Nitric oxide plays a key role in the suppressive activity of tolerogenic dendritic cells

Tolerogenic dendritic cells (DCs) are widely studied for their possible use in the treatment of inflammatory disorders, such as autoimmune diseases. One of the obstacles for the use of this cell-based therapy is the characterization of drugs that are able to modulate DCs. We have previously shown that chloroquine (CQ), an antimalarial agent, has the ability to modulate DCs towards a tolerogenic phenotype.1 These tolerogenic DCs are able to suppress the development of experimental autoimmune encephalomyelitis (EAE), a T cell-driven mouse model of human multiple sclerosis. In addition, several studies have proposed that nitric oxide (NO) plays a major role in the differentiation of regulatory T cells (Tregs) and the suppression of Th1/Th17 cells.2,3 However, little is known about the role of DC-derived NO in the modulation of inflammatory autoimmune responses. Thus, we aimed to evaluate whether NO plays a role in the tolerogenic activity of CQ-treated DCs (CQ-DCs). We found that CQ induces DC production of NO and expression of indoleamine 2,3-dioxygenase (IDO), as well as inducible nitric oxide synthase (iNOS). In addition, CQ-DCs stimulated the differentiation of Tregs at the expense of Th1/Th17 cells. On the other hand, iNOS−/− DCs did not acquire a tolerogenic phenotype following CQ treatment. Rather, CQ-DCsiNOS−/− stimulated the differentiation of Th1/Th17 cells as well as Tregs. In a therapeutic approach, CQ-DCsiNOS−/− were unable to suppress the development of EAE. Gene expression analyses of central nervous system (CNS) tissue from mice that received CQ-DCsiNOS−/− showed an increased expression of inflammatory modulators compared with mice that received CQ-DCsWT. In this work, we show that NO is an important factor in the modulatory activity of tolerogenic dendritic cells. DCs are antigen-presenting cells that can dictate the course of the immune response via the modulation and activation of naive T cells. DC modulation is a possible approach to address the immunosuppression that is often caused by tumors4 and the exacerbated immune response observed in autoimmune diseases.5 Multiple sclerosis, one such autoimmune disease, is a debilitating condition that affects the CNS. Studies in EAE, an experimental mouse model of multiple sclerosis, have found that much of the immunological etiology of the disease development is due to the activity of Th1/Th17 cells, and these studies have found that NO plays a major role in disease progression.2 To verify whether NO is involved in the modulatory activity of tolerogenic DCs, we generated DCs from bone marrow precursors obtained from wild-type (DCsWT) and iNOS−/− (DCsiNOS−/−) mice and treated these DCs with CQ or vehicle (PBS-DCsWT). All protocols involving laboratory animals were approved by the institutional committee (protocol no. 2687-1). NO measurements revealed that CQ treatment induced DCsWT to produce large amounts of NO in an iNOS-dependent manner (Figure 1a). It has been demonstrated that CQ administration results in NO production by the endothelium as well as expression of endothelial nitric oxide synthase in some cell types, leading to renal failure and oxidative distress.6,7,8 We also found that CQ stimulated the expression of iNOS in DCsWT (Figure 1b). Interestingly, the expression of IDO, an immunomodulatory enzyme, was also elevated, and this effect was observed both in CQ-DCsWT and CQ-DCiNOS−/− (Figure 1b). To verify the possible influence of NO in the functional activity of CQ-DCs, modulated DCs were cocultivated with naive T cells, and the differentiation of Tregs and IFN-γ- and IL-10-producing cells was analyzed by flow cytometry. We found that both CQ-DCsWT and CQ-DCsiNOS−/− stimulated the development of Tregs. However, CQ-DCsiNOS−/− were not as efficient as CQ-DCsWT in their ability to stimulate Treg development (Figure 1c). Additionally, CQ-DCsWT failed to induce IFN-γ production, whereas IL-10 was significantly upregulated (Figure 1d). On the other hand, CQ-treated DCsiNOS−/− did not induce IFN-γ production, but were able to stimulate the production of IL-10 (Figure 1d), although not to the degree of CQ-DCsWT. These data support the conclusion that NO may be important for the modulatory activity of CQ-DCs in vitro. To determine whether NO is essential for the tolerogenic function of DCs, we adoptively transferred CQ-DCsWT or CQ-DCsiNOS−/− into WT EAE-bearing mice. EAE was induced as previously described.9 The adoptive transfer was performed 10 days after MOG35–55 immunization. Corroborating our previous report,1 CQ-DCsWT were able to significantly reduce the clinical signs of EAE compared with the control group (PBS-DCsWT) (Figure 1e). However, in the absence of NO production, the modulatory effect of CQ-DCs was abolished as demonstrated by the exacerbation of EAE in mice that received PBS-DCsiNOS−/− and CQ-DCsiNOS−/− (Figure 1e). As NO production has been shown to suppress EAE and the generation of Th17 cells,2,3,10 we then sought to determine whether inflammation in the CNS was affected. Our analysis showed that the transfer of CQ-DCswt into EAE mice (CQ-DCswt+EAE) provoked a significant increase in the CNS expression of anti-inflammatory genes (FOXP3, IL-10, TGF-β, IL-4, iNOS and IDO) alongside decreased expression of pro-inflammatory genes (IFN-γ, IL-17, TNF-α and IL-6) when compared with control mice (PBS-DCsWT+EAE) (Figure 1f). Surprisingly, the transfer of CQ-DCsiNOS−/− into EAE mice (CQ-DCsiNOS−/−+EAE) stimulated the expression of anti-inflammatory mediators, such as FOXP3, IL-10, TGF-β and IDO, as well as pro-inflammatory molecules, such as IL-17, IL-6, IFN-γ and TNF-α (Figure 1g). It has been shown that in the absence of NO, mice develop a more severe form of EAE.2 Additionally, NO stimulates the generation of a regulatory repertoire of T cells.3,10 Notwithstanding, myeloid suppressor cell-derived DCs were shown to be potent modulators of allogeneic T cells through mechanisms that are dependent on cell contact and NO production.11 These observations support the notion that NO plays a modulatory role in the immune system. However, the cellular source of the NO, which helps regulate autoimmune encephalomyelitis, was not known. In this short report, we present data that support the hypothesis that tolerogenic dendritic cells produce large amounts of NO, and our data suggest that the tolerogenic function of DCs is dampened in the absence of NO. Interestingly, the in vitro stimulation of Tregs by CQ-DCsiNOS−/− suggests that NO-independent mechanisms must exist. Further studies must be conducted in order to characterize the extent of NO-mediated immune regulation. This work was supported by the São Paulo Research Foundation (FAPESP, grant nos. 2011/17965-3 and #2013/08194-9). RT received a FAPESP scholarship (grant no. 2014/02631-0).