Molecular genotypes of culture-adapted clinical isolates
Seventeen P. falciparum clinical isolates were successfully culture-adapted and underwent genotypic profiling. The parasite isolates were classified into 4 groups based on the molecular markers. Group 1, parasite isolates 14 and 17, contained K13-, Exo- and PfCRT-wild-type with a single copy of pfpm2. Group 2, parasite isolates 13, harboured C580Y-K13 mutation, Exo- and PfCRT-wild-type with a single copy of pfpm2. Group 3, parasite isolates 4 and 9, had C580Y-K13, E415G-Exo, and PfCRT mutations with a single copy of pfpm2. Group 4, parasite isolates 1, 2, 3, 5, 6, 7, 8, 10, 11, 12, 15, and 16, carried C580Y-K13, E415G-Exo, and PfCRT mutations with multiple copies of pfpm2 (Table 1).
Correlation of ART- and PPQ-resistance markers with RSA and PSA phenotypes
The phenotypic analysis was performed to determine whether the molecular markers were associated with ART- and PPQ-resistance. Parasites without the C580Y-K13 mutation (Group 1) exhibited % RSA survival rate of less than 1 (a cut-off for ART resistance), while parasites with the C580Y-K13 mutation (Group 2-4) all had % survival rate of greater than 1, clearly demonstrating the correlation between the C580Y-K13 marker and ART resistance (Fig. 1A, Table S3).
The relationship between pfpm2 copy number (CN), the E415G-Exo and novel PfCRT mutations and the % PSA survival of the parasites was assessed (Fig 1B, Table S3). PPQ-sensitive parasites P. falciparum W2, IPC-4884, and IPC-5202 were used as controls. Parasites in Group 1 and 2, lacking PPQ-resistance markers (pfpm2 multiple CN, the E415G-Exo and novel PfCRT mutations) showed % PSA survival of less than 10, indicative of PPQ-sensitivity. Parasites from Group 3 and 4, which contained E415G-Exo and PfCRT mutations, exhibited % PSA survival higher than 10, indicative of PPQ-resistance. Although parasites from group 3 had a single copy of pfpm2 and parasites from group 4 had multiple copies of pfpm2, there were no observed differences in % PSA survival between these two groups.
In vitro drug susceptibility and cell morphology of culture-adapted clinical isolates
To get a better understanding of cross-resistance, the parasites were tested against a panel of antimalarial drugs (Fig. 2 and Table S4). The samples were categorized into PPQ-resistant (PPQ-R) (red bars, Groups 3 and 4) and PPQ-sensitive (PPQ-S) (blue bars, Groups 1 and 2) isolates. PPQ-S and PPQ-R parasites had similar IC50 values for AS, DHA, DOX, and ATQ. However, the PPQ-R parasites had higher IC50 for CYC and lower IC50 values for LUM, QN, CQ, and MQ. This suggests reciprocal drug resistant pattern for PPQ and the following drugs, LUM, QN, CQ, and MQ. To assess whether the PPQ resistant field isolates exhibited a second peak of survival around 0.1-10 µM (Bimodal dose-response) as was shown by Bopp et al.  in lab strains, the starting PPQ concentration was increased from 0.5 to 50 µM and the dilution series extended from 12 to 24 points (Fig. 3). PPQ-R parasites exhibited the second peak of survival around 78 -20,000 ng/ml (or 0.08-21 µM), indicating a bimodal dose-response curve. Unlike PPQ-R parasites, PPQ-S parasites did not show the second peak, and their dose-response curves were similar to that of the reference clone W2.
Ross et al.  has shown previously that pfcrt-edited Dd2 parasites developed a distended and translucent DV phenotype during the development from mid-trophozoites to mid-schizonts. This trait was specific to the pfcrt-edited Dd2 with F145I-, M343L- and G353V-PfCRT mutations and a single copy of pfpm2 but not observed in the PPQ-resistant Cambodian lines PH1008-C (multiple copies of pfpm2 and M343L-PfCRT) or PH1263-C (multiple copies of pfpm2 and H97Y-PfCRT) . To validate if this phenotype could be observed with PPQ resistant parasites collected from the field, schizont morphology was examined. Among the parasites analysed only isolate 9 exhibited the swollen and translucent DV (Fig. 3). This isolate contained F145I-PfCRT and E415G-Exo mutations but a single copy of pfpm2. Parasite isolate 4 had a similar pattern of PPQ-resistance markers (except I218F-PfCRT instead of F145I-PfCRT) but no distended DV was observed. Parasites carrying both an F145I-PfCRT mutation and multiple copies of pfpm2 (such as parasite isolates 3 and 6) did not show an altered DV morphology. This is similar to the observation found in the PPQ-resistant Cambodian lines PH1008-C and Ph1263-C .
Cloning of Cambodian P. falciparum isolate
Clinical isolates of P. falciparum are a genetically heterogeneous population of parasites. To obtain stable strains of the parasites for long term experiments, a rapid method of cloning was developed using a combination of limiting dilution and plaque assay . Several attempts were carried out to clone 8 Cambodian P. falciparum isolates (isolates 3, 4, 6, 9, 12, 14, 15, and 17) and while all of the selected samples generated a single plaque after 7 days, only clones from isolate 14 could be expanded. After 1 month, four clones from isolate 14 were established including 14-B5, 14-C6, 14-C7, and 14-F5. These clones were genotypically and phenotypically characterized and compared to parent isolate and standard lab clones P. falciparum 3D7 and W2 (Fig. S1). All 4 clones possessed the identical genotypes to the parent isolate, and the RSA and PSA survival assays reflected sensitivity to ART and PPQ (Fig. S1).
Drug susceptibility profiles of the clones and parent isolate are illustrated in Fig. 4 and Table S5. Compared to 3D7 (CQ-sensitive), all clones and the parent isolate exhibited high IC50s toward CQ similar to that of W2 (CQ-resistance). MQ IC50s for the clones were much higher than for W2 isolate (MQ-sensitive), and comparable to IC50 of D6 (MQ-resistance, IC50-MQ = 130.8 ± 15.96 nM). Collectively, based on the drug susceptibility profile and survival assays all four clones (14-B5, 14-C6, 14-C7, and 14-F5) were classified as ART- and PPQ-sensitive, but CQ-resistant and having reduced MQ sensitivity.
Drug combination testing of P. falciparum 14-B5
Since we successfully obtained a new clone from parasite field isolates, this clone was evaluated for sensitivity against a diverse array of drug combinations utilizing the HRPII-ELISA to establish synergistic, additive, and antagonistic in vitro anti-malarial drug interactions. As validation controls, the fixed ratio combinations of DHA-PPQ, CQ-CQ, and ATQ-PG were first tested against P. falciparum 3D7, W2, D6, C2B, and IPC-5202 strains as well as our 14-5B clone. Tables 2 and S6 represent summary of drug interaction and the ΣFIC50s of tested fixed drug ratio combination, respectively. DHA-PPQ revealed antagonistic interactions, as reported previously . CQ-CQ, serving as an experimental drug combination control, showed the additive interaction while ATQ-PG revealed a synergistic interaction as previously reported by Co et al. .
Potential novel combinations with either ATQ or PG were tested. When PND or MQ were combined with ATQ, we observed antagonism/toward antagonism, except for MQ-ATQ in the C2B strain (Table 2). In TQ-ATQ combination, responses varied across strains although only one line, 14-B5 clone, revealed antagonistic interaction. When combined with PG, both PND and TQ showed different responses across strains, with 14-B5 showed antagonism between TQ-PG. All of the tested parasites revealed toward synergistic drug interaction against MQ-PG combination.