The blood-feeding behaviour is present in several orders of insects that have acquired the genetic and morphofunctional resources to suck, digest and use the blood of their vertebrate hosts . Over millions of years of evolution, hematophagous mosquitoes have developed a complex pharmacological cocktail in their saliva, which clearly modulates host vascular and immune systems. Nevertheless, our knowledge of these processes is incomplete. The present study aimed to investigate the putative immunomodulatory effects of salivary components of A. aegypti mosquito vector on the differentiation, maturation and function of DCs and on the proliferation of T lymphocytes.
Pioneering work has shown that saliva of Rhipicephalus sanguineus, the “brown dog tick”, is able to inhibit differentiation and maturation of murine DCs . In addition, Sá-Nunes et al. (2007) were the first to isolate and characterize prostaglandin E2 (PGE2) as the major DC modulator in saliva of Ixodes scapularis ticks, the Lyme disease vector . Recently, PGE2 found in the saliva of Dermacentor variabilis ticks was also shown to regulate macrophage migration and cytokine production by these cells . In addition, the presence of PGE2 in R. sanguineus saliva was also demonstrated, although in smaller amounts than I. scapularis. However, the capacity of R. sanguineus saliva to modulate DCs is complemented by the presence of adenosine . Additional studies have demonstrated the immunoregulatory and anti-inflammatory activity of crude tick saliva [40, 41] and other proteinaceous components capable of modulating the function of DCs, such as Salp15  and sialostatin L , both identified in the I. scapularis saliva. Although these previous pieces of evidence show a clear effect of the tick saliva on DCs, very little is known about the modulation of these cells by saliva of blood feeding insects. Costa et al. (2004) demonstrated that SGE of Lutzomyia longipalpis sandflies, one of the leishmaniasis vectors in the new world, affects cytokine production and costimulatory activity of human DCs . Some years later, it was shown that SGE of P. duboscqi and P. papatasi sandfly species induced the production of PGE2 and IL-10 by DCs . The observed effects of P. papatasi SGE on DCs was due to the presence of adenosine and adenosine monophosphate (5’ AMP) and this seems to be, at least partially, the mechanism by which the SGE of this species was able to decrease the arthritis symptoms in an autoimmune model induced by collagen .
DCs comprise distinct developmental and functional subsets present in lymphoid and non-lymphoid tissues and are involved in the activation of adaptive immune responses, but also in tolerance to self-antigens . However, their frequencies in the tissues limit their experimental use. For example, Langerhans cells (the DC population from the epidermal layer of the skin) account for 3-5% of epidermal cells . Accordingly, classical DCs such as those found in the dermis, represent 1-5% of total cells from peripheral tissues . In addition, the increasing number of DC phenotypes described and isolation protocols employing enzymatic digestion which temporarily destroy surface markers, are other factors to consider . Thus, although the BMDCs preparations employed in the present work do not precisely represent the population of epidermal and dermal DCs that possibly interact with mosquito saliva, the use of these cells to investigate the biological effects of salivary preparations or their purified components is accepted by most studies in the field [9, 24, 27, 37, 44, 47]. To our surprise, A. aegypti SGE had no effect on DCs differentiation (Figure 1), maturation (Figure 2) and antigen presentation to T lymphocytes (Figure 3A). Corroborating with this data, it has been demonstrated that A. aegypti SGE did not affect the viability of a murine DC line . These results contrast greatly with data described in other species of arthropod vectors and, in the specific case of A. aegypti, our results are original in demonstrating that direct modulation of DCs by salivary components does not seem to be a property of the saliva from this mosquito species.
Interestingly, addition of A. aegypti SGE to cultures of CD11c+ cells after washing caused a significant inhibition in antigen specific (Figure 3B) and polyclonal (Figure 3C) proliferation of CD4+ T lymphocytes. This data confirms that SGE acts directly on T lymphocytes and not on antigen-presenting cells, and corroborates with findings in the literature showing the negative modulation of lymphocyte function by A. aegypti salivary components [16–18]. We also observed that SGE of other insect species (namely An. aquasalis, and P. duboscqi) had no inhibitory effect on T cell proliferation (Figures 3E and 3F). Wanasen et al.  also observed the absence of effects on T cell proliferation employing SGE of Culex quinquefasciatus, which belongs to the same mosquito subfamily. To our knowledge, such inhibitory activity was only found in the SGE of another nematoceran species, the black fly Simulium vittatum.
We have also explored the mechanisms by which the A. aegypti SGE inhibits lymphocyte proliferation. Our results show that this inhibition occurs due to induction of apoptosis in spleen cells, more specifically on T (CD4+ and CD8+) and B (CD19+) cells (Figure 4A). The specificity of such biological activity is evidenced by DC assays, since differentiation, maturation and function were not affected, even when these cells were incubated with 40 μg/mL SGE, a 4-fold increase in the maximum concentration effect on the proliferative response. As previously reported, decreased T cell proliferation induced by A. aegypti SGE was due to diminished cell viability, as evaluated by PI and 7-AAD expression, both DNA markers [16, 18]. It is important to emphasize that these markers are not specific for apoptotic cell death. Our findings unveiled that A. aegypti SGE induces apoptosis in T and B lymphocytes as assessed by exposure of phosphatidylserine (labeled with annexin V) at the surface of these cells (Figure 4A). Furthermore, our results suggest that caspase-3 (an executor caspase) and caspase-8 (an initiator caspase), but not Bim (a proapoptotic member of the intrinsic pathway), are likely to be involved in the apoptosis signaling induced by A. aegypti SGE (Figures 4B and 4C). As T and B cells are components of the adaptive immune response, it is reasonable to imagine that their effector functions (antibody secretion, cytotoxic granules or helper activities) are somehow deleterious to the mosquito life cycle. In fact, a classical study has demonstrated a decrease in the fecundity of mosquitoes fed on rabbits or guinea pigs immunized with a A. aegypti whole body homogenate .
Memory T cells are known to be more resistant to apoptosis than naïve T cells due to the increased expression of anti-apoptotic proteins . Because cells from Bim−/− mice are as susceptible to apoptosis as cells from Bim+/− mice (Figure 4C), and it is well known that this differential resistance is considerably dependent on the neutralization of Bim-mediated apoptosis by increased levels of Bcl-2 [29, 30], we investigated whether the A. aegypti salivary components would also have an effect on memory cells. Initially, mice were sensitized with A. aegypti bites and the effect of SGE on proliferative response of spleen cells from these animals was evaluated. Unlike naïve spleen cells (Figure 5A), those derived from sensitized animals proliferate in the presence of SGE (Figure 5B). In addition, when these cells were stimulated with Con A, only partial inhibition was achieved, suggesting that the memory cells present in the culture continue to proliferate even in the presence of SGE inhibitory factor(s). We rule out the role of neutralizing antibodies produced by B lymphocytes present in the culture on this effect, since spleen cell supernatants from sensitized mice, used as conditioned media, did not block the effect of SGE on T cells from non-sensitized mice (Additional file 3). Other explanations/mechanisms cannot be ruled out, such as peripheral tolerance and development of regulatory T cells, and will be explored in future work. We also demonstrated that SGE does not inhibit proliferation if memory cells are generated to other antigens. For example, spleen cells from An. aquasalis-sensitized mice proliferate when cultured with SGE of the same species, even in the presence of A. aegypti SGE. In addition, when these cells are co-cultured with Con A plus A. aegypti SGE, the inhibition of the proliferative response is only partial, reinforcing again our assumption that only naïve cells are affected by the presence of SGE in culture (Figure 5C).
This hypothesis is further supported by analyzing the proliferation of spleen cells from BALB/c mice receiving cells from a DO11.10 donor and subsequently immunized with OVA. When these cells are co-cultured with Con A plus SGE, the proliferation is partially inhibited (Figure 5B), whereas in cells from non-sensitized animals, SGE completely inhibits proliferation (Figure 5A). Remarkably, antigen-specific proliferation induced by OVA only occurs in sensitized mice, as expected, and it is not affected by A. aegypti SGE (Figure 5D and 5E). Regarding cell phenotype, SGE affects naïve cells (CD62LHIGH/CD44LOW) in both, control and immunized animals (Figure 5F and Figure 5G, respectively), while TCM cells (CD62LHIGH/CD44HIGH) and TEM cells (CD62LLOW/CD44HIGH) are proportionally more resistant in immunized animals (Figure 5G). So far, there is a single study showing a selective action of a salivary component in subpopulations of T cells. Such work demonstrated that Salp15, from the I. scapularis tick, binds to CD4 but not CD8 T cell co-receptor . However, this is the first time that arthropod saliva has been shown to discriminate between naïve and memory T cells.
The blood feeding strategies greatly diverge between many hematophagous arthropods. While hard ticks maintain prolonged contact with host skin, some others, like mosquitoes and sandflies, are transient feeders and leave their host in minutes or even seconds. Undoubtedly, their strategies to modulate the host vascular and immune system may vary as well.