The current study provides information on genetic polymorphisms in the pfmdr1 gene in P. falciparum isolates from Jazan region, a low malaria transmission area, 11 years after the discontinuation of CQ and the adoption of ACT in its place. The findings revealed that 11.5% and 86.3% of the isolates carried the pfmdr1 86Y and 184F point mutations, respectively. Accordingly, three pfmdr1 haplotypes were identified, with the single (NFSND) and double (YFSND) mutated haplotypes found in 74.9% and 11.4% of the isolates, respectively.
The prevalence of the pfmdr1 86Y mutation found by the current study is far lower than that reported by the only two studies available from Saudi Arabia, which were conducted in Jazan region in 2010 and 201226,27. In 2010, the 86Y mutation was found in 40.4% of 109 P. falciparum isolates from the Aseer and Jazan regions27. Two years later, 30% of 165 P. falciparum isolates from Jazan region were found to be carrying the pfmdr1 86Y point mutation26. Moreover, the current finding that the 184F point mutation dominated the pfmdr1 polymorphism of the analysed isolates is consistent with the sole local study that reported 96% (159/165) of this mutation in the region despite an approximately 10% decline demonstrated by our result26. Likewise, the current study found no mutations at codons 1034, 1042 and 1246, similar to the study conducted in Jazan region in 201226.
The current findings suggest that there has been a significant decline in the prevalence of pfmdr1 86Y (and conversely an increase in the wild N86 allele) since the implementation of the ACT treatment policy in Saudi Arabia, which is consistent with previous reports elsewhere15,17,28,29. However, a conclusion based on two previous studies26,27 and the current study might not be reliable. Therefore, further analyses of the SNPs of the pfmdr1 gene over time using archived samples are required to enable a better understanding of the trend in pfmdr1 mutations in Jazan region. Interestingly, the current findings corroborate the results of our separately published study on the same isolates, which found a significant decline in the prevalence of pfcrt 76T and the re-emergence of pfcrt K76 CQ-sensitive P. falciparum strains in Jazan region30. Altogether, about two-thirds (69.2%; 18/26) of the isolates that carried the pfmdr1 86Y mutation were also found to be carrying the pfcrt 76T mutation. The pfmdr1 86Y point mutation is the second most important molecular marker of CQ resistance after pfcrt 76T, and several studies have suggested that there may be a synergistic association between these pfcrt and pfmdr1 point mutations31,32. Moreover, a remarkable decline in the prevalence of mutant pfcrt 76T and pfmdr1 86Y alleles has also been observed in low malaria transmission settings where drug pressure is likely to be lower33.
In Yemen, a neighbouring country with high malaria transmission, previous studies showed that the prevalence of the 86Y mutation was higher than that reported by the current study, varying between 16–20% in the governorates of Hadhramoud (southeastern Yemen) and Hodeidah (northwestern Yemen)34,35,36. However, striking results were reported in Taiz governorate (southwestern Yemen) in which all of the 50 isolates analysed were found to be carrying the wild N86 allele37. On the other hand, mutation at codon 184 was found at a fixation level (100%)34,36. Interestingly, a high prevalence of mutations (up to 70%) for 1034C and 1042D has been reported across Yemen34,36. In Iran, the 86Y mutation was detected in 23% of isolates while mutations at the other four codons were not detected in any isolate38. Variations in the geographical distribution of pfmdr1 mutations at codons 86 and 184 and haplotypes have also been reported in different African and Asian countries13. The current study also found that there was significant variation in the distribution of pfmdr1 polymorphism, with the highest prevalence of the 86Y mutation and YFSND haplotype in isolates from Aldair and Alharth governorates. These governorates border Yemen and most of the isolates were from Yemeni patients. Thus, the higher percentage of 86Y and YFSND might be attributed to the source of malaria infection circulating in northern Yemen.
The current study also demonstrated that three pfmdr1 haplotypes are circulating in Jazan region, with the single-mutant NFSND (74.9%) dominant followed by the wild NYSND (13.7%), and the double-mutant YFSND (11.4%) haplotypes. This pattern is different from that reported in 2012, where 66% and 31% of the examined isolates carried the single NFSN and double YFSN haplotypes while only 3% carried the wild NYSN haplotype (codon 1246 was not involved)26. Studies from other countries such as Yemen and Ghana have reported up to five different haplotypes32,36.
Thus, the mutations of pfmdr1 gene are complex. Moreover, the mechanism for modulating multiple antimalarial drug resistance in P. falciparum parasites as well as their associations with the mutations of other genes such as pfcrt is not well understood. Mutation at codon 86 has been associated with CQ resistance in several studies from different regions while mutations at other codons of the pfmdr1 gene have been found to have complementary roles13. Some previous studies have reported that mutation at codon 86, either alone or in combination with mutations at codons 184 and 1246 (YFSNY haplotype), enhances the susceptibility of P. falciparum to ACT, specifically AS + MFQ (the first-line treatment in Saudi Arabia)16,39. On the other hand, it has been reported that the carrying of the wild-type N86 together with the mutant 184F and wild D1246 (NFSND haplotype) enhances parasite tolerance to AL32,40. The prevalence of the NFSND haplotype increased in different countries such as Yemen, Mozambique, Gabon, and Tanzania after the introduction of AL treatment16,36,41,42.
Based on the findings of the current study, it is possible that P. falciparum resistance to AL (the second-line treatment for uncomplicated falciparum malaria in Saudi Arabia) may emerge in Jazan region in the near future. However, this hypothesis needs further evaluation based on continued in vitro and in vivo parasite susceptibility testing to the drugs plus monitoring of the molecular markers of resistance to antimalarial drugs. Moreover, a selection of the pfmdr1 86Y and 1246Y mutations has been observed post-AQ exposure43, whereas mutations at codons 1034C, 1042D and 1246Y have been found to be associated with the low parasite susceptibility to QN44. In the current study, these mutations (1034C, 1042D and 1246Y) were not detected. Thus, it could be concluded that QN treatment, which is the third-line treatment (or the rescue treatment) for both uncomplicated and severe falciparum malaria in Saudi Arabia remains effective. However, it should be noted that AQ has not been used in Saudi Arabia. As a whole, the different patterns of circulating pfmdr1 haplotypes that have been identified thus far should be considered when formulating and updating antimalaria drug strategies in endemic areas.
In addition, the current study found significant associations between the pfmdr1 86Y mutation and the YFSND haplotype and the nationality and the residence of the participants while associations with age and gender were not significant. The percentages of 86Y and YFSND in isolates from non-Saudi patients were significantly higher than in isolates from Saudi patients. This is consistent with the fact that more than half (53.8%; 14/26) of these markers (86Y and YFSND) were detected in isolates from Yemeni patients. These results corroborate the results of our separately published study on the same isolates that showed that 95% of the isolates from Yemeni patients carried the pfcrt 76T mutations30. In a previous study in Jazan, associations between pfmdr1 markers and the patients’ nationality and gender were found to be not significant26. Similarly, studies from Yemen, Sudan and Qatar found no association of pfmdr1 mutations with the age and gender of patients31,35,36,45. In contrast, a study in Kenya found that males are more likely to harbour CQ-resistant P. falciparum parasites with high pfmdr1 mutations as compared females. This difference was attributed to a practice among males of delayed treatment seeking, which may have placed them at higher risk of accumulating super-infections and developing resistance29.
The current study also demonstrated that the percentages of the pfmdr1 86Y mutation and the YFSND haplotype in isolates from patients living in urban areas were almost three times higher than in isolates from patients in rural areas while the percentages of the 184F mutation and the NFSND haplotype were comparable between both groups. While there is as yet no clear explanation for this finding, a previous study in Burkina Faso attributed a higher rate of CQ failure in urban settings to higher drug pressure in those areas as compared to rural areas46. However, conversely, other studies have found a higher prevalence of the 86Y mutation in rural areas, and this has been explained by the practice of self-medication41,47. In addition, the findings of the current study also revealed that the pfmdr1 86Y mutation and the YFSND haplotype occurred more at low parasitaemia, which is consistent with findings by earlier studies elsewhere48,49,50. Although causality cannot be inferred, the association between the pfmdr1 86Y mutation and the parasitaemia level has been linked to multiplicity of infection and parasite virulence, with parasites with decreased virulence carrying the mutant pfmdr1 86Y allele50,51.