Study site and period
Ujjain district is located in the western part of Madhya Pradesh, central India. The population of the district is 1.9 million as per 2011 Census [19]. The climate is tropical and transmission may occur throughout the year provided the relative humidity levels support the vector survival. Ujjain district has low transmission of malaria with an annual parasite index (API) <0.1 [20, 21]. Peak malaria transmission occurs during the warm and humid months of July to September. Data collection was from June to October in 2009, 2010, 2015 and 2016. Data from 2009 and 2010 were published previously and these samples used for K13 analyses only in the current study.
Recruitment of patients and sample collection
Samples and data were collected by the nine major pathology laboratories located in Ujjain city, Madhya Pradesh, India. Individuals or groups of pathologists reported the results from participating laboratories. Laboratories used microscopy to examine peripheral blood smears or rapid diagnostic tests (RDT) for the diagnosis of P. falciparum malaria. Inclusion criteria were microscopically or RDT verified malaria. For all patients that were smear positive for malaria, a drop of blood was put onto filter papers (WhatmannTM 3MM). RDTs were collected from patients in whom RDT only was used to diagnose malaria. RDTs or filter papers were labelled with the patient’s age and sex, dried and then placed inside individual sealed plastic bags.
Sample Storage, DNA extraction
Filter papers and RDTs were stored at room temperature. DNA was extracted from two 3 mm Ø punches obtained from the filter paper or the whole RDT strips. DNA was extracted with Chelex®100 resin (Bio-Rad Laboratory, Hercules, CA) using the boiling method with minor modifications from the original protocol using 0.2% saponin / phosphate-buffered saline and 10% Chelex [22]. DNA was stored at −20°C until use
Molecular analyses
Amino acid positions pfdhfr 16-185, pfdhps 436-632 and K13 407-689 were amplified using previously described PCR protocols and then sequenced commercially [13, 15, 23-26]. The Sequencher™ software version 4.6 (Gene Codes Corporation, Ann Arbor, MI, USA) was used for sequence analysis. The P. falciparum 3D7 clone sequences obtained from NCBI database were used as references for pfdhfr, pfdhps and K13.
Previously described multiplex PCR-RFLP (restriction fragment length polymorphism) methods with minor modifications were used to identify pfdhfr N51I, C59R and S108N and pfdhps S436F/A, A437G and K540E when sequencing failed and pfcrt K76T and pfmdr1 N86Y SNPs [23, 27, 28].
PCR and restriction products were resolved on 2% agarose gels (Amresco, Solon, OH, USA). All gels were stained with a nucleic acid gel stain (GelRedTM, Biotium Inc, Hayward, CA, USA) and visualized under UV transillumination (GelDoc®, Biorad, Hercules, CA, USA). PCR products were purified and sequenced commercially (Macrogen Inc, Seoul, Korea).
Statistical analyses
The exact incidence of P. falciparum malaria was not known and we therefore decided to collect as many samples as possible over a two year period. SNP frequencies were calculated by dividing the number of SNPs by the number of patients in whom a certain the allele could be identified. Allele frequencies in 2015 and 2016 were compared using Chi-squared tests.
Ethics
Patients with uncomplicated malaria were enrolled and samples collected after informed oral consent of the patient or in the case of minors informed proxy-consent of their parent or guardian. The study was approved by the Institutional Ethics Committee of R D Gardi Medical College in Ujjain, Madhya Pradesh, India (61/2009 and 494/2016) and the Regional Ethics Committee in Stockholm, Sweden (2011/832-32/2).