Plasmodium ovale curtisi and Plasmodium ovale wallikeri circulate simultaneously in African communities

Abstract

It has been proposed that ovale malaria in humans is caused by two closely related but distinct species of malaria parasite, Plasmodium ovale curtisi and Plasmodium ovale wallikeri. It was recently shown that these two parasite types are sympatric at the country level. However, it remains possible that localised geographic, temporal or ecological barriers exist within endemic countries which prevent recombination between the genomes of the two species. Here, using conventional and real-time quantitative PCR (qPCR) methods specifically designed to discriminate P. o. curtisi and P. o. wallikeri, it is shown that both species are present among clinic attendees in Congo-Brazzaville, and occur simultaneously both in lake-side and inland districts in Uganda and on Bioko Island, Equatorial Guinea. Thus P. o. curtisi and P. o. wallikeri in these localities are exactly sympatric in both time and space. These findings are consistent with the existence of a biological barrier, rather than geographical or ecological factors, preventing recombination between P. o. curtisi and P. o. wallikeri. In cross-sectional surveys carried out in Uganda and Bioko, our results show that infections with P. ovale spp. are more common than previously thought, occurring at a frequency of 1-6% in population samples, with both proposed species contributing to ovale malaria in six sites. Malaria elimination programmes in Africa need to include strategies for control of P. o. curtisi and P. o. wallikeri.

Graphical abstract

Fig. 1

Discrimination between Plasmodium ovale curtisi (Poc) and Plasmodium ovale wallikeri (Pow) using nested PCR amplification of the repeat region of poctra and powtra genes. (A) Alignment of the translated amino acid sequence encoded by a large portion of the P. ovale sp. tryptophan-rich antigen gene (potra), showing the four different forms sequenced (Poc, Poc + 6, Pow1, Pow2). Black shading indicates amino acid identity in at least three sequences, grey shading indicates the occurrence of different amino acids with similar biochemical properties such as charge, size or hydrophobicity, and absence of shading indicates the presence of different amino acids with distinct biochemical properties. Amplification primers (PoTRA5 pair) are situated at the arrows, and are designed to maximise visible size differences in PCR products by gel electrophoresis. Some Nigerian isolates of P. o. curtisi have an additional six amino acids in the repeat region (Poc + 6) (Sutherland et al., 2010). (B) Agarose gel showing size distinction between Poc and Pow isolates after a single PCR using PoTRA5 primers. Fragment sizes: poctra = 317 bp; poctra + 6 = 335 bp; powtra = 245 bp. (a) shows four isolates from Bioko Island, Equatorial Guinea (one Poc from Punta Europa, one Poc from Luba, two Pow from Luba) and four from Bogoigo, Uganda. Previously-characterised isolates of each species from the UK Malaria Reference Laboratory are shown in (b).

Fig. 2

Discrimination between Plasmodium ovale curtisi and Plasmodium ovale wallikeri using melt profile differences. (A) Alignment of a 120 bp region of the porbp2 (reticulocyte-binding protein 2) gene between amplification primers Porbp2TMfwd and Porbp2TMrev of P. o. curtisi (pocrbp2) and P. o. wallikeri (powrbp2), showing mismatches (boxed) predicted to affect product T_m. (B) Output of melt analysis on a Rotorgene thermo-cycler, expressed as the change in fluorescence over time (δF/δT), displaying clear discrimination between the two species: (a) the amplification product of P. o. curtisi exhibits a melt peak at 73 °C, whereas that of P. o. wallikeri exhibits a melt peak at 74 °C. (b) Melt profiles of isolates (M70C1, M02C2) from two children sampled in Uganda, each run in duplicate. All controls and samples were run in duplicate.