Leishmaniasis is considered by the World Health Organization as one of the six major infectious diseases in the whole world  and affects over 1.5 million people every year worldwide. In Brazil, the majority of visceral and tegumentary cases are due to L. infantum and L. braziliensis, respectively. Parasite glycoconjugates have long being incriminated in a variety of events during parasite-host interactions modulating important host cell functions [11, 22, 34, 35]. Among these glycoconjugates, LPG is the best studied especially in Old World species of Leishmania. In this study, the role of the LPGs, from two epidemiologic important Leishmania species in Brazil, in interfering in signaling mechanisms was assessed in murine macrophages.
Since it has been reported that there is no significant difference between procyclic and metacyclic L. major LPG NK cell activation  and that the conserved GPI anchor is important for LPG activity  this work used stationary phase LPG. Although metacyclic promastigotes can be readily prepared from culture by several methods, only a small percentage of parasites, less than 5% , differentiate into these forms. For this reason, all experiments described here could not have been done with metacyclic LPG.
One of the most important events in the initial steps of Leishmania infection is the production of NO by macrophages. In many models, its production is dependent on a combination of IFN-γ and TNF-α via TLR-dependent mechanisms . Our results indicated that LPG from both species could induce the production of NO in IFN-γ-primed macrophages. Its production was higher in macrophages stimulated with L. braziliensis LPG than that of L. infantum. C57BL/6 macrophages incubated with both LPGs showed a higher production of NO, IL-1β and IL-6 than BALB/c. On the other hand, TNF-α production after stimulation by L. braziliensis LPG was higher in BALB/c macrophages (Figures 2, 3, 4). Similar results were also observed for GIPLs , these differences of activation between C57BL/6 and BALB/C mice may be due to the genetic background of mouse strains [37, 39]. In the present study, no macrophage activation was observed in WT, TLR2−/− and TLR4−/− murine cells after incubation with live promastigotes.
Previous reports have shown that in vivo, pro-inflammatory cytokines such as IL-1β, IL-6 and TNF-α, as well as chemokines, are induced in the initial events of L. major and L. donovani, causative agents of cutaneous and visceral leishmaniasis infection in the Old World, respectively. Similar results were observed here using L. braziliensis and L. infantum LPGs since macrophages stimulated with L. braziliensis LPG exhibited higher cytokine and NO production compared to that of the visceral form L. infantum (Figures 2 and 3). This finding was confirmed after incubation of the LPGs with CHO cells demonstrating that L. braziliensis LPG was able to induce NF-kB translocation. These data reflect a similar stimulation pattern between Old World and New World species that causes similar disease outcomes. More importantly, the lack of IL-10, IL-12 production, persistent MAPKs activation and the lack of NF-kB translocation via TLR4 ensure that no traces of endotoxins were present in our preparations.
Because most Leishmania glycoconjugates are on the external surface of the cell plasma membrane or secreted, they are able to modulate important functions in cell biology . The interspecies variations observed in L. infantum and L. braziliensis may be dependent on the action and specificity of glycosyltransferases [23, 24]. For example, in L. donovani (MONGI strain) critical glycosyltransferases are down regulated in the metacyclic phase . Such intra and interspecies variability is likely to have implications in antigenicity enabling carbohydrates to be important sources of biological diversity . In this work, the differential pattern of macrophage activation might be due to carbohydrate polymorphisms in the LPG of these two species. Our results with the two New World species of Leishmania are consistent with the reports from many Old World Leishmania species and strains which showed that LPG with its varied structural polymorphisms induced different levels of NO and TNF-α in murine macrophages [20, 21, 34].
No IFN-gamma, IL-10 or IL-12 production was observed by cells stimulated with LPGs from both New World species. Similar observations can be made in the human visceral leishmaniasis where immune suppression and a mixed Th1/Th2 profile modulate most of the immune response . The lack of IL-12 production by cells stimulated with Leishmania GPI-anchored glycoconjugates was also observed elsewhere, where mouse peritoneal macrophages failed to produce IL-12 when co-incubated with L. braziliensis or L. infantum GIPLs and when also stimulated with IFN-γ or LPS . It is also important to note that the lack of IL-12 production was not due to IL-10 release, since we did not observe any production of this cytokine (Figure 2A). This is similar to that observed when macrophages were treated with synthetic L. major LPG  and L. braziliensis and L. infantum GIPLs .
In the present work we also evaluated the role of TLRs on the recognition and signaling of LPG. TLR2 was first incriminated as the LPG receptor in macrophages and NK cells [34, 42]. Additional in vivo experiments demonstrated the importance of TLR3, TLR4 and TLR9 in different Leishmania species . By using RNA interference methodologies, it was shown that both TLR2 and TLR3 were implicated in the recognition of L. donovani LPG in IFN-γ primed macrophages . In vivo, it was shown that TLR4 deficient mice are more susceptible to Leishmania infection, failing to efficiently resolve the lesions  while TLR2 shows a more regulatory role in L. braziliensis-infected dendritic cells . Here, we demonstrated in vitro with macrophages and CHO cells that TLR2 and, to a lesser extent, TLR4 were recognized by LPGs from both species clearly suggesting their participation in the LPG signalling. The inability of L. infantum LPG to activate NF-kB and ERK could be suggested as an evasion mechanism compared to L. braziliensis LPG.
Given that the LPGs were able to induce NO and cytokine production, we investigated whether activation of MAPK signalling was affected. In contrast to GIPLs (which fail to activate MAPKs) , LPG from both species activated MAPKs, but with different kinetics. L. infantum LPG was not able to activate ERK1/2. In contrast, L. braziliensis LPG strongly activated MAPK activity after 15 min. Interestingly, p38 and JNK activation exhibited a gradual and transient profile in L. infantum and L. braziliensis, respectively (Figure 5). Although a punctual MAPK activation was observed for L. braziliensis LPG after 5 min, this may not be a sufficient stimulus for IL-12 production. Activation of p38 appears to be important for controlling Leishmania infection since anisomycin reduced parasite survival upon p38 activation . Feng et al.  reported that ERK1/2 and p38 are important for NO and TNF-α production by macrophages stimulated with LPS. Consistent with those observations, our data suggests that the LPGs from New World species are also able to differentially activate MAPKs.
The data presented here demonstrated a differential production of NO, cytokines and MAPK activation profile by L. braziliensis and L. infantum LPG stimulated macrophages. The LPG from these species have a limited proinflammatory potential since they fail to activate important proinflammatory cytokines such as IL-12 and only activate low amounts of IL-1β while inducing activation of early inflammatory cytokines such as TNF-α and IL-6. Additionally, as seen in Figure 7, preincubation with LPG prior to stimulation with LPS reduced the nitrite production to basal levels, indicating that the dynamics of infection must be well regulated and consistent with the long-known proposal that LPG acts as a multifunctional virulence factor for Leishmania.