Polymorphisms analysis of the Plasmodium ovale tryptophan-rich antigen gene (potra) from imported malaria cases in Henan Province

Plasmodium ovale was first described in 1922 [1], as the fourth malaria parasite of humans [2, 3]. Generally, a P. ovale infection is of low parasitaemia, and the morphology of the parasite is similar to Plasmodium vivax. Also, it frequently presents as a mixed infection with the other Plasmodium species [4‐9]. As a result, P. ovale attracted less attention compared to other species, and its prevalence has apparently been underestimated. It has long been considered predominantly found in Africa and some islands of Western Pacific [10, 11], with confirmed cases occasionally found in other endemic regions [12, 13].

Currently, P. ovale spp. may be classified into two different subspecies by molecular genotyping: P. o. curtisi (classic type) and P. o. wallikeri (variant type) [14]. The nuclear genome sequences further confirmed that the two species were genetically different, but morphologically indistinct [15], and their duration of latency were seemly different [16]. Both species were considered to exist sympatrically in Africa and Asia, and even both parasites were infected simultaneously [17‐21].

Because of the generally low parasitaemia of P. ovale infections, sensitive molecular methods to detect and identify the two subspecies must be used in future investigations, with polymorphic markers as a method to discriminate the different strains. Many protocols showed that the SS rRNA genes [17, 22, 23] were suitable for identification but not for genotyping. The recent study showed that the gene encoding P. ovale tryptophan-rich antigen (potra) could be used to distinguish the two P. ovale subspecies [14]. The sequence and size of the tryptophan-rich antigen gene was variable among the P. ovale subspecies (poctra and powtra) [14]. A nested PCR detection assay was exploited to discriminate the species by the size of the amplified fragments (299 or 317 bp for poctra; 245 bp for powtra), where the conserved sequences were chosen as primers for these two genes [19]. Additionally, a semi-nested PCR protocol was developed by Tanomsing et al. [24] with which the two P. ovale subspecies could be discriminated efficiently, and more fragment sizes were found comparing with previous studies, the 299 bp fragment was overlapping between the two subspecies. This would invalidate amplified fragment size difference, as a means of distinguishing between P. o. curtisi and P. o. wallikeri. The amplified fragment size variations resulted from differences in the number of repeated units, which suggested that a broader range of size variants might occur. In this study, more variations of potra gene were observed.

Malaria elimination is a long-term goal to be achieved worldwide. As one species of human Plasmodium, the identification of P. ovale is more widespread than formerly known. Plasmodium ovale, like P. vivax, has hypnozoites that cause relapses [26, 27], and it consists of two different subspecies: P. ovale curtisi and P. ovale wallikeri [14]. Therefore, the differentiation of the two P. ovale species, especially with respect of molecular phylogeny will need to be better understood.

In 2011, Oguike et al. [19] published the discrimination of the two P. ovale subspecies by the size of the amplified fragments of the potra gene (299 or 317 bp for P. o. curtisi; 245 bp for P. o. wallikeri), using nested PCR. Although this technique was specific for P. ovale spp., the sizes of the amplified fragment varied with the number of repeat units, which reduced the discrimination between species: for P. o. curtisi, 299, 317 and 353 bp, and for P. o. wallikeri, 245, 299 and 335 bp [24]. Tanomsing et al. [24] suggested the number of potra size variations might be more than those evaluated and this speculation was confirmed in this study. Using the same primes and method, more fragment sizes were identified in this study, while some sizes overlapped between the two subspecies: for P. o. curtisi having sizes 299, 317 and 335 bp, and for P. o. wallikeri having sizes 227, 245, 263, 281 and 299 bp. The results of the three studies [19, 24] were also combined, as shown in Table 1. Four different sizes for P. o. curtisi and six sizes for P. o. wallikeri have been reported, and that the fragment sizes of 299 and 335 bp were overlaps between the two species. As more samples were analysed, it was likely that the number of potra size variants would be more than expected. Conceivably, more size variants may be identified in future studies.

Potra gene was used to discriminate the two P. ovale species, because the tryptophan-rich antigen was encoded by a repeat pattern of variable length 3-amino acid [14]. Sutherland et al. [14] and Oguike et al. [19] had also proposed that the potra gene of P. o. curtisi (poctra) could be identified by the pattern of the six amino acids of the repeat region, TITPIS, while the potra gene of P. o. wallikeri (powtra) were different in two non-synonymous positions. By alignment of the amino acid sequence of potra fragments, 11 genotypes of potra were found from the 110 isolates in this study. Combination with the previous studies of Oguike et al. [19] and Tanomsing et al. [24], showed ten subtypes of potra gene for P. o. wallikeri and four subtypes for P. o. curtisi, and 14 genotypes of potra gene were under analysis. This study showed that the amino acid sequence of potra fragments were composed with multiple amino acid units, six amino acids were as one unit. There was different dominant amino acid repeat of potra for the two P. ovale species, which could be used to discriminate the subtype of P. ovale spp. The repeat of six amino acids as one unit to analyse the difference of the genotypes between the two P. ovale species was first reported, which could make the results more clear and simple. The sizes of 277, 245 and 263 bp for P. o. wallikeri all had two different subtypes, and this phenomenon did not find in the P. o. curtisi.

The sizes of the reference sequences KX417700–KX417704 [25] from Genbank were short for discriminating the differences of dominant amino acid repeats. In our study, the genetic relationship of the P. ovale spp. was analysed by the neighbour-joining tree mainly based on the number of the dominant amino acid repeats including MANPIN, AITPIN, TINPIN or TITPIS.

The distribution of the two P. ovale spp. was different. The isolates of P. o. wallikeri was more than that of P. o. curtisi, and the genotype of the 245 bp size was the predominant type for P. o. wallikeri in most Africa countries, but the other genotypes were less. For P. o. curtisi, the genotypes of the 299 and 317 bp size were commonly in Africa and the genotype of 335 bp size was less.

Molecular epidemiological studies on genetic diversity of Plasmodium vivax have been based mainly on single copy polymorphic genes which code for parasite surface antigens such as circumsporozoite protein (csp), merozoite surface protein-1 (msp-1) and merozoite surface protein 3 alpha (msp 3α) [28]. Pvcsp comprises of central domain of tandem repeated sequences flanked by two non-repeated conserved sequences [29‐32]. Two types of repeat elements, either VK210 or VK247 types were detected in clinical isolates of P. vivax and thus pvcsp serves as a useful tool for genotyping [33, 34]. Potra has the similar characteristics with pvcsp, which also could be used for parasite genotyping.

Considering the change of malaria epidemiology and the approaching of malaria elimination, P. ovale spp. deserves more attention. Molecular techniques are a good tool for detecting and identifying the two P. ovale subspecies and their relative distribution.