pathogens. 
In certain viral or parasitic infections (Anderson etal., 2007; Brooks etal., 2006; Ejrnaes etal., 2006), high amounts of stimulation may lead to the chronic nonhealing infection shown to be regulated by IL-10. During the course of infection, after initial triggering with antigen, Tcells migrate to the tissue encountering high doses of antigen and factors produced by the innate immune response. Under these conditions, we speculate that Th1 cells will be induced to express high amounts of IL-10, in keeping with reports that IL-10-producing Th1 cells were found in CD4+ clones isolated from BAL but not blood of TB patients (Gerosa etal., 1999). Similarly, the immune response to a clinical isolate of L. major, which produces heavily infected nonhealing lesions, was found to be regulated by IL-10 derived from Foxp3- Th1 cells that coproduce IL-10 and IFN-gamma (Anderson etal., 2007), and the immune response during T. gondii infection was found also to be regulated by Foxp3- Th1 cells (Jankovic etal., 2007). It is likely that IL-10 production by Th1 cells is evoked under conditions of high inflammation and antigenic stimulation, whereas regulatory CD4+ Tcells producing IL-10 may operate to regulate the immune response under conditions in which the pathogen is clinically controlled, such as in infection with L. major (Friedlin strain) (Belkaid etal., 2002; Suffia etal., 2006). We now also reported that CD4+ Tcells cultured with high antigen dose and IL-12 differentiate into canonical Th1 effector cells , which, in addition to expressing large amounts of IFN-gamma and IL-10, lose their IL-2 expression as described before in certain chronic infection models (Sallusto etal., 2004). Our demonstration that loss of IL-2 is accompanied by production of IL-10 offers potential additional mechanisms whereby effector Tcell responses may be dampened during chronic disease. 
Using an invivo transfer model of DO11.10 TCR transgenic cells (Castro etal., 2000), we showed that IL-10-producing Th1 cells were differentiated in the presence of high doses of OVA protein and LPS. We showed here that this induction of IL-10 inTh1 cells invivo was markedly, but not totally, reduced in STAT4-deficient Tcells as observed during T. gondii infection (Jankovic etal., 2002). A high antigenic activation during T. gondii infection or high antigen doses delivered in the presence of LPS, as seen in our system, may compensate for an absolute requirement for IL-12 in the induction of IL-10 by Th1 cells. 
In our invitro system, repeated stimulation of Th1 cells with high antigen doses allowed the development of Th1 cells producing IL-10 in an IL-12-dependent manner. IL-10 production by Th1 cells induced by high antigen dose and IL-12 was independent of IFN-gamma, in keeping with previous findings (Jankovic etal., 2002). However, a role for IFN-gamma in mediating IL-10 reactivation by Th1 cells during secondary infection with T. gondii has been suggested (Shaw etal., 2006). We have found that CD4+ Tcells exposed to a high dose of antigen do not express IL-10 upon restimulation, but can be induced to produce IL-10 upon re-exposure to a high dose of antigen in the recall phase in the absence of added IL-12. However, this is dependent on the induction of IL-12 by antigen-presenting DCs. The combination of both high antigen dose and IL-12 resulted in the highest levels of IL-10 production and correlated with the high levels of ERK1 and ERK2 activation. The increased expression of IFN-gamma observed during the secondary phase will induce increased IL-12 production by DCs and suggests that repeated high-level TCR activation feeds back to upregulate IL-12 production by DC. It is thus likely that in T. gondii infection invivo (Shaw etal., 2006), the requirement for IFN-gamma to induce IL-10, was for feedback upregulation of IL-12 by DCs, which in turn induced IL-10 in the Th1 cells. 
Although IL-10 may be differentially regulated in Th1 and Th2 cells as has been reported (Chang etal., 2007; Wang etal., 2005), some studies suggest the existence of common pathways, but the molecular basis for these is as yet unclear. Costimulatory OX-40 signals have been shown to negatively regulate IL-10 production (Ito etal., 2005) both in Th1 and Th2 cells, whereas ICOS signaling has been suggested to induce IL-10 (Ito etal., 2007; Witsch etal., 2002) in both Th1 and Th2 cells. However, in some cases, ICOS signaling also regulates IL-4 production and Th2 responses (Greenwald etal., 2005). We now provide a common mechanism of ERK1 and ERK2 activation for the regulation of IL-10 production in Th1, Th2, and Th17 cells, although each subset differentiates along a distinct and subset-specific transcriptional pathway. This reinforces the fact that IL-10 is not a Th cell-subset-specific cytokine, but instead is produced in a tightly regulated fashion during each differentiation pathway. Of note, a role for ERK1 and ERK2 activation in the induction of IL-10 production has already been described for macrophages and DC (Agrawal etal., 2006; Hacker etal., 1999). 
Differential transcriptional regulation of IL-10 in Th1 and Th2 cells has been suggested (Chang etal., 2007; Wang etal., 2005), and extensive histone acetylation of the IL-10 gene is detectable in fully polarized Th2 cells, but not Th1cells (Chang etal., 2007). We provide evidence that IL-10 is produced in canonical Th1 cells and that its expression correlates with the expression of T-bet and the highest IFN-gamma production, in keeping with our observations that high-dose antigen stimulation and IL-12 signaling are required for IL-10 and IFN-gamma expression. It has also been shown that maintenance of IL-10 expression is conditional on IL-12 or IL-4 unless the IL-10 gene is imprinted by GATA-3 (Chang etal., 2007), which can remodel the IL-10 locus, thus explaining the highest amounts of IL-10 produced by Th2 cells (Chang etal., 2007; Shoemaker etal., 2006). We show here that high antigen dose and IL-12 drastically downregulate Gata-3 expression, suggesting that additional factors are in place to induce IL-10 expression in Th1 cells, albeit transiently. Expression of c-maf was greatly diminished by high antigen doses in Tcells and yet was unexpectedly maintained by IL-12 and present in Th17 cells. That c-maf expression is common to IL-10-producing Th1, Th2, and Th17 cells and, like IL-10, is dependent on ERK activation in Th1 and Th17 cells for its expression is of interest because c-Maf has been shown to be an essential transcription factor for IL-10 expression in macrophages (Cao etal., 2005). 
In summary, we show that although Th1, Th2, and Th17 CD4+ Tcell subsets differentiate along distinct signaling and transcriptional pathways, they can all be induced to make IL-10. ERK1 and ERK2 activation is required for IL-10 production by all these Th cell subsets. With regard to the expression of IL-10 by Th1 cells, our data provide a mechanism for how IL-10 expression is induced and then amplified and regulated by the levels of antigen and IL-12 encountered in the environment. This provides a mechanism whereby a Th1 cell responds to extrinsic signals, reflecting increased inflammation in the tissue, to tightly regulate the production of IL-10 so as to allow a protective response to eradicate a pathogen with minimal damage to the host and also prevent chronic infection. Moreover, our findings have important implications for the regulation of IL-10 production during an inflammatory Th1 response in infection and may be of relevance for the design of vaccines and for strategies in immunotherapy in infectious diseases. 
