Implementation of a novel PCR based method for detecting malaria parasites from naturally infected mosquitoes in Papua New Guinea.

Background: Detection of Plasmodium species in mosquitoes is important for designing vector control studies. However, most of the PCR-based detection methods show some potential limitations. The objective of this study was to introduce an effective PCR-based method for detecting Plasmodium vivax and Plasmodium falciparum from the field-caught mosquitoes of Papua New Guinea.

Methods: A method has been developed to concurrently detect mitochondrial cytochrome b (Cyt b) of four human Plasmodium species using PCR (Cytb-PCR). To particularly discriminate P. falciparum from P. vivax, Plasmodium ovale and Plasmodium malariae, a polymerase chain reaction-repeated fragment length polymorphism (PCR-RFLP) has further been developed to use with this method. However, due to limited samples number of P. ovale and P. malariae; this study was mainly confined to P. vivax and P. falciparum. The efficiency of Cytb-PCR was evaluated by comparing it with two 'gold standards' enzyme linked immunosorbent assay specific for circumsporozoite protein (CS-ELISA) using artificially infected mosquitoes; and nested PCR specific for small subunit ribosomal RNA (SSUrRNA) using field caught mosquitoes collected from three areas (Kaboibus, Wingei, and Jawia) of the East Sepic Province of Papua New Guinea.

Results: A total of 90 mosquitoes were artificially infected with three strains of Plasmodium: P. vivax-210 (n = 30), P. vivax-247 (n = 30) and P. falciparum (n = 30). These infected mosquitoes along with another 32 unfed mosquitoes were first checked for the presence of Plasmodium infection by CS-ELISA, and later the same samples were compared with the Cytb-PCR. CS-ELISA for P. vivax-210, P. vivax-247 and P. falciparum detected positive infection in 30, 19 and 18 mosquitoes respectively; whereas Cytb-PCR detected 27, 16 and 16 infections, respectively. The comparison revealed a close agreement between the two assays (kappa = 0.862, 0.842 and 0.894, respectively for Pv-210, Pv-247 and P. falciparum groups). It was found that the eight CS-ELISA-positive mosquitoes detected negative by Cytb-PCR were false-positive results. The lowest detection limit of this Cytb-PCR was 10 sporozoites. A highly concordance result was also found between nested PCR and Cytb-PCR using 107 field caught mosquitoes, and both tests concordantly detected P. falciparum in an Anopheles punctulatus mosquito collected from Kaboibus. Both tests thus suggested an overall sporozoite rate of 0.9% (1/107) in the study areas. Subsequently, PCR-RFLP efficiently discriminated P. falciparum from P. vivax for all of the Cytb-PCR positive samples.

Conclusion: A single step PCR based method has been introduced here that is highly sensitive, efficient and reliable for identifying P. vivax and P. falciparum from mosquitoes. The reliability of the technique was confirmed by its ability to detect Plasmodium as efficiently as those of CS-ELISA and nested PCR. Application of the assay offers the opportunity to detect vector species of Papua New Guinea and may contribute for designing further vector control programmes.