Chloroquine resistance is associated to multi-copy pvcrt-o gene in Plasmodium vivax malaria in the Brazilian Amazon

The study was conducted from January 2012 to April 2013 at the FMT-HVD, an infectious disease referral centre located in Manaus, in the Western Brazilian Amazon.

The study included patients with P. vivax malaria of both genders, aged 6 months–60 years, weight greater than 5 kg, with blood parasite density from 250 to 100,000 parasites/mL and axillary temperature ≥ 37.5 °C or history of fever in the last 48 h. Use of anti-malarials in the previous 30 days, impossibility to be followed up for 42 days and clinical complications were considered exclusion criteria [17]. Patients received supervised treatment with 25 mg/kg of CQ phosphate over a 3-day period (10 mg/kg on days 0 and 7.5 mg/kg on days 1 and 2. PQ was prescribed at the end of follow-up or if malaria recurred during follow-up, at the dosage of 0.5 mg/kg per day, during 7 days.

Patients were evaluated on days 0, 1, 2, 3, 7, 14, 28 and 42 and, if they felt ill, at any time during the follow-up period. Day 0 (D0) is defined as the admission day and day of recurrence (DR) as the day of recrudescence for the same patient. Both samples from the same patient were analysed whenever available. For some patients, only a one-time sample (D0 or DR) was available. Thick blood smears and full blood counts were collected from all patients on each visit. On D0 and in DR, whole blood was also collected for DNA storage. CQ and desethylchloroquine (DCQ) plasmatic levels were determined only in case of parasitological failure. Three aliquots of 100 μL of plasma from DR samples were spotted onto filter paper for later analysis by high performance liquid chromatography to determine the levels of CQ and DCQ, as previously described [2, 18]. CQRPv was defined as a parasite recurrence presenting plasma concentrations of CQ plus DCQ higher than 100 ng/mL when added up. Susceptible-CQ P. vivax group consisted of patients with no parasite recurrence during the follow-up period.

Thick blood smear was performed by the Walker technique [19] and evaluated by an experienced microscopist. Parasite densities were calculated by counting the number of parasites per 500 leukocytes and the number of parasites/µL per patient was determined [20]. Real-time PCR was performed to confirm P. vivax monoinfection. The extraction of total DNA from whole blood was performed using the QIAamp DNA Blood Mini Kit (Qiagen, USA), according to the manufacturer’s protocol. The DNA was amplified in an Applied Biosystems 7500 Fast System using primers and TaqMan fluorescence labeled probes for real-time PCR [21].

Quantitative PCR (qPCR) was performed using Taqman^® Universal PCR Master Mix (Applied Biosystems) and 100–200 ng of template DNA were prepared from each sample. PCR products were amplified and identified on a 7500 FAST (Applied Biosystems). Sequences of primers used to determine copy number variation of pvcrt-o [22] and pvmdr-1 [23] genes were described in Additional file 1.

Cycling parameters for PCR were an initial denaturation step at 95 °C for 10 min, 40 cycles of 15 s at 95 °C and 1 min at 60 °C. The single-copy β-tubulin gene was used as a reference gene (normalizer), and a field sample with a single copy of the target gene was used as a calibrator. The ΔΔCt method was used to determine the number of copies of pvcrt-o and pvmdr-1 genes relative to a standard calibrator sample. Samples were considered to have a single copy of the gene when the estimated value was 0.5–1.5, while values above 1.5 were defined as gene amplification. Each DNA sample was assayed in triplicate.

To define the calibrators for copy number variation (CNV) analysis using real-time PCR assays, plasmids containing a single copy of each gene: pvcrt-o, pvmdr-1 and pvtubulin were constructed. The presence of a unique copy of each gene was confirmed by sequencing. Using the single-copy plasmids, a set of field samples to define the gene copy number using relative quantification in real-time PCR was evaluated. Thus, samples with a single copy of pvcrt-o and pvmdr-1 were selected to be used as calibrators in the CNV assays.

PCR primers and different reaction conditions used to amplify pvcrt-o and pvmdr-1 gene sequences are described in Additional file 2. Sequencing reaction was performed in triplicate. Nucleotide sequences were analysed using the NCBI BLAST algorithm and compared with reference sequences of GenBank [24]. Amino acid sequences were aligned and compared with sequences of the reference chloroquine transporter from P. vivax Sal1 (GeneBank ID: NC_009906) using Geneious^® software (Biomatters, v6.0.5).

For the genotyping procedures for comparing day 0 and recurrences samples, 3 highly polymorphic genes were chosen according to diversity and discriminatory potential in the region, msp1F3, MS2, and MS8 [25, 26]. PCRs were performed in 20 µL reactions with 10 µM of each primer, 2 µL of 10 × Buffer B (50 mM KCl, 10 mM Tris, pH 8.3), 2.5 mM each dNTP, 25 mM MgCl2, 1.5 U Taq DNA polymerase and 5 µL genomic DNA, as described in Additional file 3. All reagents were used from Invitrogen. 3 µL of diluted primary PCR product was used as template for the nested PCR. PCRs were performed with conditions as follows: initial denaturation at 95 °C for 1 min, followed by 29 cycles (primary PCR) or 24 cycles (nested PCR) of denaturation at 95 °C for 30 s, annealing at 59 °C for 45 s, elongation at 72 °C for 1 min, followed by a final elongation at 72 °C for 5 min. PCR products were sent to Macrogen for capillary electrophoresis-based sequencing and analysed with Gene Marker version 2.6.0. To distinguish reinfection, recrudescence, and relapse by P. vivax, was used the classification method recommended by WHO for P. falciparum [27]. For each recurrence, samples were classified as homologous if at least 1 allele for each locus investigated was detected in both paired samples (D0 and DR) on two different markers and as heterologous if all alleles for a given marker were different on two or more markers [11].

Data were analysed using SPSS^® version 21.0 for Windows (SPSS Inc.^® Chicago, IL, USA). Normal distribution of data was evaluated with the Kolmogorov–Smirnov test. Asexual parasitaemia at D0, D1, D2 and D3 was compared in patients with CQRPv and CQ-susceptible P. vivax using T-Student test. Parasitaemia clearance rates at D1, D2 and D3 were compared in patients with CQRPv and CQ-susceptible P. vivax using Chi square or Fisher’s test. Chi square or Fisher’s test was used to test differences in proportions of polymorphism of the promoter region and copy number variation. A p value of 0.05 or lower was considered to be statistically significant. The genetic variation for each microsatellite locus was measured by calculating the expected heterozygosity (H_E). H_E was calculated using D0 and DR for each locus as = [n/(n − 1)] [1 − ∑p _i ² ] where n is the number of isolates sampled and p_i is the frequency of allele i.

For pvcrt-o CNV analysis, with 28 patients with CQ-susceptible P. vivax and 25 patients with CQRPv being analysed (14 samples from D0 to 18 from DR). At D0, 7.1% of the CQ-susceptible P. vivax presented multiple copies of pvcrt-o, 7.7% of the CQRPv showed multiple copies of pvcrt-o at D0 while at DR 33.0% had multiple copies of pvcrt-o (Fig. 1a). There was a significant difference between CQ-susceptible P. vivax and CQRPv at DR (P = 0.02).

For pvmdr-1 CNV analysis, a total of 28 CQ-susceptible and 25 CQRPv (9 from D0 to 17 from DR) P. vivax samples were analysed. It was verified that 10.7% of the CQ-susceptible P. vivax samples, 11.1% of the CQRPv samples at D0 and 0.0% of the CQRPv samples at DR presented multiple copies (Fig. 1b). There was no significant difference between CQ-susceptible and CQRPv groups regarding pvmdr-1 CNV (P > 0.05). Samples that did not meet the criteria used in the methodology were excluded from the analysis of these genes. Overall, pvmdr1 and pvcrt-o CNV was not related to parasitaemia and age for individuals infected with P. vivax with either with multiple copies of pvmdr-1 or with one copy of the genes (P > 0.05).

Nucleotide sequencing of a region flanking the promoter and exon I of the gene pvcrt-o showed a deletion of 19 bp in the promoter region at position 32,083–32,102 and an insertion of three bases in exon 1 leading to 5AAG repeats (AAG) after codon nine, leading to the K10 insertion compared to the reference sequence NC_009906-Chloroquine resistance transporter-P. vivax Sal1, with 4AAG repeats termed wild type here (Fig. 2).

A deletion of 19 bp was found in 11/23 (47.6%) of the patients with CQ-susceptible P. vivax and 3/10 (23.1%) of the samples with in CQRPv at D0. At day DR, 55.5% of the samples with CQRPv had the 19 bp deletion. There was no significant difference between groups (P ≥ 0.05, Fisher’s exact test) (Table 3).

All the samples with 19 bp deletion presented 4AAG repeats in the exon I (Table 2). Of the samples with wild type sequence, 8 (33.3%), 8 (60.0%) and 3 (25.0%) exhibited 5AAG (K10) repeats in the exon I of the gene in the samples with CQ-susceptible P. vivax, CQRPv at D0 and CQRPv at DR, respectively (Table 3).

For the pvmdr-1 gene, nucleotide sequencing was performed in 66 samples. Sixteen samples with CQ susceptible P. vivax and 23 CQRPv were sequenced, 14 in the D0 and 9 in the DR. There was no variation in the analysed gene compared to the P. vivax reference Sal-1 available in the NCBI gene bank.