Expression Levels of pvcrt-o and pvmdr-1 Are Associated with Chloroquine Resistance and Severe Plasmodium vivax Malaria in Patients of the Brazilian Amazon

Molecular markers associated with the increase of chloroquine resistance and disease severity in Plasmodium vivax are needed. The objective of this study was to evaluate the expression levels of pvcrt-o and pvmdr-1 genes in a group of patients presenting CQRPv and patients who developed severe complications triggered exclusively by P. vivax infection. Two different sets of patients were included to this comprehensive study performed in the Brazilian Amazon: 1) patients with clinically characterized chloroquine-resistant P. vivax compared with patients with susceptible parasites from in vivo studies and 2) patients with severe vivax malaria compared with patients without severity. Quantitative real-time PCR was performed to compare the transcript levels of two main transporters genes, P. vivax chloroquine resistance transporter (pvcrt-o) and the P. vivax multidrug resistance transporter (pvmdr-1). Twelve chloroquine resistant cases and other 15 isolates from susceptible cases were included in the first set of patients. For the second set, seven patients with P. vivax-attributed severe and 10 mild manifestations were included. Parasites from patients with chloroquine resistance presented up to 6.1 (95% CI: 3.8–14.3) and 2.4 (95% CI: 0.53–9.1) fold increase in pvcrt-o and pvmdr-1 expression levels, respectively, compared to the susceptible group. Parasites from the severe vivax group had a 2.9 (95% CI: 1.1–8.3) and 4.9 (95% CI: 2.3–18.8) fold increase in pvcrt-o and pvmdr-1 expression levels as compared to the control group with mild disease. These findings suggest that chloroquine resistance and clinical severity in P. vivax infections are strongly associated with increased expression levels of the pvcrt-o and pvmdr-1 genes likely involved in chloroquine resistance.


Introduction
Mechanisms underlying severe malaria triggered by Plasmodium vivax have been poorly appreciated, mainly the likely higher and more-severe disease burden imposed by increasingly chloroquine-resistant P. vivax parasites (CQRPv) [1,2]. Most common manifestations of severe vivax malaria include severe anemia and respiratory distress, being particularly associated to young age [2][3][4]. Of interest, the majority of reports on severe vivax malaria come from regions where drug-resistant P. vivax parasites significantly thread the radical cure and control of this infection [5]. The first reports of CQRPv were made in 1989 [6,7] from Papua New Guinea and Indonesia. In Brazil, the first case of properly ascertained in vivo CQRPv was from a patient treated in Manaus, in the Brazilian Amazon [8]. In this city, a subsequent trial assessed the efficacy of standard supervised CQ monotherapy and the proportion of failures was 10.1% [9]. In parallel to the emergence of CQRPv in the Brazilian Amazon, reports of clinical severity exclusively associated with P. vivax infection increased substantially [3,[10][11][12].
Based on observational studies, an association between CQRPv and disease severity in P. vivax was proposed in the Indonesian Papua [2], although this has not been properly and systematically evaluated. CQR in P. falciparum is associated with mutations in the pfcrt-o and pfmdr-1 genes [13,14] but similar studies with the orthologous genes of P. vivax, pvcrt-o and pvmdr-1, failed to demonstrate a similar association [15][16][17]. Interestingly, a pioneer case of severe disease exclusively associated with P. vivax, demonstrated a significant increase on expression levels, not mutations, of these genes [18]. Altogether, these studies suggest the possibility of a significant association of CQRPv and disease severity and that expression levels of pvcrt-o and pvmdr-1 should be better explored as molecular markers of both phenomena to further enhance the understanding of the clinico-epidemiological. The aim of this study was to evaluate the expression levels of pvcrto and pvmdr-1 genes in a well-characterized group of patients presenting CQRPv and in a group of patients who developed severe complications triggered by P. vivax infection.

Ethics statement
The study was approved by the Ethics Review Board of the Fundação de Medicina Tropical Heitor Vieira Dourado (FMT-HVD) (approval number 343/2009). Participants were instructed about the objectives of the study and signed an informed consent. In the case of under 18 years, the consent form was signed by the guardians. Patients diagnosed with malaria were treated according to the guidelines of the Brazilian Ministry of Health [19].

Study area
The study was carried out from June 2011 to December 2012 in the FMT-HVD, an infectious disease referral center located in Manaus, Western Brazilian Amazon.

Selection of patients for the CQRPv study
The study included patients with P. vivax malaria of both sexes, aged 6 months-60 years, bodyweight greater than 5 kg, presenting blood parasite density from 250 to 100,000 parasites/ml and axillary temperature $37.5uC or history of fever in the last 48 hours. Exclusion criteria were: use of antimalarials in the previous 30 days, refusal to be followed up for 42 days and any clinical complication [19]. Patients received supervised treatment with 25 mg/kg of chloroquine (CQ) phosphate over a 3-day period (10 mg/kg on days 0 and 7.5 mg/kg on days 1 and 2). Primaquine was prescribed at the end of follow-up or if disease recurred for 7day period in the dosage of 0.5 mg/kg per day. Patients who vomited the first dose within 30 minutes after drug ingestion were re-treated with the same dose. 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. In all sample, blood smear, full blood counts and PCR were performed. On admission (D0) and in the day of recrudescence (DR), 20 ml of intravenous whole blood was also collected for DNA and RNA storage. CQ and desethylchloroquine (DCQ) plasmatic levels were determined only in case of parasitological failure [20]. Three aliquots of 100 ml of whole blood from DR samples were spotted onto filter paper for later analysis by high performance liquid chromatography (HPLC) to estimate the levels of CQ and DCQ as previously described [21,22]. CQRPv was defined if: 1) recurrences of the parasite with plasma concentrations of CQ higher than 100 ng/dl, or 2) recurrences with smaller concentrations but presenting DR and D0 with the same alleles. The control group consisted of patients with no parasitemia recurrence during the same follow-up period.
Selection of patients with severe P. vivax malaria and patients with mild symptoms During the same period, admitted patients classified as severe if presenting any of the WHO severity criteria [23], and patients with mild manifestations from the outpatient clinics. On admission, 20 ml of intravenous whole blood was collected for complete blood count, blood biochemistry and for DNA and RNA extraction. Other tests were requested at physician's discretion. To exclude patients with infection other than Plasmodium, all patients also had a blood culture performed for aerobic bacteria and serological tests for leptospirosis (IgM), HIV-1/HIV-2, hepatitis A (anti-HAV IgM), hepatitis B (HBsAg), hepatitis C (anti-HCV), hepatitis D (total anti-HDV) and dengue (RT-PCR). Patients presenting any sign of severity were treated with intravenous artesunate, as recommended by the WHO guidelines [19].

P. vivax malaria diagnosis
Thick blood smear was prepared as recommended by the Walker technique [24] and evaluated by an experienced microscopist [25]. Parasite densities (parasites/ml) were calculated by counting the number of parasites per 500 leukocytes in high magnification fields, and the number of leukocytes/ml per patient. In addition, differential counting of asexual forms (ring-stage parasites, mature trophozoites and schizonts) was made to ensure that there was no difference between groups of cases and controls.
Afterwards, real-time PCR was performed to confirm P. vivax mono-infection. 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 [26].

RNA extraction and expression of pvcrt-o and pvmdr-1 genes
For RNA extraction, 5 ml of whole blood were collected and processed using saponin. Total RNA was purified using Trizol reagent (Invitrogen) according to the manufacturer's instructions. Furthermore, 2% agarose gel electrophoresis was made to guarantee RNA quality. Amplification reactions were performed using Power SYBR Green PCR Master Mix (Applied Biosystems) and 45 ng of template cDNA prepared from each sample. PCR products were amplified and detected on a 7500 FAST (Applied Biosystems). The primers were pvmdr-1F (AAGGATCAAAGGCAACCCA), pvmdr-1R (TCAGGTTGTTACTGCTGTTGCTATT), pvcrt-oF (ATGTCCAAGATGTGCGACGAT), pvcrt-oR (CTGGTCCCTGTATGCAACTGAC), pvtubulinF (CCAAGAATATGATGTGTGCAAGTG), pvtubulinR (GGCGCAGGCGGTTAGG) [18]. Cycling parameters for PCR were an initial denaturation step at 95uC for   Reproducibility of the method (accuracy inter-runs) was assessed through measures of Cts obtained by cDNA quantification of one isolate analyzed 10 times in two runs that were performed on different days, using the same equipment. The results were expressed as mean, standard deviation (SD) and coefficient of variation (CV%). The repeatability of the method (accuracy intra-run) was evaluated by measures of Cts obtained by cDNA quantification of one isolate measured 10 times in the same run. The results were expressed by mean, standard deviation (SD) and coefficient of variation (CV%).

Sequencing analysis of pvcrt-o and pvmdr-1 genes
All samples from patients presenting CQRPv or patients with severe symptoms and the respective control groups, obtained prior to treatment (on D0), were sequenced for searching mutations suspected to be associated to CQR. Extraction of whole DNA was carried out using a QIAamp DNA Mini Kit (QIAGEN, Germany) according to the manufacturer's protocol. PCR primers and different reaction conditions used to amplify pvcrt-o and pvmdr-1 gene sequences were made as previously described (Table S1). The amplification products were quantified by NanoDrop 2000 (Thermo Scientific) and sent for sequencing to Macrogen (South Korea). Amino acid sequences were compared with sequences of Sal I (Salvador I): GenBank accession nos. ADE74979.1 for pvcrto [17] and EU333979.1 for pvmdr-1 [28]. The deduced amino acid sequences were aligned and analyzed using Mutation Surveyor. The pvmdr-1 and pvcrt-o gene sequences were deposited in GenBank under the accession numbers KM016489 to KM016517 and KM016518 to KM016532, respectively.

Polymorphism analysis of msp1F3 and MS2 molecular markers for patients with CQRPv
PCRs were performed in 20 ml reactions containing 10 mM of each primer, 2 ml of 106buffer B, 2.5 mM each dNTP, 25 mM MgCl2, 1.5 U Taq DNA polymerase and 5 ml genomic DNA (Table S2). 3 ml of diluted primary PCR product was used as template for the nested PCR (MS2 and msp1F3). These polymorphism markers were chosen based on preliminary data from other patients from Manaus (unpublished data). PCRs were performed in a thermocycler with conditions as follows: initial denaturation at 95uC for 1 min, followed by 29 cycles (primary PCR) or 24 cycles (nested PCR) of denaturation at 95uC for 30 seconds, annealing at 59uC for 45 sec, elongation at 72uC for 1 min, followed by a final elongation at 72uC for 5 min. PCR products were stored in the dark (plates wrapped in foil) at 4uC until be sent to Macrogen for capillary electrophoresis-based sequencing and analyzed with GeneMarker version 2.6.0.

Statistical Analyses
Data were analyzed using SPSS version 16

Reproducibility, repeatability of measurements of Cts and specificity
For reproducibility, amplification of pvcrt-o and pvmdr-1 genes showed average Ct values ranging from 35.14 to 37.93 (mean Ct value 36.82) and from 34.99 to 36.19 (mean Ct value 35.44), respectively. Among the twenty measured values, doing in to different runs, the maximum coefficient of variation was 2.34% and 1.12%, for pvcrt-o and pvmdr-1 respectively. For the repeatability, the coefficient of variation was 1.35% for the pvcrt-o and 0.75% for pvmdr-1 in the first run, and 2.77% for the pvcrt-o and 1% for pvmdr-1 in the second run (Table S3).
After the relative quantification, only one peak was observed in the dissociation curves, confirming a good specificity of the primers in real time PCR.

Gene expression levels from CQR parasites
In this period, in a parallel study to evaluate efficacy of CQ with the 42-day follow-up, with supervised drug administration, 135 patients were enrolled, and 16 were CQR (11.8%). Of these, 12 with adequate RNA storage were included in the analysis. Clinical and laboratorial characteristics of the 12 patients with CQRPv included in this study are presented in Table 1. Out of this total, 83.3% patients were male, with a mean age of 29.0 years. The mean hemoglobin was 13.1 g/dl at D0 and 12.7 g/dl in DR. The geometric mean parasitemia was 1781.7 parasites/ml at D0 and 4206.5 parasites/ml at DR. It was found that mean blood levels of CQ plus DCQ were greater than 100 ng/ml on the DR in 10 of Table 2. Frequency of polymorphisms in the pvmdr1 and pvcrt-o genes for the different set of patients. 12 (83.3%) patients with recurring infection. Five patients with CQ-susceptible parasites and mild malaria were selected to generate the relative expression of 12 CQRPv and 15 CQsusceptible isolates.
In patients R1 to R7 gene expression was analysed at D0 and DR, while in patients R8 to R10 and R11 to R12, due to insufficient RNA amount, gene expression was measured only at D0 or DR, respectively. Patients with CQ-resistant P. vivax parasites presented a higher gene expression of pvcrt-o and pvmdr-1 at D0 and DR when compared to the susceptible group ( Figure 1A and 1B). For the CQR patients, median gene expression values at D0 and DR, presented 2.4 fold (95% CI: 0.96-7.1) and 6.1 fold (95% CI: 3.8-14.3) increase in pvcrt-o levels ( Figure 1A and 2A) compared to the susceptible patients at D0 with 0.12 fold (95% CI: 0.034-0.324) ( Figure S1A and 2A). Median gene expression for pvmdr-1 presented 2.0 fold (95% CI: 0.95-3.8) and 2.4 fold (95% CI: 0.53-9.1) increase levels at D0 and DR, for the CQR patients ( Figure 1B and 2B) versus 0.288 fold (95% CI: 0.068-0.497) for the susceptible patients at D0 (Figure S1B and 2B). Patients R2, R4, R6 and R7 showed a higher expression level of both genes in relation to the control group at D0 and DR (Figure 1). It was not observed significant differences in expression levels between D0 and DR for resistant parasites (p = 0.375 for pvcrt-o and p = 0.844 for pvmdr-1). However, when we compared expression levels between resistant and sensible parasites at D0 for both genes, values were significantly greater for treatment failures (p,0.05 for both genes). No correlation was found between CQ concentration and gene expression of DR for pvcrt-o (p = 0.286) and pvmdr-1 (p = 0.433) genes and between gene expression of pvcrt-o and pvmdr-1 and concentration of RNA in D0 and DR (p.0.05).
To discard that difference in gene expression levels was not due to difference in the distribution of asexual blood stages, parasites distribution was counted in both, CQRPv and CQ-susceptible samples (Tables S4 and S5). Of notice, no significant differences were observed between immature and mature asexual forms in both groups (p.0.05). Moreover, it was found that the pvcrt-o and pvmdr-1 gene mutations studied were not related to CQ-resistance (p.0.05) ( Table 2).
In order to characterize the genetic variation for each individual population, two neutral microsatellite loci (msp1F3 and MS2) were analyzed in D0 and DR samples. The analysis of microsatellites showed at least one concordant allele (83.3%) in primary infection and recrudescence (Table 1). Patients which presented CQ concentration less than 100 ng/ml, showed the same microsatellite allele in D0 and DR. The expected heterozygosity (H E ) was different in parasites from primary and recrudescence episodes (for msp1F3, primary: H E = 0.915 and recrudescence: H E = 0.681; and for MS2, primary: H E = 0.531 and recrudescence H E = 0.771).

Gene expression levels in severe vivax malaria parasites
In the period of study, 70 patients were hospitalized with P. vivax. However, due to many comorbidities, and absence of any criterion of severity, only 16 could be characterized as severe vivax malaria. Of these, 7 patients presenting mono-infection by P. vivax with any criterion of severity were included. Of these patients, 4 had severe anemia, 1 had severe anemia paralleled with acute respiratory distress syndrome (ARDS) and acute renal failure, 1 had acute renal failure and prostration and 1 had ARDS alone. Demographic and clinical characteristics of the patients enrolled are presented in Table 3. Of these patients, 51.0% were male, with a mean age of 38.6 years. The mean disease duration was 6.3 days, median hemoglobin was 6.3 g/dl and mean parasitemia was 11,180 parasites/ml. Table 3. Description of 7 patients presenting severe vivax malaria admitted to a tertiary health center, Manaus, Amazonas, Brazil. Ten patients diagnosed with non-severe malaria were selected and included as controls. The same 5 CQ-susceptible isolates from patients with mild malaria above cited were used to generate the relative expression of seven severe malaria patients and 10 nonsevere malaria patients. The group of patients with severe vivax malaria had a 2.9 (95% CI: 1.1-8.3) and 4.9 (95% CI: 2.3-18.8) fold increase in pvcrt-o and pvmdr-1 expression level versus 0.7 (95% CI: 0.3-1.1) and 1.5 (95% CI: 0.7-2.4) for the control group with mild disease (p = 0.007 and p = 0.01) (Figure 2, Figure 3 and Figure S2). To discard that difference in gene expression levels was not due to difference in the distribution of asexual blood stages, parasites distribution was counted in both, severe vivax malaria (Table S6) and non-severe. No significant difference was found between immature and mature asexual forms among patients with severe and non-severe malaria (p.0.05) ( Table S6). The pvcrt-o and pvmdr1 gene mutations studied were not related to severity (p.0.05) ( Table 2).
In a linear regression analysis, it was found a negative correlation between hemoglobin values (of severe and non-severe patients) and the log of the expression of the pvcrt-o (p,0.001) and pvmdr-1 (p = 0.007) genes among hospitalized patients (Figure 4).

Discussion
There have been recent reports on CQR from different regions of the world [8,9,15,[28][29][30][31], including Brazil [3,10,11]. Although the relevance of this problem, the molecular mechanisms of CQRPv are poorly understood. In this work, expression levels of pvcrt-o and pvmdr-1 genes were determined in patients with CQRPv from a well-characterized cohort under supervised treatment, pointing to a link between in vivo CQR and overexpression of both genes. Furthermore, these patients showed higher concentrations of CQ and microsatellite revealed the presence of the same clonal nature at D0 and DR, bringing evidence that the re-emergent infection is the same as the primary infection, therefore indicating recrudescence.
It is likely that CQR contributes to the burden of severe vivax malaria in regions where resistance is emerging [4,32], but there is paucity of available data regarding this issue. In this work, the group of patients with severe vivax malaria presented an increase in pvcrt-o and pvmdr-1 expression levels, as compared to the control group with mild disease. These two genes have been suggested as genetic markers of CQR in P. vivax [33,34].
Our results are in concordance with those reported by Fernandez-Becerra et al. [18], where high levels of pvcrt-o and pvmdr-1 gene expression were observed in a patient with severe vivax malaria compared to three non-severe. A prospective study conducted in the Papua New Guinea highlighted P. vivax as a major cause of severe malaria, particularly in settings with established or emerging chloroquine resistance, supporting indirectly our findings [2]. One speculates that in areas in which CQR P. vivax circulates, acute febrile illness potentially becomes chronic, recurrent and severe, due to recurrent parasitemia [5]. Moreover, chronic infections could lead to diserythropoiesis, as is seen in P. falciparum-induced anemia [35].
In this investigation, CQR and severe P. vivax malaria were observed in parallel with a high expression of pvcrt-o and/or pvmdr-1 genes in patients seen at a reference center in the Brazilian Amazon. Even though it is tempting to conclude about a causal association, a clear limitation of this study is that CQR has not been evaluated in the same patients with severe vivax malaria, especially because severe patients were in use of intravenous artesunate (ACT), making it impossible to evaluate CQ-susceptibility in vivo phenotype. Actually, patients with CQRPv were identified from following WHO criteria, excluding patients with any clinical complication [19]. Another clear limitation is the small number of samples used in this study. Moreover, so far it is not possible to rule out a differential virulence concurring with dissimilar drug resistance phenotypes. Studies on P. falciparum laboratory strains and field isolates of P. vivax suggest that there is marked variability in growth rates of isolates ex vivo with CQR isolates growing faster than CQ-susceptible isolates [36,37]. Since parasites from patients with severe disease have greater ex vivo multiplication rates compared to those from patients with non- severe disease [38], the possibility arises that the highly CQR isolates may be more virulent.
Mefloquine (MQ) was used in the Brazilian Amazon until the beginning of the 2000s. Indeed, CQ and MQ resistance parallel in this region [8]. Thus, even after more than one decade of nonexposure to MQ, we could not discard this effect in the present. Actually, increased pvmdr-1 copy number was found in two isolates in the Western Brazilian Amazon [39]. Unfortunately, pvmdr-1 copy numbers were not measured in our samples.
In conclusion, our study suggests that CQR and severe P. vivax malaria are associated with increased expression levels of the pvcrt-o and pvmdr-1 genes. CQRPv could be rising in the Brazilian Amazon and probably this fact could be contributing to the simultaneous spread of vivax malaria and clinical severity related to this parasite. Further studies are needed with larger number of patients and using other molecular markers to confirm the underlying pathogenesis of severe disease, and the degree to which this is related to the emergence of multidrug resistant strains of P. vivax.