Amaral et al. revealed that sera levels of RvD1 in BT and LL leprosy patients were similar, but increased in comparison with the sera of healthy individuals (17). Interestingly, after MDT serum levels of RvD1 in BT and LL patients were reduced to those of healthy controls (17). These data indicated that RvD1 is being produced in response to inflammation and possibly also associated with the presence of the pathogen or pathogen products. However, induction of RvD1 production via M. leprae infection has not been investigated.

A comprehensive study to define the biological activity of the D-series resolvins (RvD1 and RvD2) and maresin-1 on the adaptive immune response demonstrated that these SPMs reduce the production of IFN-γ and IL-17 by Th1 and Th17 cells, respectively (26). Moreover, RvD1 was shown to promote the de novo generation of FoxP3⁺ Treg cells, the expression of CTLA-4 (a surface marker of Treg cells) and IL-10 secretion. The similar levels of RvD1 in BT and LL patients, does not correlate well with this laboratory assessment of RvD1 activity, since BT patients present a strong Th1 and Th17 responses (3, 4, 6) and LL patients are characterized by T-cell anergy and increased frequency of Treg cells (4, 6). Nevertheless, it would be premature to conclude that RvD1 does not participate in the dichotomous immune responses of TT/BT and BL/LL patients. It is possible that the higher level of RvD1 down-modulates the Th1 immune response in TT/BT as well as BL/LL patients. Martins et al. demonstrated that peripheral mononuclear cells (PBMCs) from paucibacillary (TT/BT) leprosy patients possess a lower capacity to produce IFN-γ than healthy individuals exposed to M. leprae (3). Thus, the adaptive immune response in TT/BT individuals is still reduced as compared to healthy controls. Furthermore, it could be that RvD1 activity is related to the level of expression of its cognate receptors, GPR32 and ALX/FPR2. Thus, studies that assess the presence of these receptors in the T cells of TT/BT and BL/LL patients are required to fully understand the potential influence of RvD1 on the adaptive immune response across the spectrum of leprosy. Polymorphisms in the promoter region of the ALX/FPR2 gene resulting in a reduced expression of this receptor are known (100, 101). Thus, it would also be interesting to investigate whether polymorphisms exist between TT/BT and BL/LL patients in the promoter or functional regions of the GPR32 and ALX/FPR2 genes.

Besides the ability to reduce the activity of Th1 and Th17 cells, RvD1 also controls the activity of macrophages (102, 103). RvD1 induces efferocytosis in monocytes/macrophages (73), a process that engulfs apoptotic cells and is reported to play an important role in the clearance of Mycobacterium tuberculosis and Mycobacterium avium (104, 105). However, De Oliveira and colleagues indicated that this process might promote the persistence of M. leprae (106). Specifically, in the presence of M. leprae, efferocytosis alters the phenotype of the pro-inflammatory M1 macrophage toward anti-inflammatory M2 phenotype with increased the uptake and survival of M. leprae. Therefore, in paucibacillary patients, where apoptotic bodies are present in higher number (107, 108), efferocytosis may play an important role in the in vivo persistence of M. leprae. The increased levels of RvD1 in TT/BT patients could help drive this process (Figure 3).

Adding to the immunomodulatory activity of efferocytosis, it is recognized that M. leprae inhibits the capacity of macrophage to respond to IFN-γ stimulation (47) and impairs the production of pro-inflammatory cytokines (e.g., IL-6 and TNF-α) (109). Macrophages infected with M. leprae have been found to preferentially prime Treg cells over Th1 or cytotoxic T cells (110). Thus, RvD1 may have an additive or synergistic effect on macrophage function that further reduces the innate responses against M. leprae and consequently allows the survival of the pathogen in leprosy patients with a robust Th1 and Th17 cells response (Figure 3). However, studies are required to determine whether RvD1 preferentially drives the response of M. leprae-infected macrophage, as well as enhancement of M. leprae uptake in the context of efferocytosis. While we would hypothesize that RvD1 would influence macrophage polarization in the context of M. leprae infection, the involvement of other lipid mediators in this process cannot be excluded.

T1R is a major complication in borderline leprosy patients (BT, BB, and BL) and occurs before, during and after MDT (111). The increased inflammation of T1R driven by Th1 and Th17 cells in skin lesions and/or nerves can result in permanent loss of nerve function (112, 113).

A higher bacillary load and MDT are factors associated with the development of T1R pathology (114–116). Thus, it has been hypothesized that the release of M. leprae antigens promoted by MDT drive an enhanced immune-inflammatory response, especially in multi-bacillary patients (116, 117). Interestingly, the levels of RvD1 in leprosy patients decrease after the conclusion of MDT (17). Thus, a reduction in circulating SPM may remove suppressive activity being placed on Th1/Th17 cells and contribute to susceptibility of developing T1R in the presence of M. leprae antigens (Figure 3). Recently, a metabolomics study of sera from leprosy patients with and without T1R, and that had not started MDT, confirmed that the level of RvD1 was significantly increased (9.01-fold) in non-T1R leprosy patients as compared to T1R leprosy patients and healthy controls (18). These findings indicate a direct correlation with reduced RvD1 levels and destructive inflammation due to enhanced Th1/Th17 activity and revealed that reduced RvD1 production could occur during active disease.

As the balance of pro-inflammatory and pro-resolving lipid mediators are important in the development and control of inflammation, it is important to note that RvD1 also down-regulates the production of the pro-inflammatory lipid mediator LTB₄ (68). LTB₄ promotes chemotaxis of Th1 (32) and Th17 cells (33) and enhances the production of pro-inflammatory cytokines associated with Th1 responses (TNF-α and IFN-γ) (22). Although the concentration of LTB₄ in BT and LL patients are similar to healthy individuals (17), Silva and colleagues observed a significantly increased level of serum LTB₄ during T1R (18). Studies to define the mechanisms of RvD1 activity revealed that this SPM inhibits the translocation of 5-LO to the nucleus and this inhibits the synthesis of LTB₄ (68). This mechanism would explain why the levels of LTB₄ were not increased in leprosy patients without T1R, but with a reduction of RvD1, they become elevated in T1R patients. However, it does not explain why the levels of LTB₄ did not increase after MDT in leprosy patients without T1R since this treatment reduced RvD1 concentrations (17). It is possible that therapeutic elimination of infection reduces signals and stimuli leading to LTB₄ production, as well as those that drive RvD1 production.

In conclusion, although increased RvD1 levels may favor M. leprae infection by modulating the protective innate and adaptive immune responses (i.e., bad with it), at the same time, RvD1 is likely important to avoid exacerbated inflammation that may cause skin and nerve injuries. Once the levels of the RvD1 drop in a leprosy patient (e.g., because of MDT or other factors), we hypothesize that this increases susceptibility to pathogenic Th1 and Th17 responses against M. leprae antigens (i.e., worse without it).

The study of Amaral et al. demonstrated that LXA₄ is increased in leprosy patients (17). However, the biological function of LXA₄ in M. leprae infection is not well understood, but has been studied in M. tuberculosis infection, another model of chronic infectious disease. In the murine model of tuberculosis, Bafica et al. showed that after 1 week of M. tuberculosis infection, LTB₄ and LXA₄ increase in abundance as compared to uninfected animals, but the levels of LTB₄ decrease after 10 days while those of LXA₄ persist during chronic M. tuberculosis infection (20). Interestingly, mice deficient for 5-LO (5-lo^−/−) did not produce LXA₄ increasing the resistance against M. tuberculosis due to higher production of Th1-derived cytokines (INF-γ and IL-12). Conversely, the 5-lo^−/− mice treated with a LXA₄ analog reduce the levels of Th1 cytokines resulting in increased susceptibility to M. tuberculosis (20). These results indicate that LXA₄ has a more predominant effect than LTB₄ during M. tuberculosis infection and that a high LXA₄ favors the mycobacterial infection. Similar to the animal studies with M. tuberculosis, infection of humans by M. leprae and the presentation of leprosy, are associated with increased levels of LXA₄, but not LTB₄ (17). This likely reflects the capacity of an M. leprae infection to pass unnoticed for years (1–10 years), presumably due to a protective and non-pathogenic immune response. However, as observed for household contacts, a gradual increase in bacillary load and continuous exposure to antigen, down-modulates the immune response against M. leprae (3, 118). Thus, we hypothesize that the reduced capacity of the host to respond to M. leprae, even during an increase in the bacillary load, is exacerbated by a higher production of LXA₄. Once this SPM and RvD1 are produced in sufficient amounts they would inhibit the production of LTB₄ (68), and thus elevated levels of LXA₄, together with RvD1, might favor the chronic infection of M. leprae.

It is suggested that LTB₄ and LXA₄ modulate the expression or the effects of TNF-α, a pro-inflammatory cytokine involved with the resistance/susceptibility to leprosy (21, 22, 119). Moreover, an imbalance in the ratio of the pro-resolving LXA₄ to pro-inflammatory LTB₄ (LXA₄/LTB₄) is related with a poor control of the immune-inflammatory response in humans (120, 121). Collectively, metabolomics data produced with sera of leprosy patients indicate that the balance between LXA₄ and LTB₄ is altered (17, 18). However, the mechanisms by which altered ratios of LXA₄/LTB₄ affect the immunopathology of leprosy remain undefined.

Previous works from Tobin et al. demonstrated that the LXA₄/LTB₄ ratio was an important factor in susceptibility of zebrafish larvae to Mycobacterium marinum, due to the modulation of TNF-α expression (21, 88, 89). Specifically, shunting LTA₄ into LXA₄ synthesis resulted in an increase in the LXA₄/LTB₄ ratio and consequently a down-modulation of TNF-α expression (21, 88, 89). This culminated in a high bacterial burden, death of infected macrophages and increase in the severity of the disease. In contrast, accumulation of LTB₄ enhanced TNF-α expression and enabled macrophage control of infection, but an excess of TNF-α results in the necrosis of macrophages and a higher burden of infection (88, 89). Previous findings support a correlation between the levels of TNF-α and LXA₄/LTB₄ ratio in leprosy patients. Both paucibacillary and multi-bacillary leprosy patients exhibited similar levels of TNF-α, LTB₄ and LXA₄ (11, 17, 122). On the other hand, leprosy patients with T1R possess a lower LXA₄/LTB₄ ratio (18), which agrees with increased inflammation and higher levels of TNF-α observed in these patients (123). Thus, the balance between pro-inflammatory and pro-resolving lipid mediators is important to the outcome of infection.

Furthermore, support for the importance of a LXA₄/LTB₄ balance is provided through population genetics in humans (21). Vietnamese and Nepali individuals homozygous for a common promoter polymorphism at the human LTA4H locus display lower protection against tuberculosis and multi-bacillary leprosy, respectively. This polymorphism is associated with deficient (low activity alleles) or excessive (high activity alleles) expression of the LTA4H gene. Conversely, heterozygous individuals displayed a moderated expression of LTA4H gene and consequently a more balanced production of LXA₄ and LTB₄, due to the presence of both a low-activity allele and a high-activity allele (21, 88). As a consequence, heterozygous LTA4H individuals exhibited better protection against mycobacteria infection.

The connection between LTA₄H and TNF-α is reciprocal, as TNF-α is able to modulate the expression of LTA4H (124–126). This suggests that the synthesis of TNF-α and the LXA₄/LTB₄ ratio could be regulated by a feedback loop generated by expression of TNFA and LTA4H (details in Figure 4). Interestingly, polymorphisms in the promoter region of the TNFA are associated with human susceptibility to leprosy (119, 127, 128).

Existing data strongly support the hypothesis that the LXA₄/LTB₄ ratio in leprosy disease is an important factor in regulation of TNF-α and hence the susceptibility or resistance to M. leprae infection. We hypothesize that an increase in the LXA₄/LTB₄ ratio leads to lower TNF-α secretion and reduced control of M. leprae replication (Figure 4). However, a decrease in LXA₄/LTB₄ ratio would promote higher TNFA expression and an intense inflammatory response as observed for leprosy patients with T1R.

PGE₂ and PGD₂ are increased in LL patients (17), and previous studies indicate that foamy macrophages/Schwann cells, a classical hallmark of LL patients, are the main source of prostaglandins (129, 130). The higher levels of PGE₂ in LL patients (17) together with the lower levels in T1R patients (18) suggest that PGE₂ is related to the different clinical forms of leprosy. Indeed, this lipid mediator impairs the proliferation of T cells (39, 40) and inhibits the activation of macrophages by IFN-γ in M. leprae infection (47). Thus, levels of PGE₂, produced by foamy macrophages/Schwann cells, can contribute to the inhibition of Th1 responses against M. leprae in LL patients. This may also indicate that lower levels of PGE₂ in T1R patients favors the exacerbated acute responses of Th1 cells. Moreover, PGE₂ has the ability to augment the suppressive capacity of human CD4⁺CD25⁺ Treg cells and up-regulate the expression of transcription factor FOXP3 (46). Garg and colleagues demonstrated that PGE₂, but not PGD₂, promotes the expansion of Treg cells during M. tuberculosis infection (45). Thus, the higher frequency of Treg cells, as well as the anergy of Th1 and Th17 cells in LL individuals, could be related with increased amounts of PGE₂ secreted by foamy macrophages/Schwann cells (Figure 5). Other mechanisms through which higher levels of PGE₂ might affect the differentiation of Th17 and Th1 cells in LL patients include, modulating the secretion of IL-23 by dendritic cells (Figure 5) (23) and impairment of IL-12 production by dendritic cells (19).

There is evidence that at the proper concentration and in the presence of a co-stimulatory signal, PGE₂ also stimulates Th1 response. Yao and colleagues showed that treatment of naive T cells with PGE₂ and antibody stimulation of CD28 induces the differentiation of Th1 cells (24, 44). It is well known that PGE₂, through interaction with EP2 and EP4, inhibits the differentiation of Th1 cells by increasing intracellular levels of cAMP (42, 43). However, with a concomitant stimulation of CD28, T cells are rescued from the inhibitory effects of cAMP and therefore differentiate to Th1 cells (24). Interestingly, M. leprae antigens are able to reduce the expression of B7-1 and CD28 molecules in PBMC cultures from healthy controls (131), and the levels of B7-1 and CD28 molecules in BL/LL patients, but not in BT patients, are reduced. Therefore, the higher levels of PGE₂ that leads to an increase in the intracellular levels of cAMP together with lower expression of CD28 could inhibit the differentiation of Th1 cells in LL patients. Conversely, BT patients that secrete basal levels of PGE₂ and express higher levels of CD28 would be expected to propagate and maintain a Th1 response. T1R patients also exhibit a basal level of PGE₂ (18). Hence, our hypothesis is that lower PGE₂ levels promote Th1 and Th17 cell activities in BT and T1R patients, but in LL patients, the higher concentration of this prostaglandin inhibits Th1 and Th17 responses (Figure 5). Together, these studies highlight the controversial role of PGE₂ in the human adaptive immune response and underscore the need for studies to determine other possible roles of PGE₂ in leprosy.

The prostaglandin PGE₂ has been shown to also interfere with the control of cell death (48) and the production of NO by phagocytic cells (41). Studies using an experimental animal model of pulmonary tuberculosis demonstrated that at the early phase of M. tuberculosis infection, BALB/c mice produce lower amounts of PGE₂ and this promotes the expression of the inducible form of NO synthase (iNOS). In contrast, at later stage of infection, higher amounts of PGE₂ are produced and inhibit the expression of iNOS (41). These assays support the idea that lower production of PGE₂ favors the bacterial control, and at higher concentrations, PGE₂ inhibits microbicidal mechanisms in the murine model. In line with these observations, skin lesions of BT leprosy patients exhibit a higher expression of iNOS than those of BL patients (11), and macrophages isolated from BT patients secrete higher concentrations of nitrite, a marker for iNOS activity, than macrophages derived from LL patients (132). Thus, we hypothesize that the lower levels of PGE₂ in BT patients (17) directly promote the microbicidal activities of phagocytic cells to control M. leprae replication as well as enhance the Th1 responses. Interestingly, the higher production of NO may cause nerve damage in BT patients as hypothesized in previous work (15). On the other hand, higher concentrations of PGE₂ secreted by foamy macrophages/Schwann cells would inhibit these same antimicrobial activities and thus favor multi-bacillary disease (Figure 5).

A potential mechanism by which PGE₂ would inhibit the production of NO in LL patients is through the induction of insulin-like growth factor I (IGF-I). PGE₂ induces the expression of IGF-I in murine macrophages (133) and osteoblasts (134, 135), and IGF-I inhibits the NOS2 pathway (136). A recent study has demonstrated that increased amounts of IGF-I are found in the skin lesions of LL patients and that IGF-I inhibits signaling cascades required for NO production (137). Therefore, it is possible that the elevated levels of PGE₂ could be linked to the inhibition of NO production via the induction of IGF-I in LL patients.

The production of IGF-I, possibly mediated by PGE₂, may also promote M. leprae survival by inhibition of apoptosis. Live M. leprae induces the production of IGF-I in Schwann cells and this was found to prevent apoptosis (138). The inhibition of apoptosis could be a significant advantage for M. leprae since this mechanism of cell death promotes the presentation of mycobacterial antigens to T cells (139). Thus, via an IGF-I network, PGE₂ may directly impact antigen presentation and favor M. leprae replication (Figure 5). However, a direct functional link between increased IGF-I and PGE₂ levels in LL individuals and apoptotic activity needs to be experimentally established.

It is interesting to highlight that the role of PGE₂ in M. leprae infection may greatly differ from the function of PGE₂ during M. tuberculosis infection. It appears that, during the early phase of infection, virulent M. tuberculosis (H37Rv) inhibits the synthesis of PGE₂, by inducing synthesis of LXA₄, to prevent apoptosis and consequently inhibit early T-cell activation and promote necrosis of macrophages (48, 49, 139, 140). In contrast, at the chronic stage, PGE₂ is highly produced (41), which could control the bacillary load by apoptosis. Furthermore, macrophages infected by the avirulent strain of M. tuberculosis (H37Ra) produced increased levels of PGE₂ (48), promoting the protection against mitochondrial inner membrane perturbation and induced plasma membrane repair, crucial processes to avoid necrosis and induce apoptosis (48, 49). Thus, PGE₂ might be crucial for the resistance against M. tuberculosis but promote susceptibility to M. leprae. These possible differences between M. tuberculosis and M. leprae infections could be partially related with different modulation of EP1-4 receptors by the two pathogens and should be explored in future studies.

Based on the several findings regarding PGD₂ and its effects on the modulation of T cells we suggest that PGD₂ production via foamy macrophages/Schwann cells promotes Th2 response in LL patients. It is well established that PGD₂ decreases the numbers of CD4⁺ and CD8⁺ T cells that produce IFN-γ and IL-2, through interactions with the DP1 receptor, while contributing to the Th2 responses with induction of IL-4, IL-5, and IL-13 by binding the CRTH2 receptor (60, 61). Besides a direct effect on T cells, PGD₂ modulates the T-cell response through dendritic cells and their production of IL-12 (19, 59). Braga et al. has revealed that monocyte-derived dendritic cells from LL patients produced less IL-12 (25), and although a direct association has not been made, the decreased IL-12 levels in LL patients could be driven by increased PGD₂ production and secretion by foamy macrophages/Schwann cells (Figure 5).

One observation that does not fit with the PGD₂ immune suppressing scenario in leprosy is that PGD₂ levels increase during a T1R (18). T1R is considered a delayed type hypersensitivity (DTH) reaction (141) and several works indicate that PGD₂, or its metabolite 15d-PGJ₂ (142), is highly produced during DTH to control the inflammatory activity in animal models (143). Thus, the increasing of PGD₂ in T1R patients may be a response by the host to control inflammation.

Individuals with acute inflammatory demyelinating polyneuropathy, an autoimmune disease that directly attack the peripheral nerve myelin (144), have increased levels of PGD synthase enzyme in their cerebrospinal fluid (145). In a murine model of spinal cord contusion injury, the levels of PGD synthase are also elevated (146). Interestingly, although the expression of PGD synthase was never determined, COX-2 is increased during T1R (147, 148). Thus, an increase in PGD₂ is not unexpected during T1R as these leprosy patients suffer the most severe nerve damage. PGD₂ is known to promote the myelination of neurons (55). In addition, mice that lack PGD synthase are unable to promote myelination of the neurons. These studies, as well as the fact that mast cells that are in close proximity to the peripheral nerve fibers in the tissue are the major producers of PGD₂, support the hypothesis that increased PGD₂ is a consequence of the T1R in leprosy and not a driver of the pathology.

Given the potentially varied activities of PGD₂ at different stages of leprosy, it is important to determine not only the source of this prostaglandin, foamy macrophages/Schwann cells versus mast cells, but also the receptors that bind PGD₂ during the different manifestations of leprosy and the cells that are expressing these receptors. Additionally, PGD₂ potentiates the formation of edema (56, 57), a factor that might contribute to the nerve damage in leprosy (149). Therefore, further studies are required to determine if PGD₂, through edema formation, can contribute to the pathology of leprosy lesions.