Comparative Genomic Analyses of New and Old World Viscerotropic Leishmanine Parasites: Further Insights into the Origins of Visceral Leishmaniasis Agents

Visceral leishmaniasis (VL), also known as kala-azar, is an anthropozoonotic disease affecting human populations on five continents. Aetiologic agents belong to the Leishmania (L.) donovani complex. Until the 1990s, three leishmanine parasites comprised this complex: L. (L.) donovani Laveran & Mesnil 1903, L. (L.) infantum Nicolle 1908, and L. (L.) chagasi Lainson & Shaw 1987 (=L. chagasi Cunha & Chagas 1937). The VL causal agent in the New World (NW) was previously identified as L. (L.) chagasi. After the development of molecular characterization, however, comparisons between L. (L.) chagasi and L. (L.) infantum showed high similarity, and L. (L.) chagasi was then regarded as synonymous with L. (L.) infantum. It was, therefore, suggested that L. (L.) chagasi was not native to the NW but had been introduced from the Old World by Iberian colonizers. However, in light of ecological evidence from the NW parasite’s enzootic cycle involving a wild phlebotomine vector (Lutzomyia longipalpis) and a wild mammal reservoir (the fox, Cerdocyon thous), we have recently analyzed by molecular clock comparisons of the DNA polymerase alpha subunit gene the whole-genome sequence of L. (L.) infantum chagasi of the most prevalent clinical form, atypical dermal leishmaniasis (ADL), from Honduras (Central America) with that of the same parasite from Brazil (South America), as well as those of L. (L.) donovani (India) and L. (L.) infantum (Europe), which revealed that the Honduran parasite is older ancestry (382,800 ya) than the parasite from Brazil (143,300 ya), L. (L.) donovani (33,776 ya), or L. (L.) infantum (13,000 ya). In the present work, we have now amplified the genomic comparisons among these leishmanine parasites, exploring mainly the variations in the genome for each chromosome, and the number of genomic SNPs for each chromosome. Although the results of this new analysis have confirmed a high genomic similarity (~99%) among these parasites [except L. (L.) donovani], the Honduran parasite revealed a single structural variation on chromosome 17, and the highest frequency of genomic SNPs (more than twice the number seen in the Brazilian one), which together to its extraordinary ancestry (382,800 ya) represent strong evidence that L. (L.) chagasi/L. (L.) infantum chagasi is, in fact, native to the NW, and therefore with valid taxonomic status. Furthermore, the Honduran parasite, the most ancestral viscerotropic leishmanine parasite, showed genomic and clinical taxonomic characteristics compatible with a new Leishmania species causing ADL in Central America.


Introduction
Visceral leishmaniasis (VL) is an anthroponotic or zoonotic disease that affects human populations on five continents. According to the World Health Organization (WHO), VL causes approximately 50,000 to 90,000 new cases annually [1]. More than 90% of VL cases occur in six countries: India, Bangladesh, Sudan, South Sudan, Ethiopia, and Brazil [2]. In Latin * L. (L.) infantum chagasi from Brazil/** L. (L.) infantum chagasi from Honduras. † Previously published data [18]. # Unpublished data.
The total lengths of both genomes were 31,924,889 nt, with 36 chromosomes that varied in length, as shown in Figure 1; the total number of genes predicted for each genome was 8423. After the sequencing procedures, 36 chromosomes were initially identified for the two strains, with length profiles shown in Figure 1. A progressive increase in chromosome lengths from 1 to 36 can be observed. was 8423. After the sequencing procedures, 36 chromosomes we the two strains, with length profiles shown in Figure 1. A progres some lengths from 1 to 36 can be observed.   The genomic comparisons among L. (L.) infantum chagasi from Brazil (Sout and Honduras (Central America) and L. (L.) infantum (Europe) were also evalu on the numbers of SNPs in each genome-a genetic character linked to the anc parasite. We observed that the Honduran parasite had more than twice as m (44,627) as the Brazilian parasite (16,867), demonstrating its older ancestry The genomic comparisons among L. (L.) infantum chagasi from Brazil (South America) and Honduras (Central America) and L. (L.) infantum (Europe) were also evaluated based on the numbers of SNPs in each genome-a genetic character linked to the ancestry of the parasite. We observed that the Honduran parasite had more than twice as many SNPS  (44,627) as the Brazilian parasite (16,867), demonstrating its older ancestry among the leishmanine parasites examined (Figure 3). novani; and L. (L.) major.
The genomic comparisons among L. (L.) infantum chagasi from Brazil (South and Honduras (Central America) and L. (L.) infantum (Europe) were also evalua on the numbers of SNPs in each genome-a genetic character linked to the ance parasite. We observed that the Honduran parasite had more than twice as m (44,627) as the Brazilian parasite (16,867), demonstrating its older ancestry a leishmanine parasites examined ( Figure 3).

Discussion
Our results represent the first genomic analysis performed in Latin ) donovani], although previously molecular clock comparisons of polymerase alpha subunit gene (a highly conserved genomic region related to tionary process of leishmanine parasites [26]) among these parasites indicate Honduran parasite was significantly older (382,800 ya) than the Brazilian parasi

Discussion
Our results represent the first genomic analysis performed in Latin America on wholegenome sequences of L. (L.) infantum chagasi from Honduras (Central America), the same parasite from Brazil (South America), and sequences of L. (L.) infantum (Europe) and L. (L.) donovani (India). The three parasite genomes revealed high genomic identity (~99%) [except L. (L.) donovani], although previously molecular clock comparisons of the DNA polymerase alpha subunit gene (a highly conserved genomic region related to the evolutionary process of leishmanine parasites [26]) among these parasites indicated that the Honduran parasite was significantly older (382,800 ya) than the Brazilian parasite (143,300 ya), L. (L.) donovani (33,776 ya), or L. (L.) infantum (13,000 ya) [18]. Moreover, the Honduran parasite was found to have a single structural variation (and perhaps also a single functional variation) on chromosome 17 (Figure 2), as well as the highest frequency of genomic SNPs (more than twice that of the Brazilian parasite) ( Figure 3). The original data presented here contradict recent views concerning the origin of the Leishmania parasite responsible for AVL as well as its taxonomic position [12][13][14][15]27].
These results also appear to contradict the interpretation that the low genomic heterogeneity of L. (L.) infantum chagasi samples from Latin America (mostly from Brazil) is due to the recent introduction of that parasite into the New World [15]. Our results clearly suggest that the low genomic heterogeneity of L. (L.) infantum chagasi reflects the parasite's low genomic plasticity (due to its older ancestry) much more than its recent introduction into the New World. Likewise, another species belonging to the L.  [28][29][30][31].
Indeed, the high genomic identity (~99%) observed among these parasites [except L. (L.) donovani] has been crucial to the view that L. (L.) infantum chagasi is synonymous with L. (L.) infantum and that the causal agent of AVL was introduced to Latin America during Iberian colonization over the last 500 years [12][13][14][15]27]. However, it is interesting to note how difficult it is to explain the establishment of the parasite in Latin America in an ecological niche quite different from that of southwestern Europe. The Latin American niche includes a sandfly vector of a different genus (Lutzomyia) from that found in Europe (Phlebotomus) and wild mammalian hosts of different orders (rodent, marsupial, canid, feline, and chiroptera) [3,[32][33][34][35][36][37][38][39][40]. Adaptation of the parasite to the main phlebotomine vector of AVL in Latin America, Lutzomyia longipalpis, has been suggested [32]. In doing so, previous authors have omitted some principles of ecology of the sandfly, which lives in a natural environment in Latin America that is alongside the greatest known global genetic diversity of Leishmania parasites. It is therefore relevant to note that the Latin American geographic region currently hosts representatives of the three known subgenera of the genus Leishmania (including Leshmania, Viannia, and Mundinia) [41], with a total of fifteen different well-known causal agents of American cutaneous leishmaniasis (ACL). Thus, it does not seem likely that Lutzomyia longipalpis, a wild phlebotomine species considered permissive at the laboratory level [47][48][49], would no longer be involved in the transmission of any of those various Leishmania species in Latin America. Occasionally, there is evidence of L. (V.) braziliensis or L. (L.) amazonensis DNA in Lutzomyia longipalpis caught in the wild [50][51][52][53]-although these reports are not sufficient to incriminate this phlebotomine as a vector of leishmanine parasites. It is extremely relevant that the laboratory-bred sandfly vector can become experimentally infected after feeding on blood from the skin lesion edges of a patient with ACL caused by L. (V.) braziliensis [54]-reinforcing its potential as a vector for transmitting a Leishmania species other than L. (L.) infantum chagasi-which has never been definitively demonstrated under natural conditions. It, therefore, appears easier to accept that Lutzomyia longipalpis has developed a species-specific interaction with native L. (L.) infantum chagasi and its main wild reservoir (the crab-eating fox, Cerdocyon thous) over thousands of years and represents an ecologically balanced enzootic cycle that is widely distributed in Latin America [3,4,37,[55][56][57]. It is not imperative that major genomic differences exist between L. (L.) infantum chagasi and L. (L.) infantum for this ecologically balanced enzootic cycle to occur in the New World, but only that L. (L.) infantum chagasi existed prior to the description of L. (L.) infantum-which seems clear based on the results of this work.
Corroborating the idea of a native enzootic cycle of the parasite in the New World is recent evidence pointing to the role of a small wild rodent (Proechimys sp.) in the enzootic cycle of L. (L.) infantum chagasi in a mountainous area of primary forest in the "Serra dos Carajás" (municipality of Marabá) in Pará State in the Brazilian Amazon [33]. European/African descendants only entered that area approximately 50 years ago-an extremely short window of time for the parasite to have adapted to wild rodents after the arrival of supposedly infected dogs.
More recent evidence has corroborated the wild enzootic cycle of the parasite in the Brazilian Amazon in the central area of Amapá State in the Wajãpi Indigenous Territory (WIT). The WIT area is a rugged landscape covered by dense rainforest and is part of one of the world's largest continuous protected areas of rainforest, covering over 12 million hectares. A natural L. (L.) infantum (chagasi) infection was found there in two wild rodents (Dasyprocta sp. and Proechimys cuvieri) by PCR-based amplification of a fragment of the ITS1 gene (300-350 bp) 7-8 and by semi-nested PCR for a fragment of hsp70 (640 bp) [34]. Taken together, this information indicates the existence of a native enzootic cycle of L. (L.) infantum chagasi in the Brazilian Amazon, with the participation of not only a wild canid (Cerdocyon thous) but also small wild rodents (Proechimys sp., Proechimys cuvieri, and Dasyprocta sp.). Coincidence or not, it was precisely in this region that the original description of the AVL causal agent was made (Leishmania chagasi Cunha & Chagas 1937).
From the genomic point of view, the present results undoubtedly confirm a high similarity (~99%) among the studied parasites [except L. (L.) donovani] and, by extension, even isolated samples of asymptomatic human infection [12][13][14][15]27,58]. However, it is necessary to emphasize the presence of a single structural variation (and perhaps a functional variation) on chromosome 17 of the Honduran parasite in relation to the same parasite from Brazil, L. (L.) infantum reference strain [JPCM5] [59], and L. (L) donovani (Figure 2), although the full meaning of this variation has not yet been deciphered. Interestingly, there had been no reported AVL cases for at least ten years in the area (Isla del Tigre, Amapala District, West Coast of Honduras) when the parasite was isolated from a human case of atypical dermal leishmaniasis [ADL] (all symptomatic cases of human infection registered during that time were the ADL form). Whether there is any association between that single structural variation on chromosome 17 of the Honduran parasite and the biology of human infection is open to further investigation, although we speculate that an ecological process of host-parasite stabilization during the human-parasite ancestry process in Central American countries could have occurred. The situation in Costa Rica is an example; only the atypical cutaneous form of infection has been recorded in recent years (although a single case of AVL has now been diagnosed) [60].
Another possibility analogous to that of the Honduran parasite involves the RagC gene found on chromosome 36 of L. (L.) donovani, which controls GTPase protein expression and appears to be involved in the process of parasite visceralization in Sri Lanka in Asia [61,62]. It will therefore be necessary to screen the main genes on chromosome 17 of the Honduran parasite to search for a gene that might inhibit the parasite's visceralization, as there is strong clinical evidence that human infections are evolving towards dermal stabilization in the area of the parasite's origin [63,64].
Another original finding in this work is the high frequency of genomic SNPs in the Honduran parasite (more than twice the number found in the Brazilian parasite) (Figure 3), a fact that seems to be strongly related to the older ancestry of the Honduran parasite compared to the other parasites [18]. SNPs are known to have low mutation rates, which makes them useful genetic markers for following the inheritance patterns of chromosomal regions from generation to generation and excellent markers of ancestry [65].
There are currently two clinical forms (ADL and AVL) of human symptomatic L. (L.) infantum chagasi infections along the eastern coast of Honduras, with ADL being the most common. Previous research (using kDNA RFLP and RAPD techniques) has indicated that the two clinical forms (ADL and AVL) are caused by genetically similar parasites [66]. Additional observed genomic similarities (CG content percentages and mean coverage) supported a high genomic identity (~99%) among the parasites studied ( Table 1) and confirmed that they constitute a genomically cohesive group. Although the results of the present genomic analysis corroborate previous works [12][13][14][15]58], the major question in this study concerns not only genomic differences among the parasites but their geographical origins.
As such, the complement to comparative genomic analyses of the parasites is molecular clock comparisons of the DNA polymerase alpha subunit gene (a highly conserved genomic region related to the evolutionary process of leishmanine parasites) [26]. Molecular clock comparisons have revealed that the Honduran parasite is significantly older (382,800 ya) than L. (L.) infantum chagasi from Brazil (South America), L. (L.) donovani (India), and L. (L.) infantum (Europe) [18]. These molecular clock data represent strong evidence that L. (L.) infantum chagasi is native to the New World (more precisely, Central America) and not to the Old World, as has been suggested [12][13][14][15]27]. One might counter this evidence by noting that the genomic analysis was based on only one strain of L. (L.) infantum chagasi from Honduras and Brazil, however, similar results have been obtained by PCR analyses examining three different genetic targets (12S, 9S, and ND7 of maxi circle genes) from a larger sample of parasites (4) showing the same clinical form of the disease (ADL) as well as the same geographic origin (Isla del Tigre, Amapala District, West Coast of Honduras). These results corroborate our finding that the Honduran parasite is a different lineage than the Brazilian parasite (Fernández Figueroa, unpublished data). Moreover, Leishmania (Viannia) utingensis Braga et al. 2003, a parasite that was described based on the phenotypic characteristics of a single isolate obtained from the sandfly Lutzomyia (Viannamyia) tuberculata, remains a valid species without any other isolate of the parasite from the same sandfly, from another wild mammal, or even from humans in the Brazilian Amazon. In a punctual analysis such as this, with a focus on the origin and/or the taxonomy of these parasites, the evaluation of only one individual of each species provided highly significant results for the origin (ancestry) of these parasites, and a broad genomic spectrum analysis of different endemic regions worldwide would have led to the 31 Latin American samples being treated as L. (L.) infantum, therefore neglecting L. (L.) infantum chagasi [67].
Another point that deserves to be highlighted refers to the accuracy of our previous molecular clock comparisons [18]. Although we used only the DNA polymerase alpha subunit gene to assess ancestry, we were able to identify a phylogenetic tree consisting of the two major clades within the genus Leishmania Thus, we note that our phylogenetic design, based on DNA polymerase alpha subunit gene sequences, was able to reconstruct a phylogenetic tree with a podological profile substantially similar to trees reconstructed using other gene regions [29,68], demonstrating its application for phylogenetic studies of Leishmaniinae and related parasites. It is also important to mention the pioneering work that produced exceptionally high-value DNA and RNA polymerase gene sequences in phylogenetic studies of the genus Leishmania. Those studies considered several representative species of parasites from the L. ) braziliensis], as well as parasites of previously questionable taxonomy [Leishmania herreri, Sauroleishmania adleri, Sauroleishmania deanei, Sauroleishmania gymnodactyli, and Sauroleishmania tarentolae], which led to the proposal that Leishmania spp-infecting reptiles evolved from mammalian Leishmania spp. [69], a hypothesis that is now well-established in taxonomic classifications of the genus Leishmania [41,70].
Based on these results, there is now a clear need to determine not only the origins and historical geographical routes of dispersal of leishmanine parasites associated with visceral leishmaniasis throughout the world but also their taxonomic positions. This will be a difficult task, however, as the origins of these parasites date to approximately 90-100 million years ago [68]. The first significant step in this direction is the suggestion that "the predecessors of the Leishmania (L.) donovani group and Leishmania (L.) major would have evolved from monoxenous parasites of insects in South America   [13]. Thus, it is not surprising that the results of our previous work show the likely origin of L. (L.) infantum chagasi as the New World. However, unlike this previous proposal, L. (L.) infantum chagasi is considered to have a more specific origin in the Central American region [18]. It, therefore, seems reasonable to assume that the parasite migrated from the Central American region to Asia, following the itinerary suggested above, and then diverged into European and North African [L. (L.) infantum], and into African and Indian/Kenyan [L. (L.) donovani] [13]. The geographic dispersal of the parasite may also have been directed southwards by the migration of wolves of approximately 10 Mya from North America to Central America, finally arriving in South America [71]. There is also a strong ecological relationship between the parasite [L. (L.) infantum chagasi] and a widely distributed wild canid species in Latin America, the crab-eating fox Cerdocyon thous [19], which is a strong attraction for the main sandfly vector of the parasite, Lutzomyia longipalpis [32]. Reinforcing this hypothesis of a North American to South American axis, a similar geographic trajectory may also have been made by the North American L. Leishmania species representing the leishmanine parasite in the Central American countries that is strongly associated with atypical cutaneous leishmaniasis (ACL) or nonulcerated cutaneous leishmaniasis (LCNU). That putative species seems to be signaling a strong biological character of this parasite (a strong tropism for human skin), which is distinct from descendants that demonstrated typical visceral tropisms. In line with this proposal, it is relevant that the taxonomic classification of the genus Leishmania is not only well-defined phylogenetically but can also be described in terms of host specificity (vertebrate and/or invertebrate) or by clinical parameters [70]. Thus, the need to define the specific identity of the Central American (Honduras) ancestral parasite seems clear and will be addressed in the near future.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.