Four years’ monitoring of in vitro sensitivity and candidate molecular markers of resistance of Plasmodium falciparum to artesunate-mefloquine combination in the Thai-Myanmar border

Background The decline in efficacy of artesunate (AS) and mefloquine (MQ) is now the major concern in areas along the Thai-Cambodian and Thai-Myanmar borders. Methods The correlation between polymorphisms of pfatp6, pfcrt, pfmdr1 and pfmrp1 genes and in vitro sensitivity of Plasmodium falciparum isolates to the artemisinin-based combination therapy (ACT) components AS and MQ, including the previously used first-line anti-malarial drugs chloroquine (CQ) and quinine (QN) were investigated in a total of 119 P. falciparum isolates collected from patients with uncomplicated P. falciparum infection during 2006–2009. Results Reduced in vitro parasite sensitivity to AS [median (95% CI) IC50 3.4 (3.1-3.7) nM] was found in 42% of the isolates, whereas resistance to MQ [median (95% CI) IC50 54.1 (46.8-61.4) nM] accounted for 58% of the isolates. Amplification of pfmdr1 gene was strongly associated with a decline in susceptibility of P. falciparum isolates to AS, MQ and QN. Significant difference in IC50 values of AS, MQ and QN was observed among isolates carrying one, two, three, and ≥ four gene copies [median (95% CI) AS IC50: 1.6 (1.3-1.9), 1.8 (1.1-2.5), 2.9 (2.1-3.7) and 3.1 (2.5-3.7) nM, respectively; MQ IC50: 19.2 (15.8-22.6), 37.8 (10.7-64.8), 55.3 (47.7-62.9) and 63.6 (49.2-78.0) nM, respectively; and QN IC50: 183.0 (139.9-226.4), 256.4 (83.7-249.1), 329.5 (206.6-425.5) and 420.0 (475.2-475.6) nM, respectively]. The prevalence of isolates which were resistant to QN was reduced from 21.4% during the period 2006–2007 to 6.3% during the period 2008–2009. Pfmdr1 86Y was found to be associated with increased susceptibility of the parasite to MQ and QN. Pfmdr1 1034C was associated with decreased susceptibility to QN. Pfmrp1 191Y and 1390I were associated with increased susceptibility to CQ and QN, respectively. Conclusion High prevalence of CQ and MQ-resistant P. falciparum isolates was observed during the four-year observation period (2006–2009). AS sensitivity was declined, while QN sensitivity was improved. Pfmdr1 and pfmrp1 appear to be the key genes that modulate multidrug resistance in P. falciparum.


Background
Southeast Asia, particularly the Thai-Cambodian border, is one of the malaria-endemic region where multidrugresistant Plasmodium falciparum malaria has been reported [1]. In the 1960s and 1970s, chloroquine (CQ) resistance spread throughout the region and subsequently, in the 1980s, resistance to sulphadoxine and pyrimethamine was reported [2]. In 1984, mefloquine (MQ) was firstly introduced for clinical use as first-line treatment for uncomplicated multidrug-resistant P. falciparum malaria in Thailand, but MQ resistance was rapidly developed four years after its implementation. Following the decline in clinical efficacy of MQ, the artemisininbased combination therapy (ACT) using the artesunatemefloquine combination was introduced as first-line reatment in 1994 [3]. Cure rate was improved to over 90% and the incidence of P. falciparum malaria was markedly reduced [4,5]. Artemisinin resistance however, initially occurred during 2006-2007 in areas along the Thai-Cambodian border [6]. With regard to the Thai-Myanmar border, treatment failure following ACT has been increasing especially in Tak province [7]. Studies during 2008-2009 showed a marked decline in the 42-day cure rate from 99.2 to 72.58% [8,9]. It is unclear whether artemisinin resistance has spread from the eastern to the western border of the country. Monitoring and identifying factors contributing to this low cure rate is necessary for the country's future perspective of malaria control policy. Applying genetic analysis as a tool for detecting the genetic change of malaria parasite genes that have been shown to link with the decline in efficacy of artesunate (AS) and MQ; i.e., pfmdr1 [10], pfatp6 [11] and pfmrp1 [12][13][14], in association with the in vitro sensitivity of the parasite to both combination partners, would help to detect early changes in P. falciparum sensitivity to this combination therapy.
In the present study, the association between the polymorphisms of pfatp6, pfcrt, pfmdr1, and pfmrp1 genes and in vitro sensitivity of P. falciparum isolates to AS, MQ, as well as CQ and quinine (QN) was investigated in P. falciparum isolates collected from the Thai-Myanmar border. In addition, the change in parasite genetic patterns and in vitro sensitivity over the period 2006-2009 was also examined.

Blood samples
A total of 130 P. falciparum isolates were collected from patients with uncomplicated P. falciparum infection prior to treatment with a three-day combination regimen (25 mg/kg body weight MQ and 12 mg/kg body weight AS) for investigation of the polymorphisms of candidate molecular markers of anti-malarial drug resistance and in vitro parasite sensitivity. The study was conducted at Mae Tao clinic, Mae Sot District, Tak Province, Thailand during 2006-2009. Fifty-seven and 73 isolates were collected from patients during 2006-2007 and 2008-2009, respectively ( Figure 1). The study protocol was approved by the Ethics Committee of the Ministry of Public Health of Thailand and written informed consents for study participations were obtained from all patients before study.
Investigation of polymorphisms of pfcrt, pfmdr1, pfatp6, and pfmrp1 using PCR-RFLP Genomic DNA was extracted from all samples (cultureadapted P. falciparum; Figure 1) using chelex resin modified technique [19]. Prior to being used as a DNA template, concentration of the malaria genomic DNA was determined by spectrophotometry (Nanodrop TM , Thermo fisher Scientific, Massachusetts, USA).
Detection of pfatp6 and pfmdr1 gene copy number by SYBR Green I real-time PCR Pfatp6 and pfmdr1 gene copy number in all samples ( Figure 1) was determined by SYBR Green I real-time PCR (iCycler™, Bio-Rad, California, USA) using the default thermocycler program: 10min of pre-incubation at 95°C, followed by 40 cycles for 15 sec at 95°C and 1 min at 60°C. The oligonucleotide primers used were those previously designed by Ferreira et al. [26] with modification. Individual real-time PCR reaction was carried out in a 25 μl reaction volume in a 96-well plate containing 2 μl of DNA (50 ng), 1 μM each of sense and antisense primer and 12.5 μl of Platinum™ PCR SuperMix (Invitrogen, California, USA).
The 2 -ΔΔCt method of relative quantification was adapted to estimate copy number in P. falciparum genes. The genomic DNA extracted from P. falciparum 3D7 clone known to harbour a single copy of each gene was used as a calibrator, while Pf-β-actin 1 served as the house-keeping gene in all experiments. Dd2 genomic DNA carrying four copies of pfmdr1 was used as a second calibrator. The threshold cycle (Ct) was determined as the increase in reporter signal, which was first detected above baseline. Results were analysed by a comparative Ct method based on the assumption that the target (pfatp6 and pfmdr1) and reference (pf-β-actin 1) genes were amplified with the same efficiency within an appropriate range of DNA concentrations.
The ΔΔCt calculation for the relative quantification of target was as follow: ΔΔCt = (Ct, target gene − Ct, Pf-β-actin1) x − (Ct, target gene − Ct, Pf-β-actin1) y , where x represents unknown sample and y represents P. falciparum 3D7 clone. Results for each sample was expressed as an N-fold change in χ target gene copies, normalized to Pf-β-actin-1 relative to the copy number of the target gene in P. falciparum 3D7 clone, according to the following equation: amount of target = 2 -ΔΔCt . A minimum of two experiments were carried out for each gene and each sample. In each experiment, each individual sample was analysed in duplicate wells and the Ct of each well was recorded at the end of the reaction.

Statistical analysis
The association between in vitro sensitivity of P. falciparum isolates and polymorphisms of pfcrt, pfmdr1, pfmrp1, and pfatp6 was analysed using Chi-square and Mann-Whitney U tests. Correlation between the two quantitative variables was evaluated using Spearman correlation test. The qualitative variables are summarized as proportions and percentages and the quantitative variables

In vitro sensitivity of Plasmodium falciparum isolates
In vitro sensitivity to AS, MQ, CQ, and QN was successfully evaluated in a total of 119 P. falciparum isolates ( Figure 1 and Table 1

Pfmdr1 mutation
No mutation in pfmdr1 at the codons 1042D and 1246Y was observed (0/130 and 0/130) in any isolate collected during the two investigation periods. The prevalence of pfmdr1 wild type at the target amino acid residue 86 was significantly higher in isolates collected

Pfmdr1 copy number
Isolates which carried one, two, three, four, five, six and eight gene copies were found in 48.8% (60/123), 15.4% two, three, four, five, six and eight gene copies, respectively. There was no significant difference in the prevalence of pfmdr1 gene copy in isolates collected during the two periods.

Pfatp6 copy number
All of the 67 isolates under the analysis from the two periods carried only one gene copy.

Association between polymorphisms of candidate molecular markers of anti-malarial drug resistance and in vitro sensitivity of Plasmodium falciparum isolates
The association between polymorphisms of candidate molecular markers of anti-malarial drug resistance and in vitro sensitivity to anti-malarial drugs was investigated in 119 matched-paired P. falciparum isolates (Figure 3).

Pfatp6 mutation and pfatp6 copy number
No significant association was observed between pfatp6 mutation including pfatp6 copy number and in vitro sensitivity of P. falciparum isolates to all drugs.

Discussion
The present study investigated the relationship between P. falciparum gene polymorphisms (pfatp6, pfcrt, pfmdr1, and pfmrp1) and in vitro sensitivity of P. falciparum isolates to the ACT combination partners AS and MQ, including CQ and QN in a multidrug-resistant area along the Thai-Myanmar border during the two periods (2006-2007 and 2008-2009). Results suggest a significant change in the prevalence and pattern of pfatp6, pfmdr1 and pfmrp1 gene polymorphisms with no change in the in vitro sensitivity profiles of the parasite during the two periods. Decline in sensitivity of P. falciparum to MQ was continuously reported during 1991-1994 when MQ monotherapy was employed in Thailand [27]. However, until a decade after the switch of first-line treatment to AS-MQ combination, high prevalence and intensity of MQ resistance was reported (32%, 46% and 58% during 1998-2005 [28], 2007-2008 [29], and 2006-2009 (the current study), respectively). While evidence of artemisinin resistance was documented in Western Cambodia during the period 2006-2007 [30,31], sensitivity of P. falciparum isolates in this area to AS during the same period was generally considered satisfactory [18,29,32,33]. Subsequent reports on a decline in AS sensitivity [7,9] nevertheless, has created a major concern on the future use of AS in combination therapy. Without definite criteria for defining artemisinin resistance, the upper limit of 95% CI of median IC 50 value of AS in isolates collected from all patients (2.3 nM) was used as a cut-off level of AS resistance in the present study. Based on this cut-off level, approximately 42% of the isolates were classified as declined sensitivity to AS. The prevalence of isolates with MQ resistance and declined sensitivity to AS observed in the present study were similar to that previously reported (58 vs 57.6% and 42 vs 36.7%, respectively) [34]. This cut-off criteria for defining AS resistance (the upper limit of 95% CI of median IC 50 value of AS) was slightly lower in the current study (2.3 nM) compared with the previous study (2.8 nM) [34]. The sensitivity to AS (IC 50 : 2.0 vs 1.7 nM) and MQ (30.1 vs 34.0) were comparable with that reported in the isolates collected from Cambodia during the same period [35]. It was noted for the improvement of parasite sensitivity in this area to CQ and QN since the introduction of AS-MQ combination in Thailand in 1995. There was even one isolate that was sensitive to CQ. In 1991-1992 and 1994, the degree of CQ resistance in the country was high [median (95% CI) IC 50 193.2 (148.24-251.85) and 157.0 (124. 11-198.73 [18,36,37]. Molecular analysis in either laboratory or field P. falciparum isolates demonstrated the strong linkage between CQ resistance and pfcrt gene mutations [20,22,38]. Susceptibility of the parasite to CQ was also shown in this study to link with pfmrp1 polymorphism, of which the 191Y mutation resulted in increased susceptibility of the parasite to CQ. For QN, a decline in parasite sensitivity to the drug was obviously observed during 1991-1992 and 1994  Pfmdr1 and pfmrp1 appear to be the key genes involved in resistance of P. falciparum to the commonly used anti-malarial drugs. The pfmdr1 86Y mutation leads to increase in susceptibility of the parasite to MQ compared with wild type genotype (IC 50 of 8.1 and 30.4 nM, respectively) as well as the structurally related antimalarial QN (IC 50 of 86.4 and 236.3 nM). The results showed low prevalence of 184F allele in parasite isolates collected during the investigation period. The 184F allele was reported to be associated with increased IC 50 of MQ. High prevalence of (~86%) of the 184F allele was reported in western Cambodia where the level of MQ resistance was significant. On the other hand, the prevalence of 184F allele in eastern Cambodia was low (~32%), which was also correlated with the reduced level of MQ resistance in this region [40]. Molecular analysis in the current study revealed an obvious involvement of pfmdr1 and pfmrp1 genes with QN-resistant P. falciparum [13,14,21,41] With regard to the pfatp6 polymorphisms, there were no mutations found at codons 37, 639 and 769 in samples collected during 2006-2009. The mutation at codon 898 was found in isolates collected from 2008-2009. Although this residue is silent mutation, this observation may imply on the influence of anti-malarial drug pressure on the parasite during that period. The finding was in agreement with that reported by Jambou and colleges for the isolates in Asia including Thailand [11]. It is noted however that in vitro cultivation may cause the poor fitness of the mutant genotypes, which may explain the observation of almost absence of the mutation in pfatp6 gene in this study [42].
The current results support the role of pfmdr1 amplification in modulating the degree of AS, MQ and QN susceptibility [43,44] A trend of increasing number of gene copies and increasing IC 50 values of AS, MQ and QN was clearly observed. Approximately 51% of the isolates carried pfmdr1 copy number ranging from two to eight copies. The study conducted during 1995-2007, the period of which the clinical efficacy of AS-MQ combination was satisfactory, showed the increase in pfmdr1copy number from 30% (12/40) in 1996 to 53% (24/45) in 2006 [45]. Besides the high level of initial parasitaemia, prolongation of parasite clearance time and reduced in vitro parasite sensitivity to MQ, treatment failure following AS-MQ combination therapy was also associated with increased pfmdr1 copy number. Parasite isolates collected prior to treatment from patients with recrudescent response were found to carry higher number of pfmdr1 copies compared to those with sensitive response (mean copies of 2.7 vs 1.9). In addition, isolates collected on the day of recrudescence carried higher number of gene copies than the corresponding day 0 samples (mean copies 3.5 vs 2.6) [35]. In a previous study, about 70% of isolates collected from patients with recrudescence response before AS-MQ treatment was shown to carry more than 1 pfmdr1 copy number and increase in pfmdr1 copy number was associated with reduced parasite sensitivity to AS, or resistant to MQ, or both [34]. Although the artemisinin resistance has not been clearly defined in this study, results suggest that both the parasite (reduced in vitro sensitivity and increased pfmdr1 copy number) and host (pharmacokinetic variability) factors might contribute to artemisinin resistance [9,46].

Conclusions
Based on results of the current observation on in vitro sensitivity and candidate molecular markers of resistance, it is concluded that high prevalence of MQ resistance still remained during the four years' observation period (2006)(2007)(2008)(2009)). In addition, sensitivity of the parasite to AS appeared to be declining. Pfmdr1 gene copy number is the key molecular marker of resistance of P. falciparum isolates in this area to AS-MQ combination therapy.