Kelch Mutations in Plasmodium falciparum Protein K13 Do Not Modulate Dormancy after Artemisinin Exposure and Sorbitol Selection In Vitro

ABSTRACT

Some Kelch mutations of the Plasmodium falciparum K13 protein confer increased survival to dihydroartemisinin (DHA)-treated ring-stage parasites. Here, we asked if K13 mutations affect a dormancy phenotype allowing parasites to survive DHA exposure and then sorbitol selection. Although recrudescence from dormancy differed between two distinct parasite lines, it was similar for isogenic lines carrying single-site substitutions in K13. Therefore, K13 mutations do not alter the DHA-sorbitol combined dormancy phenotype; rather, traits from other loci likely determine this phenotype.

TEXT

Artemisinin-based combination therapies (ACTs) are powerful additions in the global effort to control malaria, and their widespread introduction has contributed to recent reductions in malaria morbidity and mortality rates (1). Early recommendations to always include a long-acting partner drug with an artemisinin compound (ART) in ACT were based on the frequent observations of recrudescence within a few weeks of ART treatment alone (monotherapy R1 failure) (2). ACTs are nevertheless threatened by spreading drug resistance, particularly of Plasmodium falciparum parasites to ACT partner drugs in Southeast Asia (3).

Since its original demonstration (4), dormancy has been described for P. falciparum with sorbitol treatment alone and with various antimalarial compounds in vitro (5), as well as in the clinical setting after artesunate treatment (6). Following exposure to dihydroartemisinin (DHA), dormant forms of P. falciparum lines may persist in culture as nonprogressing rings that downregulate most metabolic pathways (7) and then recrudesce days to weeks later as actively growing parasites (8). Early recovery from dormancy may be an advantage to parasite survival after ART treatment (9).

Multiple host and parasite factors affect the clearance of ART-treated infections, and K13 polymorphisms have been associated with clearance half-life variations in P. falciparum lines in Southeast Asia (10). Particular amino acid substitutions in the Kelch propeller region of P. falciparum protein K13, e.g., R539T and C580Y, increase the survival of young parasite ring forms exposed to pulses of 700 nM DHA for 6 h in a ring-stage survival assay (RSA) (11, 12). Whether properties of K13 polymorphisms may also affect parasite dormancy has been an open question. Here, we compare the recrudescences after the induction of dormancy by DHA and sorbitol selection of different P. falciparum lines, including isogenic pairs of parasite lines with R539T and C580Y Kelch polymorphisms.

Parasite lines from two Cambodian field isolates, KH002-009 (K13 wild-type) (13) and 967^R539T (K13 mutant R539T), and the isogenic 967 K13 transformants 967^C580Y (K13 mutant C580Y) and 967^R539 (K13 wild-type revertant allele, based on the 3D7 reference genome) (12), were used in this study (Table 1). KH002-009 and 967^R539T also present different alleles for other genes associated with delayed parasite clearance (10). The dormancy recrudescence assay was modified from Teuscher et al. (8). Each parasite line was cultivated at 5% hematocrit in 10 ml of complete medium (CM) consisting of RPMI 1640 with 25 mM HEPES and 50 μg/ml hypoxanthine (KD Medical), 0.21% sodium bicarbonate (KD Medical), 20 mg/liter gentamicin (KD Medical), and 0.5% AlbuMAX II (Life Technologies). Ring-stage cultures at approximately 2% parasitemia were treated with 700 nM DHA (WWARN) diluted in dimethyl sulfoxide (DMSO) (Sigma) for 6 h and then washed once with 10 ml CM. On the following three consecutive days, the cultures were treated every 24 h with 5% sorbitol for 30 min at 37°C to lyse any parasites that might have progressed to mature stages from the ring stage and were not dormant. After each treatment, cells were transferred to new flasks. Cultures were monitored by two-color flow cytometry, staining for DNA and intact mitochondria using SYBR green and MitoTracker deep red, respectively (14), until parasitemia reached 2%. The medium was changed every other day, and 30 μl of 50% hematocrit stock red blood cells was added to replace the loss from sampling.

TABLE 1.

Single-site polymorphisms in K13 and four other proteins in Plasmodium falciparum parasites of Cambodia

Parasite line	Date collected	Cambodian region	WT or MT (mutation)a
			Kelch	Ferredoxin	ARP10	MDR1	CRT
967R539T	2010	Pursat	MT (R539T)	MT (D193Y)	MT (V127 M)	MT (T484I)	MT (I356T)
967R539			WT	MT (D193Y)	MT (V127 M)	MT (T484I)	MT (I356T)
967C580Y			MT (C580Y)	MT (D193Y)	MT (V127 M)	MT (T484I)	MT (I356T)
KH002-009	2011	Preah Vihear	WT	WT	WT	WT	WT

^aWT, wild type; MT, mutant; ARP10, apicoplast ribosomal protein 10; MDR1, multidrug resistance protein 1; CRT, chloroquine resistance transporter.

Parasite growth was assessed without drug pressure by adding 30 μl of packed infected red blood cells at 2% parasitemia to 10 ml of CM with 5% hematocrit (estimated initial parasitemia, 0.12%). The medium was changed each day, and daily parasitemia counts were taken by flow cytometry until the culture reached 2% parasitemia.

All experiments were performed three times on independently cultured lines. Statistical analysis was performed using GraphPad Prism 7.03. The time to recrudescence to 2% following dormancy induction and untreated growth time to 2% parasitemia were analyzed using one-way analysis of variance (ANOVA) with Tukey's multiple-comparison test.

Following the DHA and sorbitol treatments, the parasitemia levels of all four lines decreased rapidly from 2% to occasional parasites in thin blood films. Microscopic examination on the fourth day showed that dead (pyknotic) bodies were frequent, while surviving parasites appeared as small rounded forms with a condensed cytoplasmic halo around the nucleus, consistent with dormant forms as previously described (15) (Fig. 1A, top and middle rows). Flow cytometry with double staining for intact mitochondria and DNA (7) detected these dormant parasites at rates of 0.013 to 0.49% (Fig. 1B). Actively growing ring forms were not observed among the dormant parasites in the first week, but they were readily found among the recrudescent populations thereafter (Fig. 1A, bottom row).

FIG 1.

Dormant parasites and recrudescences from Plasmodium falciparum lines. (A) Top and middle rows show images of dead and dormant parasites, respectively, on the fourth day after exposure of ring stages to a 700 nM DHA pulse followed by three daily sorbitol treatments. Identification is by the classification criteria of Tucker et al. (15). Bottom row shows active ring-stage parasites from recrudescencing populations taken on the day each line reached 2% parasitemia. (B) Detection of dormant and actively proliferating parasites by flow cytometry, with double staining for intact mitochondria and DNA. Replicates of the 967^R539T recrudescences are shown in red, 967^R539 in blue, 967^C580Y in green, and KH002-009 in purple. Note the scale change along the ordinate axis to improve the display of the low-level dormant parasite counts.

Populations of the 967^R539T, 967^C580Y, and 967^R539 lines recrudesced and reached 2% parasitemia on average 16.7, 17.0, and 15.3 days after DHA treatment (P = 0.58), while the average recovery of KH002-009 was on day 26.7 (P = 0.0003). Growth control cultures, which were expanded in the absence of drug exposure or sorbitol selection, grew to 2% in all four lines within 3 to 5 days (P = 0.23), suggesting that parasite propagation rates were not responsible for large differences in recrudescence time, although these in vitro growth measures might not be sensitive enough to detect smaller modulations.

Dormancy serves as a mechanism for parasite populations to cope with environmental conditions and toxic agents that threaten their survival. Parasites may enter a dormant state to avoid metabolic pathways that drugs target, to decrease drug uptake, or to prioritize repair and recovery over replication. The results of our present study show that in vitro recrudescence of P. falciparum from a dormant state is not significantly different with the K13 R539T and C580Y polymorphisms that alter RSA outcomes. It remains to be explored whether other genes, including candidates listed in Table 1 (10), may be involved in different periods to recrudescence in parasite lines, such as 967^R539T and KH002-009, perhaps through cell cycle arrest or an alternative nutrient acquisition pathway. The discovery of genes that determine dormancy will be of fundamental interest to our understanding of parasite biology as well as to the development of new medicines for a complete cure of malarial infections.