Variant Plasmodium ovale isolated from a patient infected in Ghana

It is usually assumed that the four major human malaria species Plasmodium falciparum, Plasmodium vivax, Plasmodium ovale and Plasmodium malariae have no animal reservoirs and that zoonotic Plasmodium infections in humans are highly unusual. The recent finding that Plasmodium knowlesi is found in several countries in Southeast Asia, where it has been wrongly classified as P. malariae, shows that molecular typing may change the understanding of human malaria as an infection strictly transmitted between humans[1, 2]. Plasmodium ovale was first described from West Africa in 1922 and subsequently from every continent[3]. In West Africa, a blood film P. ovale parasite positive rate between 0.7% and 10% has been found[4]. However, in other areas the occurrence of isolated cases is difficult to explain without an animal reservoir[3].

Characterization of P. ovale from Southeast Asia based on the small subunit rRNA gene and parts of the cysteine protease, ookinete surface protein and cytochrome b genes indicate that P. ovale can be divided into at least two types, classic and variant, which do not differ morphologically[5, 6]. Variant P. ovale seems to be associated with higher parasite density in humans[7‐9]. A recent study of 55 P. ovale isolates from around the world clearly showed that variant and classic P. ovale co-exist and do not recombine[10].

A study comparing sequences from the cytochrome b, cytochrome c oxidase 1 and lactate dehydrogenase genes of Plasmodium spp. from gorilla (Gorilla gorilla) and chimpanzees (Pan troglodytes) with P. ovale from humans found one variant P. ovale in a chimpanzee closely related to the human variant[11].

This study describes the molecular characterization of a P. ovale isolate from a patient infected in Ghana, showing that the isolate is a variant P. ovale based on the characterization of five genes, and that it is identical or highly similar to a P. ovale variant isolated from a chimpanzee[11] and from humans in two separate studies[5, 10]. The time of the phylogenetic split between the two species is estimated at 1,7 million years ago.

Fever developed four months after a two-week visit to Ghana and at admission, the patient had a microscopically diagnosed P. ovale infection with a parasitaemia of 0.5% and the presence of asexual forms, early schizonts and gametocytes (Figure 1). A rapid test (Binax^®) was negative. Because of the recent reports of variants of P. ovale and because the patient had a negative rapid test, the isolate was sequenced for further characterization.

Part of the asexually transcribed small subunit (18S) ribosomal DNA (possrdna) gene from the isolate was compared with classic and variant sequences published in recent studies from Asia and Africa[5, 10], revealing closer resemblance to the variant type from both studies (table 1). To further characterize the isolate, the pocytb, pocysp, pocox1 and poldh genes were sequenced and compared with sequences from three recent studies. The isolate sequences were 100% identical to the pocytb variant type sequences reported in all three studies[5, 10, 11], 99,8% identical to variant type cysp from Asian isolates[5], and 100% identical to variant type pocox1 and poldh from African isolates[11]. The isolate was also compared to pocytb, pocox1 and poldh sequences isolated from a chimpanzee[11], revealing 100% identity in all three genes (Table 1).

A phylogenetic tree was constructed based on the concatenated sequences of all five genes studied and sequences published in GenBank (Figure 2). The P. ovale variant type together with the present isolate formed a clade related to but separate from the classic type. Calculation of phylogenetic trees based on individual genes yielded equivalent results. To estimate the split between variant and classic P. ovale, Bayesian Inference was used as implemented in the BEAST software[12] with a calibration point of 6 Million Years (MYs) +/- 0.5 MYs for the human/chimpanzee speciation event as represented by the P. falciparum/P. reichenowi split, which gave an estimated age of 1.7 MY (95% Highest Posterior Density 0.25-4.44MY) for the most recent common ancestor of the variant and classic P. ovale.

The recent study of 55 P. ovale isolates from around the world clearly showed that P. ovale can be divided in two species: classic and variant (suggested named Plasmodium ovale curtisi and Plasmodium ovale wallikeri) [10]. The two species currently circulate side by side in the same geographical areas and are not recombining even though they infect the same host. It is not known whether variant P. ovale continues to circulate in both humans and primates and it is intriguing that the variant and classic strains do not recombine. One explanation for the speciation could be specificity for different Anopheles species[10, 13], another possibility is receptor divergence. Humans ceased producing N-glycolylneuraminic acid (Neu5Gc) around 2 to 3 million years ago but continued producing N-acetylneuraminic acid (Neu5Ac) in contrast to old-world primates who synthesize both[14]. Neu5Gc has been shown to interfere with the binding of P. falciparum Erythrocyte-Binding Antigen 175 (PfEBL-175) to chimpanzee erythrocytes, possibly explaining why P. falciparum does not infect chimpanzees[15]. The evolutionary timing of the loss of Neu5Gc coincides with our estimated divergence time of 1.7 million years, which is in agreement with the estimate of Sutherland et al. [10], and could explain the split between the two P. ovale variants by a similar mechanism in which only classic P. ovale is sensitive to Neu5Gc interference.

The chimpanzee can be infected with P. ovale, but the infection is self-limiting[16, 17], though it is not known whether these experiments were performed with variant or classic P. ovale. P. ovale in chimpanzees closely related or identical to human variant P. ovale has been described in Cameroon[11], and three different genes from the isolate described here (pocytb, pocox1 and poldh) are 100% identical to this variant. This suggests that the P. ovale used by Bray[16, 17] was a classic P. ovale strain. The data presented here further indicate that P. ovale which naturally infects chimpanzees can also infect humans, and that the isolate described here is identical or closely related to the variant isolates sampled in South-East Asia[5, 10] and Africa [10]. All five genes sequenced here support the segregation of P. ovale into two distinct species.

Molecular typing has clearly demonstrated that P. knowlesi infections in Southeast Asia has been misdiagnosed as P. malariae and that P. knowlesi is much more widespread than previously thought[1, 2]. Few data exist on P. falciparum, but the recent finding of a new Plasmodium in chimpanzees in Gabon closely related to P. falciparum and P. reichenowi[18, 19], show that more studies are needed to fully understand the dynamics of malaria transmission between humans and potential animal reservoirs.

Written informed consent was obtained from the patient for publication of this case report and accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.