Origin of the New World Plasmodium vivax: facts and new approaches

Abstract

Malaria is one of the most important human diseases throughout tropical and sub-tropical regions of the world. Global distribution and ample host range have contributed to the genetic diversity of the etiological agent, Plasmodium. Phylogeographical analyses demonstrated that Plasmodium falciparum and Plasmodium vivax follow an Out of Africa (OOA) expansion, having a higher genetic diversity in African populations and a low genetic diversity in South American populations. Modeling the evolutionary rate of conserved genes for both, P. falciparum and P. vivax, determined the approximate arrival of human malaria in South America. Bayesian computational methods suggest that P. falciparum originated in Africa and arrived in South America through multiple independent introductions by the transatlantic African slave trade; however, in South America P. vivax could have been introduced through an alternate migratory route. Alignments of P. vivax mitogenomes have revealed low genetic variation between the South American and Southeast Asian populations suggesting introduction through either pre-Columbian human migration or post-colonization events. To confirm the findings of these phylogeographical analyses, molecular methods were used to diagnose malaria infection in archaeological remains of pre-Columbian ethnic groups. Immunohistochemistry tests were used and identified P. vivax but not P. falciparum in histologically prepared tissues from pre-Columbian Peruvian mummies, whereas shotgun metagenomics sequencing of DNA isolated from pre-Columbian Caribbean coprolites revealed Plasmodium-homologous reads, current evidence suggests that only P. vivax might have been present in pre-Columbian South America.

Introduction

Plasmodium the etiological agent of malaria, is an intracellular parasite transmitted by the Anopheles mosquito. Plasmodium spp. are water-associated parasites, the vector uses stagnant water sources near warm-blooded animals as part of its life cycle (Carter and Mendis 2002). While malaria is a global threat, most severe cases of malaria are seen in low income tropical and sub-tropical countries since the warm and humid climate favors the replication of the vector. Although there are therapeutic drugs used that have had relative success over the years, the parasite has developed resistance to the artemisinin based drugs, chloroquine and sulfadoxine/pyrimethamine (Arrow et al. 2004; Mita et al. 2009; Vinayak et al. 2010). Vaccines are being developed to target the Plasmodium parasite, although the challenge behind the development of the malaria vaccine is the immense genetic diversity in Plasmodium strains as a result of the rapid mutation rate in response to environmental stress (Crompton et al. 2010; Vinayak et al. 2010).

Selective pressure driven by global distribution as well as ample host ranges have contributed to a large genetic diversity in Plasmodium (Good 2005; Hart 2004; Li et al. 2001; Neafsey et al. 2008). Plasmodium diversification has accumulated through the years, and evaluating the global genetic diversity using conserved gene regions in distinct populations can estimate the origin, evolution, and distribution of Plasmodium spp. To explore the possibilities of the malaria parasite in pre-Columbian America, this article will emphasize previous on phylogeographical research on P. falciparum and P. vivax in Africa, Asia and South America where the human malaria parasites reside.

The Origins of Plasmodium spp.

Theoretically, all parasites evolved from once free-living organisms, and these organisms likely lived adjacent to a future host, where the organism eventually adapted and became dependent on a specific host. These once free-living organisms likely simplified their morphology and metabolic specificities in order to accommodate to the host environmental niche. For the malaria parasite it has been suggested that the lineage diverged from other apicomplexan lineages in the Precambrian period prior to their vertebrate host (speculated origin as parasite of invertebrates) (Escalante et al. 1995a and 1998). While human malaria has arisen independently several times throughout different circumstances (primarily through host switch), human specific malaria is estimated to have derived about 129 million years ago (Escalante et al. 1998). The four recognized human malaria, P. falciparum, P. malariae, P. ovale, and P. vivax are distantly related and yet seem to be products of lateral host transfer (Escalante et al. 1995a; Rich et al. 2009). Phylogeographical analyses suggest that Plasmodium spp. follows an Out of Africa (OOA) expansion along with early human hosts.

Human malaria originated from a hominid host. The closest relative of P. falciparum is Plasmodium reichenowi (chimpanzee malaria) that form a monophyletic lineage based on separate alignments of the 30S rRNA genes (Escalante et al. 1995a) and the circumsporozoite protein (CSP) genes (Escalante et al. 1995b). However, a recent study has confirmed that P. falciparum most likely has a gorilla rather than chimpanzee origin; the finding was based on sequencing a portion of P. falciparum mitochondrial genome (Liu et al. 2010; Sharp et al. 2011). Largely, Plasmodium spp. are regarded as host specific parasites, rarely do mixed infections occur (Liu et al. 2010). P. falciparum seems to have diverged from P. reichenowi an estimated 8 million-years ago during the late Cenozoic era (Ayala et al. 1999), same estimated time as humans diverged from simians (Otto et al. 2014). The divergence of P. falciparum and P. reichenowi has led to the loss and mutation of certain genes (estimated to be about 100) generating species differentiation and adaption based on specific hosts (Otto et al. 2014). As its host migrated, Plasmodium spp. likely co-migrated with its host. Simian malaria is considered a zoonotic infection, thus species differentiation followed as a result of host transfer between simians and early humans (Escalante et al. 1995b; Yamasaki et al. 2011).

The Distribution of P. falciparum

Mitochondrial, microsatellite, and housekeeping genes of P. falciparum have revealed the population structure and genetic differentiation with geographic distribution. Genetic differentiation using microsatellite markers in P. falciparum revealed high variation in African (0.76–0.8), intermediate in Southeast Asia/Pacific (0.51–0.65) and low variation in South American (0.3–0.4) populations (Anderson et al. 2000). Multiple studies propose that low variation of South American populations reflect P. falciparum introduction through the colonization of Europeans and the African transatlantic slave trade (Anderson et al. 2000; Joy et al. 2003; Neafsey et al. 2008; Rodrigues et al. 2018; Tanabe et al. 2010; Yalcindag et al. 2012). Using Goldstein (δμ)² distance (Goldstein et al. 1995) between South American and African P. falciparum populations, resulted in a divergence time between 385 and 1,101 years ago (mean of 665 years ago) (Anderson et al. 2000). The divergence range is consistent with post-colonization events, however the range is large and faulty due to errors of estimated mutation rate and variation among loci (Anderson et al. 2000). Phylogeographical analysis and Bayesian computational methods suggest that P. falciparum originated in Africa and arrived in South America via multiple independent introduction events dividing South American P. falciparum populations into two clusters, possibly due to various slave trade routes of the European empires (Spanish and Portuguese) between the 16th and the 19th centuries (Yalcindag et al. 2012).

P. simian and P. vivax

The idea of malaria being present in America prior to the European colonization is controversial. The concept is consistent with historical records and considers the most likely scenario that the human malaria P. falciparum and P. vivax originated in Africa. However, Carter (2003) postulated that P. vivax could have been introduced twice into the New World by pre-Columbian human migrations from Asia and by the European colonization and the introduction of transatlantic slave trade. Alignment of the CSP gene and tandem repeats loci (TRs) demonstrates low genetic distance between human malaria and New World simian malaria specifically P. vivax and P. simium rendering the species morphologically and genetically indistinguishable from one another suggesting lateral host transfer (human to simians or vice-versa) (Escalante et al. 1995b; Leclerc et al. 2004a). Through an 18S rDNA cladistic analysis, Leclerc et al. (2004b) observed clustering of Asian simian malaria (P. coatneyi, P. cynomolgi, P. fieldi, P. gonderi, P. fragile, P. inui, P. knowlesi) and South American malaria P. vivax and P. simium with the exception of P. gonderi which infects Cercopithecidae (Old World simians) from Africa resulting in a bootstrap value of 80% confidence interval. 18S rRNA analysis found similar polymorphisms between New World P. simium and Old World human P. vivax (Li et al. 2001). South American P. vivax is distinct from P. simium, thus New World P. simium and P. vivax did not originate from the same source and most likely New World P. vivax was introduced twice in two separate events (Li et al. 2001). Analyses based on Duffy binding protein (DBP) suggest that New World P. vivax and P. simium are highly similar but P. simium DBP presented lower genetic variability and thus it is thought that P. simium evolved recently as a result of post-colonization events (Camargos Costa et al. 2015).

P. vivax in South America

Southeast Asian and South American P. vivax populations are related to a certain degree. In particular, P. vivax from Thailand is more closely related to P. vivax from Venezuela by 0.252 according to Fixation index estimator (Fst), while P. vivax Venezuela population is less related to the Azerbaijan, Ethiopia and Turkey P. vivax population exhibiting a Fst ranging from 0.263 to 0.609 (Leclerc et al. 2004a). Taylor et al. (2013) share similar findings, two mitochondrial genomes of P. vivax from Thailand were identical to twelve Venezuelan and several Brazil P. vivax mitochondrial genomes. Taylor et al. (2013) and Rodriguez et al. (2018) used Bayesian phylogeographical analysis of large global collection of P. vivax mitogenomes (mitochondrial DNA haplotypes) to show similarities between South America and Southeast Asia populations supporting transpacific migration. Specifically best-supported migration models assume migration between Melanesia and South American, based on Bayesian skyline coalescent models it is likely that P. vivax arrived in South America an estimated 52,149 years before the present (95% highest probability density interval, ranging about 29,896 to 60,659 years) (Rodriguez et al. 2018). It is presumed that modern humans exited Africa earlier than expected, as evidenced by excavated stone artifacts from Madjedbebe rock shelter in Australia, the artifacts were dated 65,000 years before the present with 95% probability (ranging about 60,000 to 70,000) according to Bayesian analysis (Clarkson et al. 2017). Bayesian statistical analysis of Rodriguez et al. (2018) and Clarkson et al. (2017), their time frames for human migration coincide. In theory, human malaria arrived first to South Asia then to South America via human migration. Rodriguez et al. (2018) postulates that population Y (East Asia population that migrated to Australia resulting in the Australasians population) might have introduced Melanesian strains of P. vivax to South America (Amazonian population) before the European colonization. While Taylor et al. (2013) postulates that P. vivax arrived in South America either via pre-Columbian migrations or post-colonization events with multiple sources of P. vivax introductions. Thus far, phylogeographical analysis of extant P. vivax global collection of mitogenomes hypothesized the existence of pre-Columbian malaria.

Duffy Negative Blood Group

Today, P. vivax is the most widespread pathogen throughout Asia, the Middle East, Central and South America. It was previously presumed that P. vivax originated as Asian malaria, however based on a single nucleotide polymorphism known as the Duffy negative phenotype P. vivax is an ancient parasite of African hominids. The T to C substitution of the promoter region at nucleotide −33 (in the GATA box) disrupts the binding site for the erythroid transcription factor (GATA1) thus disrupts the expression of the erythrocyte receptor known as Duffy antigen receptor for chemokines (DARC) (Horuk et al. 1993; Tournamille et al. 1995; Howes et al. 2011). Duffy negative individuals are protected from P. vivax infection, lack of the DARC inhibits the attachment of Plasmodium vivax Duffy Binding Protein (PvDBP) consequently disrupting the parasites natural life-cycle (Horuk et al. 1993; Tournamille et al. 1995; Howes et al. 2011). Today, an estimated 70% of the African population is negative for the DARC on red blood cells (Tournamille et al. 1995). Duffy negative phenotype is highly prevalent in Central and West Africa populations upwards of 95% to 100% (Howes et al. 2011), suggesting that the mutation emerged as a possible prolonged selection pressure in response to P. vivax infections (Carter et al. 2003). Considering the prior statements, Duffy negative phenotype and the rarity of P. vivax in Sub-Saharan Africa, P. vivax would have a lower likelihood of being introduced into South America through the transatlantic slave trade (Taylor et al. 2013). However, Rodriguez et al. (2018) using neighbor-joining analysis of extant P. vivax mitogenomes demonstrated connectivity between South American haplotype (Ame1) and African haplotype (Afr1), observing genetic contribution from African as well as South Asian P. vivax lineages, the African haplotype most likely arrived to South America via post-colonization events. Today, human P. vivax is more prevalent in Asia and in Latin America than it is in Central and West Africa because of the Duffy negative phenotype (Liu et al. 2014). Reportedly, great ape (chimpanzees and gorillas) P. vivax-like parasite is more prevalent in Central Africa and is more diverse than human P. vivax. Currently it is thought that human P. vivax evolved from the P. vivax-like lineage through host switch (Liu et al. 2014; Prugnolle et al. 2013).

Plasmodium vivax in the New World

Two of the four important human malaria parasites, P. falciparum and P. vivax, are related to great ape malaria. Theoretically, humans living in close proximity to non-human primates harboring the simian malaria encouraged host switching through species differentiation via host immune responses. While human migration helped spread the intracellular parasite throughout diverse regions, presumably P. falciparum and P. vivax arrived throughout different routes to South America. Although paleoparasitological analyses make use of microscopy to identify parasite infection in ancient samples, immunological tests provide direct evidence by identifying pathogens conserved antibodies in the samples. Gerszten et al. (2012) examined 155 samples of spleens and livers from Peruvian mummies dating back 3000 to 600 years before present. The samples were cut into histological sections, fixed onto microscope slides and stained with H&E to identify malaria pigments. Using ELISA tests, sixty-seven percent of the specimens were positive for P. vivax antibodies and negative for P. falciparum antibodies. Immunological tests diagnosed Plasmodium vivax infections in pre-Columbian indigenous cultures, seeing as P. vivax is genetically related to simian malaria (Leclerc el al. 2004b; Liu et al. 2014) possible interactions between primates and pre-Columbian cultures would have facilitated transmission. Howler monkey mummies were excavated from ancient human cemeteries in Peru and were suspected as possible source of food or adopted as pets by these ancient cultures (Gerszten et al. 2012).

Shotgun Metagenomic Sequencing as Means to Identify Plasmodium

Shotgun metagenomic sequencing has presented an opportunity for detecting Plasmodium-homologous reads in ancient samples. Plasmodium genes can be identified in detectable amounts in fresh human and simian feces of infected hosts (Jirků et al. 2012; Liu et al. 2014). Tito et al. (2008) identified 34 reads of Plasmodium genus after isolating and sequencing two human coprolites (dating approximately 1,300 BP) from the Cueva de los Muertos Chiquitos archaeological site (near Rio Zape, Durango, Mexico), BLASTN alignment revealed Plasmodium reads with an identity of > 80%. Similarly, our laboratory has detected Plasmodium reads in human coprolites from Vieques, Puerto Rico dating back to 1,417 to 1,787 years before the present. The Plasmodium-homologous reads were sought out using similar methods as Rivera-Perez et al. (2015). Using BLASTX alignment with a non-redundant protein database a total of 129 reads related to Plasmodium were detected. One ethnic group known as the Saladoids, resulted in Plasmodium-homologous reads (n = 127) with an average percentage of identity of 81% and an E-value of 2.05E-18 (Table. S2). Another ethnic group known as the Huecoids, resulted in Plasmodium-homologous reads (n = 2) with a lower average percentage of identity and E-value, respectively 54% and 2.22E-16 (Table. S1). The reads revealed homology to multiple organisms (both prokaryotes and eukaryotes) but assigned a higher Max Score to Plasmodium query (Data S1 & S2), which lead us to emphasize the use of the term ‘Plasmodium-homologous reads’ since the sequences could not definitely prove an affirmative identification of the organism.

Ancient DNA (aDNA) demonstrates a degree of fragmentation and degradation (Der Sarkissian et al. 2015; Pääbo, 1989). The fact that aDNA may be damaged and considering the high mutation rate of the malaria parasite (Carlton et al. 2008; Crompton et al. 2010; Vinayak et al. 2010), may have as a consequence a poor alignment of aDNA sequences with those of a modern database. Given the age of the sample, if the reads do in fact correspond to the malaria parasite we must consider over a thousand years of adaptive mutation, host switching and geographic distribution (migration of host) of the malaria parasite leading to genetic differentiation, possibly explaining low quality alignment of the Plasmodium predicted reads detected in Vieques coprolites. Yet, without physical evidence (e.g. immunohistochemical assays), the detection of Plasmodium low quality reads is somehow indicative of possible malaria infections in ancient Caribbean cultures. In this case further analysis is required to prove the presence of pre-Columbian malaria, for example performing Next Generation Sequencing (NGS) of Plasmodium spp. conserved genes (e.g. circumsporozoite protein or mitochondrial genes) prepared using mummified tissue (liver or spleen) from pre-Columbian South American cultures with positive affirmation of the malaria parasite by immunohistochemical tests. It would be recommended to use preserved liver or spleen samples since parasitized erythrocytes are entrapped in these organs causing enlargement as an immunological response, a larger sample size and accumulation of the parasite will have higher probabilities of detecting large number of Plasmodium reads. After sequencing phylogenetic inference should be performed to rule out possible false positive taxonomic assignments by alignment programs (Cleeland et al. 2013; Søe et al. 2015).

Supplementary Data

Plasmodium-homologous reads discussed in this study have been made available as Dataset_Huecoid & Dataset_Saladoid. Remote BlastX search homology results of the datasets have been made available as BlastX_Results.

Conclusions

Currently there are four reported Plasmodium species that cause malaria in humans: P. falciparum, P. vivax, P. malariae, and P. ovale. All four species are remotely related to each other, diverging at different times (Rich et al. 2009). Of the four known human malaria types, it is possible that P. vivax was present in pre-Columbian America. Genetic similarities between New World and Old World P. vivax (mainly Southeast Asia) indicate either P. vivax arrival to South America through pre-Columbian human migration or through post-colonization events. Molecular methods applied to archaeological remains has provided insight, our own data are suggestive of the presence of Plasmodium in pre-Columbian Caribbean while immunohistochemical assays of Plasmodium antibodies in pre-Columbian Peruvian mummies revealed P. vivax as the culprit in pre-Columbian malaria. Nonetheless further analyses combining molecular paleoparasitology and phylogenetic inference will be required for positive identification of P. vivax as a possible etiological agent of malaria in pre-Columbian America.