Phenotypic characterization of Leishmania spp. causing cutaneous leishmaniasis in the lower Amazon region, western Pará state, Brazil, reveals a putative hybrid parasite, Leishmania (Viannia) guyanensis × Leishmania (Viannia) shawi shawi

We phenotypically characterized 43 leishmanial parasites from cutaneous leishmaniasis by isoenzyme electrophoresis and the indirect immunofluorescence antibody test (23 McAbs). Identifications revealed 11 (25.6%) strains of Leishmania (V.) braziliensis, 4 (9.3%) of L. (V.) shawi shawi, 7 (16.3%) of L. (V.) shawi santarensis, 6 (13.9%) of L. (V.) guyanensis and L. (V.) lainsoni, 2 (4.7%) of L. (L.) amazonensis, and 7 (16.3%) of a putative hybrid parasite, L. (V.) guyanensis/L. (V.) shawi shawi. McAbs detected three different serodemes of L. (V.) braziliensis: I-7, II-1, and III-3 strains. Among the strains of L. (V.) shawi we identified two populations: one (7 strains) expressing the B19 epitope that was previously considered to be species-specific for L. (V.) guyanensis. We have given this population sub-specific rank, naming it L. (V.) s. santarensis. The other one (4 strains) did not express the B19 epitope like the L. (V.) shawi reference strain, which we now designate as L. (V.) s. shawi. For the first time in the eastern Brazilian Amazon we register a putative hybrid parasite (7 strains), L. (V.) guyanensis/L. (V.) s. shawi, characterized by a new 6PGDH three-band profile at the level of L. (V.) guyanensis. Its PGM profile, however, was very similar to that of L. (V.) s. shawi. These results suggest that the lower Amazon region – western Pará state, Brazil, represents a biome where L. (V.) guyanensis and L. (V.) s. shawi exchange genetic information.


Introduction
American cutaneous leishmaniasis (ACL) is a parasitic protozoal disease widespread in most countries of Latin America, and is caused by a variety of Leishmania spp. within the subgenera Viannia and Leishmania [22,24]. In Amazonian Brazil, [21,48,49].
The description of most of these leishmanial parasites has been based on strains isolated either from human cutaneous disease (e.g., L. The aim of our present study is to phenotypically characterize 43 Leishmania spp. isolates from human cases of ACL from western Pará state using isoenzyme electrophoresis (6PGDH, PGM, G6PD, MPI, ASAT, and ALAT) and 23 Leishmaniaspecific monoclonal antibodies (McAbs). These methods have been used for more than 20 years by the Leishmaniasis Research Group of the Instituto Evandro Chagas (IEC) in Pará state, Brazil, in their studies on the taxonomy and ecoepidemiology of leishmanial parasites causing ACL in Amazonian Brazil [23, 25-27, 30, 31, 41-46].

Study area
Our study was carried out in the lower Amazon region of western Pará state, Brazil, that is identified as the lower Amazon mesoregion of Pará by the ''Instituto Brasileiro de Geografia e Estatística'' [16]. This mesoregion is composed of three microregions: Santarém, Ó bidos and Almeirim, and 14 municipalities (Fig. 1, Table 1). These three microregions are among the largest municipalities of Pará state and the population of Santarém is comparable with that of Marabá in southeast Pará and Belém in northeast Pará.
The mesoregion is a plain containing small hills with maximum altitudes of 100 m. The climate is typically equatorial, with average temperatures ranging from 24 to 26°C and high humidity. The annual rainfall is approximately 2500 mm, and its rainy season is from January to June. The vegetation consists of an immense rainforest, mainly primary, with inundated areas referred to as varzeas and igapós.

Leishmania spp. isolated from patients
The 43 isolates of Leishmania spp. were obtained from human cases of localized cutaneous leishmaniasis (LCL) [47] examined within two periods; the first one during 1990, 1996, and 1997 when our laboratory collaborated with the Health Secretary of Santarém, to improve their diagnosis of the disease in this municipality. In that period 21 isolates were collected; in 2001 a second batch of another 22 isolates was obtained, during a formal collaboration with that Health Secretary. Of these, 33 were from Santarém, 3 from Belterra, 1 from Prainha (on the southern bank of the Amazon River), 3 from Ó bidos, and single isolates from Alenquer, Monte Alegre and Almeirim (on the northern bank of the Amazon River).

Patients
All patients were examined at the Zoonosis Control Center, Health Secretary, Santarém, and submitted to parasitological diagnosis of the disease as follows: (a) For the detection of amastigotes, smears of exudates from the lesions were rapidly air-dried, fixed in absolute methyl alcohol and stained by Giemsa's method; during 1990, 1996, and 1997, a small volume (%50 lL) of these exudates was inoculated intradermally into the feet of hamsters for parasite isolation; (b) During 2001 triturated tissue from punch biopsies was inoculated into the feet of hamsters and cultivated in Difco B45 culture medium [54].

Ethical approval
This study was approved by the Ethics Committee in Human Research of the ''Núcleo de Medicina Tropical'' of the ''Universidade Federal do Pará'', Brazil, with the protocol number 22/2000 (ECHR/TMN/FUPa/Brazil). All patients examined within the 2001 period signed an informed consent form. The patients examined during the 1990, 1996 and 1997 periods did not sign the form as they were only submitted to routine proceedings for diagnosis of the disease.

Phenotypic characterization of Leishmania spp. isolated from patients
The phenotypic characterization of Leishmania spp. isolated from patients was based on the use of McAbs against the reference strains of Leishmania spp. from the Brazilian Amazon Region [15,41] and on the comparison of the isoenzyme electrophoretic profile and the zymodeme of each isolate with these Leishmania spp. [7,26,30,31]: The 6PGDH, PGM, G6PD, MPI, ASAT, and ALAT enzyme profiles of the strains isolated in the present study were prepared according to the methods described by Miles et al. [31]. They were also analyzed according to the position of the electrophoretic bands for the six enzymes. Each electrophoretic band was regarded as a separate character and was numbered from the most distal to the anodic point in each zymogram. Zymodemes were identified according to the pattern of the electrophoretic profiles for the six enzymes [7].  The 11 strains were classified into 3 serodemes: S -I (7 isolates), recognized by the McAbs B2, B5, B12, B18, N2, and L1; S -II (one isolate) that did not express the B5 epitope; S -III (3 isolates) that did not express the B2 epitope (Table 3).  (Table 3).  (Table 3).
Of these, 4 isolates reacted with the group-specific McAbs B2, B12, and L1, indicating they belonged to the Viannia subgenus and characterized as McAb profile IV.
A further 6 isolates reacted with McAb L1, that were characterized as McAb profile V.
The McAb profile of 7 other isolates was similar to profile IV but a variable number of parasites expressed the B12 epitope, which is typical of L. (V.) guyanensis, and was designated as profile VI.  (Table 3).
However, based on their McAb reaction profiles, these could be divided into two populations. One was represented by four isolates that did not express the B19 epitope, while the other one was composed of 7 isolates that expressed the B19 epitope. This is the first record of a population of L.  (Table 3). Thus, 91% are from the southern bank of the Amazon River.  (Table 3).

Discussion
Although a high occurrence of ACL is well known in the lower Amazon mesoregion of Pará state, there is very little information on the etiological agents of the disease in this area. This region is ecologically very interesting, being at a point where there was a land link between the northern and southern banks of the Amazon River, which is now an important biological barrier for some animal species.
In this study we identified four of the seven different Leishmania species that are incriminated as the etiological agents of ACL in Pará state, namely, L. The low prevalence of ACL due to L. (V.) naiffi in Pará state, Brazil [4] may be due to a number of epidemiological factors [21]. In the first place, and unlike most neotropical species of Leishmania, this parasite frequently produces no visible lesion in the skin of the hamster, although it may be re-isolated following the in vitro culture of triturated tissue from the site of inoculation. If similar occult infections are produced in some individuals the infection rate of L. (V.) naiffi in man may be much higher than previously thought. Secondly, the sand fly Psychodopygus ayrozai Barreto and Coutinho, 1940 is generally considered as the vector transmitting the parasite among the reservoir hosts, the armadillo Dasipus novemcinctus. This insect has been found naturally infected by L. (V.) naiffi [45] and is a very frequent occupant of armadillo burrows. In the Amazonian forest, however, Ps. ayrozai is not very anthropophilic, resulting in only sporadic transmission of L. (V.) naiffi to man by this species of sand fly. Only very rare infections have been recorded in the two highly anthropophilic sand flies, Ps. paraensis Costa Lima, 1941 and Ps. squamiventris Lutz and Neiva, 1912, in spite of the large number of these two insects dissected during studies on the epidemiology of ACL, and this suggests that they are not very attracted to armadillos.
L. (V.) lindenbergi, for unknown reasons, still seems to be restricted to the type locality in the municipality of Belém, Pará state [46] and contiguous municipalities such as Ananindeua, Marituba, Benevides, Santa Bárbara, and Santa Isabel [38].
The first readily distinguishable Leishmania species identified in this study was L. (V.) braziliensis, found in 25.6% of the isolates, and all reacted with the McAb B18, which is considered to be specific for this parasite, although a Colombian population of L. (V.) braziliensis did not express this epitope [39]. All the isolates were from the southern bank of the Amazon River, 10 from Santarém and 1 from Belterra. This high frequency confirms the importance of this parasite as an etiological agent of ACL in the lower Amazon mesoregion of Pará state.
Although isoenzyme analysis did not reveal any genetic heterogeneity, we identified three serodemes of L. (V.) braziliensis that have previously been recorded [40] in other regions of Amazonian Brazil. This finding confirmed the occurrence of intra-specific McAb variation of L. (V.) braziliensis with the dominance of the serodeme I (7) followed by serodemes II (1) and III (3).
The zymodeme of our 11 L. (V.) braziliensis isolates was identical to that of the L. (V.) braziliensis reference strain, indicating the presence of a homogeneous population lacking any enzymatic polymorphism in our study area. A number of enzymatic studies in other regions have led to the opinion that L. (V.) braziliensis is polymorphic [5,10,35,36,53]. The lack of polymorphism in our study is perhaps due to the fact that only three species of phlebotomine sand flies have been incriminated as the vectors of L. (V.) braziliensis in Pará state; the two closely related species Ps. wellcomei Fraiha, Shaw and Lainson, 1971 and Ps. complexus Mangabeira Filho, 1941, and Ps. davisi Root, 1934 [27,51,52]. The reservoirs of this parasite have so far not been identified in this region and the vector/reservoir contact may favor genetic stability rather than genetic variability. The crucial role in the identification of L. (V.) braziliensis of the enzymes 6PGDH and PGM must be emphasized, especially that of the former, which is considered the best enzymatic marker for the characterization of Leishmania species [2,8,13,17,19,37].
The second most frequent Leishmania in our study was L. (V.) shawi. This species was originally described in 1989 [28] from parasites isolated from wild animals captured in the Y.L. Jennings et al.: Parasite 2014, 21, 39 levels of enzymatic homogeneity, we found high levels of epitope variability with monoclonal antibodies. We were surprised to find that the monoclonal antibody profile of one of the L. (V.) shawi populations is identical to that of L. (V.) guyanensis, while that of the other is more similar to that of L. (V.) braziliensis. This leads us to suggest that the former could be due to genetic exchange at some period between L. (V.) guyanensis and L. (V.) shawi when the two populations reencountered each other in this unique bridging zone between northern and southern Amazonia.
In discussing the origin of Amazonian species, Haffer [14] drew attention to the clear separation of closely related avian species north and south of the Amazon River. He also emphasized that species are continually being separated and later reconnected due to climatic and paleogeological changes. The area between Manaus, Amazonas state, and Ó bidos, Pará state, was joined for long periods during the Tertiary period and possibly later and allowed the exchange of fauna between the Guinian Shield and the southern Brazilian Shield. Up until the late Miocene this ''bridge'' was open and separated the upper and lower Amazon Basins [34]. Such a time scale is well within the proposed phylogenies of the American Leishmania.
Our finding of putative hybrids between L. (V.) guyanensis and L. (V.) s. shawi and sub-speciation of L. (V.) shawi in this area adds more weight to the importance of this region as an ecological bridging zone between northern and southern Amazonia, as has previously been suggested.