Hepatozoon caimani in Caiman crocodilus yacare ( Crocodylia , Alligatoridae ) from North Pantanal , Brazil Hepatozoon caimani em Caiman crocodilus yacare ( Crocodylia , Alligatoridae ) do Norte do Pantanal , Brasil

Hepatozoon species are the most common intracellular hemoparasite found in reptiles. Hepatozoon caimani, whose vectors are Culex mosquitoes, has been detected in a high prevalence among caimans in Brazil by blood smears examinations. The present work aimed to detect and characterize the Hepatozoon spp. found in 33 caimans (24 free-ranging and 9 captive; 28 males and 5 females) (Caiman crocodilus yacare) sampled at Poconé, North Pantanal, state of Mato Grosso, Brazil, using blood smears examinations and molecular techniques. Hepatozoon spp.-gametocytes were found in 70.8% (17/24) and 88.8% (8/9) of blood smears from free-ranging and captive caimans, respectively. Hepatozoon spp. 18S rRNA DNA was found in 79.2% (19/24) and 88.8% (8/9) of free-ranging and captive caimans, respectively. Comparative analysis of parasitized and non-parasitized erythrocytes showed that all analyzed features were significantly different (P<0.05) for both linear and area dimensions. Phylogenetic analysis based on 18S rRNA sequences grouped the Hepatozoon spp. sequences detected in the present study together with H. caimani, recently detected in caimans in southern Pantanal.

Culex (Meloconion) mosquitoes are the H. caimani natural vectors, where sporogonic phase takes place (LAINSON et al., 2003;VIANA et al., 2010b).Despite of that, the possibility of the infected-invertebrate hosts being eaten by caimans is minimal (LAINSON et al., 2003;PAPERNA & LAINSON, 2003).Caiman yacare become infected for the first time as juveniles, when its diet changes from ingestion of invertebrates to predation of anurans and fishes (VIANA et al., 2010a).In South America, predation of insectivorous vertebrates appears to be the main transmission route of H. caimani (VIANA et al., 2012).Hepatozoon caimani-cystozoites have been found in amphibian tissues of the following species: Leptodactylus fuscus, Leptodactylus chaquensis, Leptodactylus podicipinis, Scinax nasicus and Rana catesbeiana (LAINSON et al., 2003;VIANA et al., 2012), suggesting that anurans are paratenic hosts for H. caimani, although frogs are not part of caiman's diet.The characid fish Metynnis sp., when fed with C. quinquefasciatus previously engorged on naturally H. caimani-infected caiman showed cysts harbouring cystozoites identical to those of H. caimani, suggesting its role as paratenic hosts for this haemoprotozoa (PEREIRA et al., 2014).Although some attempts were unsuccessful in proving the experimental transmission of Hepatozoon to crocodilians (PESSÔA et al., 1972;LAINSON et al., 2003), structures similar to Hepatozoon oocystis have been recently described in these invertebrates collected from Caiman yacare's oral cavity (SOARES et al., 2017a).
Although H. caimani has been detected in a high prevalence in C. crocodilus (76.7%) in the Amazon region (LAINSON, 1977) and in C. yacare in western (71.4%) and southeastern (76-79.5%)Pantanal (VIANA & MARQUES, 2005;VIANA et al., 2010b;SOARES et al., 2017a) in Brazil, by blood smear examinations, molecular characterization studies are scarce.In fact, only recently, a molecular confirmation has been performed in a population of Caiman yacare sampled in Miranda, state of Mato Grosso do Sul, central-western Brazil (SOARES et al., 2017a) and in Cayman crocodilus in Amazon (SOARES et al., 2017b).The present work aims to detect and characterize the Hepatozoon spp.found in wild and captive Caiman yacare sampled at Poconé, North Pantanal, Mato Grosso state (MT), Brazil, using blood smear examinations and molecular techniques.

Material and Methods
Blood samples of C. yacare were collected from wild and captivity animals from August to October 2010, for a total of 2 collections.In the wild, the samples were taken at the area called Corixo Verde (16°25′,08,3″S, 56°37′ 39,2″W) at Piuval Farm in Poconé-MT (Figure 1).Thirty-three caimans (24 free-ranging and 9 captive; 28 males and 5 females) were captured during the day and night by hand or by using nooses attached to long poles.Afterwards they were tied, weighed and also marked (just free-ranging caimans) in their tail ridges with a numbered plastic ring and their snout-vent lengths were measured.Sex was determined by exposure of genitalia.
Furthermore, the blood samples were used in blood smears for microscopic examination.Slides were air-dried, fixed with methanol and stained with Giemsa.Morphologic and morphometric features of gamonts, as well as the changes caused by the presence of parasites in erythrocytes were analyzed using the CellSens Imaging software (Olympus) in the Immunoparasitology Laboratory, Department of Veterinary Pathology, Universidade Estadual Paulista (UNESP Jaboticabal).Mean morphometric variables (nuclear and erythrocytes length, width, and area) of parasitized and non-parasitized cells were compared using non-paired t-test with Welch's Correction; F test was used for comparing variance values.The results were analyzed using GraphPrism 7.03.
The phylogenetic tree of Hepatozoon spp.18S rRNA sequences clustered in a monophyletic group.The phylogenetic tree was basically in two branches: one of them composed by Hepatozoon sequences amplified in the present study and sequences of H. caimani, recently detected in caimans in southern Pantanal, and Hepatozoon spp.from other reptile taxa (snakes, lizards and tortoise) and amphibians retrieved from Genbank.The other branch grouped Hepatozoon sequences amplified from mammals and ticks previously deposited in Genbank.Adelina sp., Haemogregarina spp., Theileria sp., Isospora sp., and Sarcocystis sp. were used as outgroups (Figure 3).
Nucleotide polymorphisms and DNA divergence between sequences obtained in the present study were also analyzed.The analysis of nucleotide polymorphisms of 18S rRNA sequences showed two haplotypes (haplotype diversity (Hd): 1.000; variance of haplotype diversity: 0.25000; Standard Deviation (SD): =0.500) and nucleotide diversity (Pi) of 0.02113 (SD= 0.01057).
In the present study, morphometric alterations in parasitized cells by Hepatozoon spp.gamonts were markedly verified in all sampled yacares.The induction of changes in reptile parasitized cells by Hepatozoon spp.have already been reported in Caiman crocodilus yacare (VIANA & MARQUES, 2005;SOARES et al., 2017a) and in several species of snakes sampled in Brazil (MOÇO et al., 2002;O'DWYER et al., 2004;MOÇO et al., 2012).These changes in parasitize cells is probably mainly due to alterations in cellular erythrocyte cellular membrane's permeability (VIANA & MARQUES, 2005).
The variety of morphological and morphometric forms of gamonts verified in yacare' erythrocytes emphasizes the need for molecular confirmation of the involved Hepatozoon species.According to Moço et al. (2012), morphologically similar parasites showing different morphometric features may or may not represent different species.Furthermore, Hepatozoon spp.blood forms could show difference in their morphology depending on the host species involved (TELFORD, 2009).Besides, immature and mature gamonts, showing different morphological and morphometric features, could lead to a misdiagnosis of different species of Hepatozoon spp.(SMITH, 1996).Herein, we used a combination of morphometrical and phylogenetic assessment to confirm the occurrence of H. caimani in sampled caimans.
It seems that Hepatozoon species show limited host specificity, switching easily between different host species (MAIA et al., 2011).Regarding this high spectrum of host specificity found among Hepatozoon species, the host ecology appears to play a more important role in the complex relationship Hepatozoondifferent host species than the phylogenetic relatedness among the hosts properly (SLOBODA et al., 2007).For instance, similar Hepatozoon spp.isolates appears to infect different genera of lizards, as previously reported (MAIA et al., 2011).On the other hand, some Hepatozoon species found in snakes have shown high specificity being restricted to determined hosts (TELFORD et al., 2001).Because of that, identification of new Hepatozoon species based solely on blood smears examinations should be avoided (MAIA et al., 2011).The identification of Hepatozoon species in different caiman species in Brazil should be done based on blood smears examinations and molecular phylogeny.Also, future studies should assess the phylogenetic relationships among Hepatozoon sp.isolates in different alligator species, Culex mosquitoes (vectors), anurans (paratenic hosts) and leeches in Brazil.Regarding the molecular characterization based on rDNA coding regions, such as that used in the present study (18S rRNA), although highly conserved and used frequently in molecular characterization of Hepatozoon spp. in several hosts (ANDRÉ et al., 2010;BORTOLI et al., 2011;MOÇO et al., 2012;O'DWYER et al., 2013;SOARES et al., 2017a;SOUSA et al., 2017), this target sequence may not be suitable for phylogenetic inferences of parasites which share some of intraspecific divergences (MOÇO et al., 2012), preventing attempts to differentiate those species.The search for new targets genes which allow a better phylogenetic differentiation of this group of parasites are much needed as an attempt to elucidate the diversity of Hepatozoon species that parasitize reptiles.
In summary, the present work showed through morphological, morphometric and molecular approaches, the occurrence of H. caimani in free-ranging and captive Cayman yacare in northern Pantanal.

Figure 1 .
Figure 1.Capture site.Map of Mato Grosso state, central-western Brazil, showing the locality where caimans were sampled in the present study.

Figure 3 .
Figure 3. Phylogenetic tree based on an alignment of 1200bp fragment of Hepatozoon spp.18SrRNA sequences from Caiman yacare blood samples, using Maximum Likelihood (ML) method and TVM+G+I evolutionary model.Bootstrap values for ML (≥50) are given above relevant nodes.The sequences indicated in bold represent those from this study and original sample codes were named.