Malaria: what can apes teach humans?

Human malaria is caused by four human-host restricted or adapted species of the genus Plasmodium, listed here in their order of risk for causing severe disease Plasmodium falciparum, Plasmodium vivax, Plasmodium ovale and Plasmodium malariae – or so we thought. There were a few exceptions to this accepted wisdom, not least, the very close similarity between P. vivax and P. simium and P. malariae and P. brazilianium (the non-human hosts being New World monkeys) [1], but overall the story appeared complete. P. falciparum remained the cause of unacceptably high morbidity and mortality, with the more efficient artemisinin-derivative drugs often not being made available to those most in need [2–4].

In many places policies have changed for the better and artemisinin-based combination therapies are much more widely available at affordable prices [5]. Dogmas regarding disease have altered too. A recent focus on the more neglected vivax malaria suggests that P. vivax is more virulent than formerly thought [6,7]. Postmortem material from African children in coma treated for cerebral malaria died of causes other than malaria in a significant proportion of individuals [8]. Furthermore, zoonotic Plasmodium knowlesi, masquerading as P. malariae, has entered the human population of southeast Asia [9]. Are these newly emergent phenomena or has the application of modern tools combined with an inquisitive spirit discovered new paradigms in hitherto hidden information?

The discovery and epidemiologic characterization of P. knowlesi malaria would not have been feasible without nucleic acid technologies and access to sequence databases [9]. P. knowlesi, a parasite most often associated with natural infections of long and pigtailed macaques [10], is morphologically similar to but genetically and phenotypically distinct from P. malariae. P. knowlesi, unlike its more benign lookalike, causes severe disease in humans in approximately 10% of those infected and at least 1% of infections are fatal [11,12]. Although, in global comparative terms, the number of human infections with knowlesi are relatively small, it is now accepted as the fifth human Plasmodium and in certain foci it is an important disease [13]. Molecular evidence confirmed that knowlesi has entered the human population, yet in the early days of study there was general skepticism because the possibility of a larger focus of human P. knowlesi infections contradicted the medical texts and general malaria precepts. However, the finding, once accepted, appears to have rekindled interest in zoonotic malaria. Also highlighted was the advantage of molecular tools over the less sensitive classic microscopy, to identify the Plasmodium spp. infecting non-human vertebrate hosts, including those of primate populations. The urgent need to control and reduce the burden of infections by P. falciparum and P. vivax remain a priority but the field of zoonotic malaria and evolutionary genetics has forced a much needed, refreshing and dynamic update of some static notions, particularly on vertebrate-host specificity, that had become etched in malariology.

For example, the entry of P. knowlesi into the human population demonstrates a parasite cross event into a non-natural vertebrate host species. In addition naturally acquired human infections were geographically widespread, discontinuous with multiple co-incident entries [11]. The human host supported parasite growth and some individuals rapidly succumbed to hyperparasitemia allowing comparison of severe malaria caused by parasites from two different lineages.

Fatal knowlesi malaria has already added new perspectives on the otherwise difficult to study severe falciparum malaria [14]. That P. knowlesi was naturally permissive in humans was not entirely unexpected because experimental knowlesi infections of a variety of hosts, including humans, had appeared in the early malaria literature [10]. In addition, a single natural human infection was described in 1965 [15]. Relaxed vertebrate-host specificity in P. knowlesi was thought to be unusual among the plasmodia of primates. However the potential of this characteristic was not overlooked and P. knowlesi continues to be widely used in primate models to study malaria. By contrast a much stronger host specificity was observed when similar experiments were conducted with P. falciparum. The failure to develop a representative animal model for P. falciaprum was interpreted as extreme human-host restriction, a view that may now require re-evaluation [16,17].

Information generated from the Plasmodium spp. of old world monkeys, the macaques, and construction of molecular phylograms resulted in a consensus view that P. vivax most likely crossed into humans from Asian macaques [18]. In the effort to discover the origin of P. falciparum, molecular phylogenies were not informative. Historical reasons, including protection of the African apes, limited the information available on the Plasmodium spp. of our nearest relatives. The other primate host species studied were not infected with Plasmodium spp. related to P. falciparum. P. falciparum appeared isolated and apart from a single representative of Plasmodium reichenowi taken from a chimpanzee [19] there were no known close relatives of this lonely pair in the molecular phylogenies constructed for Plasmodium.

The single P. reichenowi isolate was experimentally maintained and has survived into the molecular era but surprisingly more representatives have not been added to the collection. Was P. reichenowi extinct or was protective legislation just too difficult to obtain approval to collect the necessary samples from the African apes? Recently more samples for the molecular detection of the Plasmodium spp. infecting the African apes have become available and exciting new information has emerged.

In 2009 Ollomo et al. discovered a new Plasmodium spp. related to P. falciparum in chimpanzees from Gabon [20]. This find was important and immediately added to the P. falciparum and P. rechenowi phylogeny. The suggested name for the species was P. gaboni and P. gaboni appeared to have emerged or diverged approximately 21 million years ago, predating the divergence of P. falciparum and P. reichenowi. At the same time Duval et al., reported that P. ovale, a human-adapted parasite, was also a parasite of chimpanzee [21]. Human P. ovale infections were, at times, missed by PCR detection methods and this was later attributed to two P. ovale ‘types’, termed classic and variant, in the human population. Duval found the P. ovale variant form in chimpanzees and a third novel ovale-related species. P. ovale had few close relatives in the phylogenies and, like P. falciparm, the ovale clade has now expanded. Duval concluded that the findings represented an exchange between humans and chimpanzee in the Cameroon where the chimpanzee’s samples were sourced. On the basis of two more recent studies on gorillas and bonobos this hypothesis may also be need to be revised [16,17]. The ovale story became even more interesting when Sutherland et al. reported that the two human-host adapted P. ovale types were two non-recombining species and perhaps the result of two independent and divergent emergences of P. ovale into humans [22]. Multiple independent entries of P. knowlesi into humans in southeast Asia would support the possibility of multiple entries of P. ovale into human or earlier Homo hominid hosts. Again this will become clearer as more human, chimp and perhaps other primate-adapted ovale sequences are added to the analyses.

Interesting as these reports are the most exciting information has come from the increased membership of the P. falciparum and P. reichenowi clade, including the addition of P. gaboni.

Last year, Rich et al. published DNA sequence data from eight new P. reichenowi isolates from chimpanzee blood samples collected in Côte d’Ivoire and Cameroon [23]. Of relevance, this represented approximately 8% positives in the animals screened. Even so the measured diversity in the eight P. reichenowi isolates was greater than all of that calculated for a larger global sample of P. falciparum and provided robust support for the hypothesis that extant P. falciparum is the result of a single emergence of P. reichenowi or common ancestor into the Homo hominid lineage. Dating this event remains difficult and has been suggested to be between 2–3 million years ago to 10,000 years ago. Nonetheless, the P. falciparum/P. reichenowi clade is now branched and more informative.

Earlier this year, two publications irrevocably changed the notion that P. falciparum is human-host specific and lacks diversity. Prugnolle et al. published results that included PCR screening for Plasmodium spp. on chimp and gorilla feces from non-captive animals in Gabon and Cameroon [17]. They were able construct a complex phylogeny of the formerly sparse P. falciparum/P. rechenowi clade. The group discovered that gorillas were infected with two new Plasmodium spp. (termed Gor A and Gor B) that were close relatives of P. falciparum and P. reichenowi. Perhaps the most intriguing discovery was that gorillas were also infected with P. falciparum[17]. When analyzed the new sequences enabled the resolution of two sister clades with 15 members between them including P. falciparum, P. reichenowi and P. Gor B in clade B and P. Gor A and P. gaboni in clade A. Suddenly we find that P. falciparum has many close relatives, a find that will surely result in a paradigm shift on how we rationalize Plasmodium spp. specificity, host switch events and lack of diversity in the human-host adapted Plasmodium spp. Concurrent with the report of P. falciparum in gorillas and the worrying implication of reservoir parasite populations to the malaria eradication program, Krief et al. reported that bonobos are also infected with P. falciparum[16]. The bonobo isolates were genetically distant from each other and from the much less divergent global P. falciparum population sample from human infections. Although reaffirming animal reservoirs of P. falciparum the Krief report gives some reason for optimism in that there was no evidence of recent crossing of P. falciparum from bonobos to humans or vice versa. This tends to support the single crossover historical event followed by adaptation to the human host for the human-associated P. falciparum population. The result also hints at an evolutionary event much more complex than the vertebrate-host flexibility currently observed in P. knowlesi infections of humans.

No doubt there will be much sifting through these recent reports and it is expected that the phylogenies will be refined as more genes, more representatives of each species and more species are identified. In the meantime the malaria community needs to properly digest the full significance of these reports. Particularly, in assessing just how threatening these parasite reservoirs, including P. knowlesi and P. ovale, are to the resurrected support for a new malaria eradication program [24].

The nonhuman apes appear to be living well with their P. falciparum isolates and one must consider that these are stable host–parasite relationships. How old are they, when did extant P. falciparum enter Homo hominids, and why, are questions that may just be within our grasp to properly address.

We now know P. falciparum is diverse and is not human-host restricted but are we any closer to the study of P. falciparum pathophysiology and immunity in a permissive animal model? Probably not in the nonhuman great apes. However, the finding that severe knowlesi malaria in humans has notable similarities with severe falciparum malaria is important [14]. The well-established permissive animal models for P. knowlesi together with genome comparisons of P. falciparum and its various hosts will enable cause and effect studies on the pathophysiology of malaria. The combined new information will, if not help us to eradicate malaria, provide essential clues to elicit disease immunity in the highly susceptible human host.