Several methods are used to identify eating habits in phlebotomines, such as immunological methods, through the precipitin technique [4, 8, 11, 28–30] and immunoenzymatic tests [7, 9, 15, 31–32]; and recently molecular methods using PCR [33–42]. Studies that seek to determine the dietary pattern of phlebotomines are of great ecological and epidemiological relevance, since they may at the same time increase the knowledge about the habits of this vector, but mainly suggest potential reservoirs of Leishmania spp., which can help in the planning of better strategies [11, 33, 37, 40].
The literature has recorded, on the basis of field work, the eclecticism of L. (L.) longipalpis in feeding on a wide range of mammals, including dogs, pigs, horses, cattle, and birds [17, 43]. In addition to the favorable environmental conditions, the abundance of food sources is a determinant factor in the population growth of the vector, especially in periurban areas, which brings it closer to humans.
The data from the present study revealed that the specimens of the populations of Araguaína, Fortaleza and Sobral, areas that were classified as intensive transmission for AVL, fed mainly on humans. The anthropophilia of L. (L.) longipalpis has already been described in several studies [12, 17, 43], being one of the essential criteria for the species to be incriminated as a vector [44]. In Ceará we have already described the attraction of L. (L.) longipalpis by humans, including individuals and populations of the vector originated from Massapê, through the ELISA test proving the ingestion of human blood [15].
Research in Campo Grande (Mato Grosso do Sul) and Marajó Island (Pará) shows the high attraction of the vector by humans [7, 14]. However, investigations in Colombia and Araçatuba (São Paulo) have suggested low anthropophilia [28, 29]. The anthropophilia of L. (L.) longipalpis observed in the populations of Araguaína, Fortaleza and Sobral, certainly plays a relevant role in the classification of intense transmission to AVL, which is attributed to these municipalities. On the other hand, the data so far do not suggest that the populations of the different biomes (Cerrado and Caatinga) have different habits regarding the attraction to humans. The fact is that the species is highly anthropophilic, but due to its opportunistic character, characteristic of some species of phlebotomines, one can not expect the same pattern of feeding.
In Araguaina, the feeding of the analyzed specimens showed the attraction to dogs, as observed in the populations of, Fortaleza and Sobral, which are areas that present a high incidence of canine visceral leishmaniasis. In the 1950s, in Sobral [43], dogs had already been suggested as important domestic reservoirs of L. (L.) infantum chagasi and as sources of infection for L. (L.) longipalpis, in the home environment; in Araçatuba (São Paulo), the highest percentage of canine blood female feeding was observed, correlating its epidemiological role in the AVL transmission chain, with the hypothesis that canine enzootia precedes human transmission [30]. In other countries of South America, such as Colombia, in an area endemic to AVL, the population of L. (L.) longipalpis showed a strong attraction to dogs [28]. Studys in Montes Claros (Minas Gerais) shows that even with the adoption of euthanasia for seropositive AVL dogs, the canine population will be continue serving as a source of Leishmania sp. infection for sand flies [45]. The data obtained in this study confirm the importance of dogs in the maintenance of the AVL transmission.
However, it is possible to observe a weak attraction of L. (L.) longipalpis to birds, different from the analyzes that demonstrate opposite behavior, in Northeast areas [15] and in Raposo (Maranhão) [14]. It is well known that coops near houses act as attractive for phlebotomines [43], which serve as breeding grounds, since immature forms develop in soil that is rich in organic matter. In this context, the contact between vectors and humans increases, besides allowing the domiciliation of the vector, and consequently the transmission of AVL within the home, a situation that is observed both in the rural area and in the periphery of the cities [12, 17, 27]. In terms of environmental determinants of occurrence of AVL transmission in the home environment, the maintenance of coops near homes increases the risk of human exposure to the phlebotomine vector [40].
The analyses of the present study did not reveal positivity of L. (L.) longipalpis to opossum blood, a synanthropic animal, which was suggested as a reservoir of L. (L.) infantum chagasi, as already observed in populations of the vector from Sobral (Ceará ), Massapê (Ceará) and Jequié (Bahia), Brazil [15] and Colombia [28]. Some studies have reported the natural infection of Didelphis marsupialis, by Leishmania spp. possibly L. (L.) infantum chagasi, thus discussing the role of these mammals as potential reservoirs for AVL [2, 46, 47].
Literature has demonstrated that L. (L.) longipalpis is an eclectic, and opportunistic, species in the search for food sources [7, 12, 14, 15, 28–30]. Thus, the non-positivity in the general context of this study, about 71% of the analyzed specimens, may have a direct relation with the chosen targets. In the catch areas, besides the sources tested (humans, dogs, birds and opossums), it was possible to observe other potential sources of food, such as pigs, cats, oxen, horses, rodents, goats, sheep and foxes. Another possibility is the viability of DNA. Studies indicate that molecular assays are able to detect host DNA only within 5 days after the blood supply [33, 34, 40].
In relation to the other females that were fed, where it was not possible to detect the food source, probably other animals may be involved in the maintenance of the local population of L. (L.) longipalpis. Rodents must be taken into account, which are frequently observed during catches in areas of work and in the home environment. Rattus norvegivus and Rattus Rattus were found to be infected with L. infantum (= L. (infantum chagasi)), with analysis using Leishmania Nested PCR and hsp70 PCR-RFLP, suggesting the possibility of participation of this rodent in the zoonotic cycle of the LVA [48, 49].
From the analyses performed with the few (13) copies from Rio de Janeiro (RJ), one specimen was detected with human blood DNA. This fact corroborates with the findings that already indicated the local transmission, demonstrating the first urban focus of the city in the Bairro do Caju; the autochthonous human case, canine cases due to visceral leishmaniasis and vector findings (L. (L.) longipalpis) [50–53].
New studies should be done, seeking to better understand the ecoepidemiological dynamics of LVA in the city of Rio de Janeiro. The literature suggests that this recent (and first) urban focus has been installed from passive vector dispersion, due to the transfer of used soil to the flower pots of the Caju Cemetery, from presence of L. (L.) longipalpis vector [54]. On the other hand, it can be assumed that in the area there was already the presence of the vector, a population that was in equilibrium and well adapted to the dens of rodents, which are very common in the cemetery. The canine transmission occurred by the introduction of an infected animal, because in the area adjacent to the cemetery there was a kennel, which triggered the local transmission to a considerable number of dogs.
The rate of naturally infected phlebotomines in endemic areas as well as the correct identification of the Leishmania species are of great importance for a better understanding of the epidemiology of leishmaniasis. The classic methodology, used until then, consists of the direct search of the parasite by microscopy, through dissection of the digestive tract, followed by microscopic analysis, isolation and culture of parasites, with subsequent identification/diagnosis of the species [55–58]. Although it is considered a "gold standard" technique, in some cases it is very laborious and unfeasible, given the amount of material to be analyzed, time and specific training for this process, sometimes rendering the technique imprecise and not always allowing the isolation and the correct identification of the parasite.
Diagnostic assays using PCR have been developed based on different molecular targets such as ribosome minor subunit genes, mini-exon gene spacer sequences, single-copy nuclear sequences such as the DNA polymerase alpha gene, but the minicircle kDNA are the ones that present the highest sensitivity for detection of Leishmania [22, 58–63].
In the present study, the molecular analyses were performed individually, since the real scenario of infection rate and of identification of the food source was searched. When working with a female pool, some information may be lost as it would not be possible to detect the exact amount of positive females (for infection or food source analysis). In addition, we sought to associate the two approaches, food source and natural infection with Leishmania sp, in view of this association, individual analysis became mandatory. When parasite load analyses are performed on phlebotomines, one should consider the possibility that parasitic load may be different in each infected female, and the level of individual infection may interfere with the total pool load. Thus, if a sample showed a high parasitic load, this may be due to one or more infected phlebotomines [64]. In Camaçari (Salvador) was performed a compared study in diferents periods of captures from L. longipalpis, the results of parasite load was low and did not vary regardless of the season, despite the number of collected sand flies [65].
In this study, it was possible to detect the natural infection of L. (L.) longipalpis by L. (L.) infantum chagasi in Araguaína and Sobral, which are areas of intense transmission of AVL, with rates of 4.9% and 9, 8% respectively. In Fortaleza, a minimum natural infection rate of 3.7% was detected [61]. These results reinforce the practicality of the use of molecular techniques in the identification of natural infection, since it is known that infection rates are low in nature, even in areas endemic for AVL.
It was also possible to carry out the association between food identification and the detection of natural infection by L. (L.) infantum chagasi in 5 specimens; four specimens of L. (L.) longipalpis performed blood meal in human, and one in dog, the latter with its role already widely discussed in the transmission of AVL [2, 12, 43]. The non-identification of the source of the food source in a female L. (L.) longipalpis, from Sobral, corroborates the idea that, possibly, other mammals (in this untested study) can serve as food sources for well-known eclectic phlebotomines.