Molecular surveillance of anti-malarial resistance Pfdhfr and Pfdhps polymorphisms in African and Southeast Asia Plasmodium falciparum imported parasites to Wuhan, China

Blood samples were collected from P. falciparum-infected migrant patients with uncomplicated malaria in Wuhan of Hubei Province from 2011 to 2019. These samples were examined by microscopy with stained thick and thin blood smears and detected by rapid diagnostic tests (RDTs) for Pf-HRP2 and pLDH, as previously described [16‐18]. RDTs, were carried out according to the manufacturer's manual (Wondfo, Guangzhou, China). Subsequently, two or three drops of blood were spotted on Whatman 3MM filter paper, air dried, and stored in an individually coded sealed plastic bag containing silica desiccant beads. The bags were stored at 4 °C until use. The study was approved by the Ethical Review Committee of the Hubei University of Medicine and Wuhan City Center for Disease Prevention and Control Ethics Committee. Informed consent was obtained and signed by all participants or their guardians before inclusion in the study and before sample collection.

Genomic DNA (gDNA) from dried blood spot samples was isolated using the TIANamp Blood DNA Kit (Tiangen Biotech Co., Ltd., Beijing, China) following the manufacturer’s recommendations. DNA samples were stored at − 20 °C for further genotyping.

To identify polymorphisms in the Pfdhfr (Gene ID: PF3D7_0417200) and Pfdhps (Gene ID: PF3D7_0810880) genes, purified gDNA templates were amplified by using nested PCR in the Mini MJ Thermal Cycler (Bio-Rad), following protocols described previously with minor changes [19, 20]. Briefly, primary PCR, was performed in a total of 20 μl containing 10 μl of 2 × Phusion PCR Master Mix (40 units/ml Phusion DNA polymerase, 400 μM deoxynucleoside triphosphate [dNTP] mixture, 2 × Phusion high-fidelity [HF] buffer, and 3 mM Mg²⁺), 1 μl of each primer (10 μM), and 2 μl of gDNA. For the second round, 1.0 μl of primary PCR products were amplified with a 50 μl reaction system. All PCR conditions conducted were as follows: predenaturation at 95 °C for 3 min, followed by 30 cycles of denaturing for 30 s at 95 °C, annealing for 30 s at 55 °C and extension at 72 °C for 30 s, plus a final extension at 72 °C for 5 min. The PCR product was electrophoresed on a 1.0% agarose gel stained with GelRed®Safe DNA Gel Stain (Boitium, USA) and visualized with a UV transilluminator. Sequencing was outsourced to Genewiz, Soochow, China, whereby for the Pfdhfr gene, the nested PCR products were purified and sequencing in reverse directions, while for the Pfdhps gene, the nested PCR products were purified and sequenced by Bidirectional DNA sequencing. Polymorphisms were analysed by creating consensus nucleotide sequences with the reference sequences in PlasmoDB using DNAStar (DNASTAR Inc., Madison, WI, USA).

The frequencies of single-nucleotide polymorphisms (SNPs) and haplotypes were calculated as the percentage of the number of successful sequencing samples. Mixed infections containing wild-type and mutation were excluded from further analysis. The comparison and trend estimation of haplotypes were assessed by unpaired T-test and linear regression analysis respectively with GraphPad Prism 5.

In total, 303 isolates from 2011 to 2019 were included for the analysis of Pfdhfr and Pfdhps mutations. Table 1 summarizes the different SNPs observed in the samples collected over these years. For Pfdhfr, mutations at different loci, including N51I, C59R, S108N, and I164L, were investigated. Of the 300 samples successfully genotyped for the Pfdhfr gene, 90.7% (272/300) harboured the mutant allele N51I, while six samples (2.0%) had a mixed type. For the Pfdhfr codon at 59, 84.7% (254/300) samples had the mutant allele C59R, and 9 (3%) had mixed genotypes. At codon 108, 94.7% (284/300) harboured a mutation, 3.0% (9/300) were wild type, and 2.3% (7/300) mixed infection were found. Only 1 sample out of 300 (0.3%) derived from Myanmar in 2011 had the highly drug resistant mutation I164L. However, we did not observe any mutation at codon 50 of Pfdhfr. For Pfdhps, mutations at different loci, including S436A, A437G, K540E, A581G, and A613S, were surveyed. Of the 290 (290/303, 95.7%) successful genotyped samples, 27.2% of the isolates harboured S436A, 79.0% harboured A437G, 21.7% harboured K540E, 9.0% harboured A581G and 11.0% harboured A613S. For wild type, 66.0% of the isolates harboured S436A, 13.1% A437G, 70.3% K540E, 85.5% A581G and 89.0% A613S, respectively. For mixed genotypes, 6.9% of the isolates harboured S436A, 7.9% harboured A437G, 7.9% harboured K540E, 5.5% harboured A581G and 0.3% harboured A613S.

In total, five unique haplotypes at the Pfdhfr locus and 11 haplotypes at the Pfdhps locus were found. The prevalence of the Pfdhfr and Pfdhps haplotypes in different years and various geographical regions are illustrated in Fig. 1. The Pfdhfr mutations were the most prevalent, with the triple mutation (IRNI) in almost 84.4% (238/282). Compared to the haplotypes of NCSI, ICNI, and NRNI, the IRNI displayed a higher prevalence during 2011–2019 (P < 0.0001, P < 0.0001, and P < 0.0001, respectively) (Fig. 1a). As the predominant haplotype, the IRNI was found in all geographical areas in Africa and SEA (Fig. 1b). For NCSI (wild type), the frequency has increased from none in 2011 and 2012 to 9.1% (3/33) in 2018, and then reduced to 2.5% (1/40) in 2019. The relative frequencies of the ICNI mutations decreased from 16.67% (1/6) in 2011 to 0 in 2012 and increased to 14.3% (6/42) in 2013; finally stabilized at 15.0% (6/40) in 2019. For NRNI, there was only a marginal difference over these years (F = 0.6316, P = 0.4529). The allele with quadruple mutations (IRNL), which conferred a high level of resistance to antifolates, was only found at a low frequency (0.3%, 1/300) in 2011. For Pfdhps, the predominant haplotype was single-mutant SGKAA (37.8%, 90/238). It was increased from 14.3% (1/7) in 2011 to 30.0% (9/30) in 2019 (Fig. 1c). Although a decreasing trend was observed in the prevalence of the quadruple-mutant AGKGS from 18.5% in 2012 to 9.8% in 2016 and finally to 0% in 2019, these differences were not statistically significant (F = 0.0063, P = 0.9391). Similarly, the prevalence of the Pfdhps wild type SAKAA genotype has decreased from 42.9% in 2011 to 3.33% in 2019, but it was not statistically significant (F = 4.456, P = 0.0727). The Pfdhps double-mutant AGKAA genotype decreased from 14.3% in 2011 to 7.3% in 2016 and finally increased to 16.7% in 2019 (F = 0.6879, P = 0.4342). A concomitant increase in the prevalence of the Pfdhps double-mutant SGEAA genotype was observed, increasing in prevalence from 3.7% in 2012 to 33.3% in 2019, which was statistically significant (F = 9.034, P = 0.0198) (Fig. 1c). As the predominant haplotype, the SGKAA was mostly found in WA, CA, and SA with a prevalence of 93.4, 47.5, and 42.9%, respectively (Fig. 1d). SGEAA, was mostly distributed in SA (36.7%), EA (76.2%), and CA (17.0%) (Fig. 1d).

A total of 230 samples were subjected to combined haplotypes analysis. The 28 haplotypes were verified by combining both genes. The most common haplotype was IRNI-SGKAA with a 30.0% (69/230) frequency, followed by IRNI-SGEAA (20.4%, 47/230), IRNI-AGKAA (9.8%, 22/230), and IRNI-AGKGS (7.8%, 18/230). Moreover, the frequency distribution of the different Pfdhfr-Pfdhps haplotypes was compared among the analysed samples collected at different times. The results showed that the most prevalent haplotype observed during the study period was IRNI-SGKAA, which also remained the most frequent one in WA (32.7%), SA (23.4%), and CA (44.4%) (Fig. 1e). For IRNI-SGEAA, an increasing trend was detected during 2011–2016, and 2017–2019, respectively (Fig. 1e). For regional distribution, IRNI-SGKAA was mainly distributed in WA (32.7%), CA (44.4%), and SA (23.4%) (Fig. 1F). Similarly, IRNI-SGEAA was mainly found in EA (71.4%) and SA (31.9%), followed by CA (16.7%) and WA (7.1%) (Fig. 1f). An additional 20 minor haplotypes with a prevalence of less than 2% constituted only 9.1% (21/227) of the overall haplotypes (Additional file 1: Table S1 and Additional file 2: Table S2).