Influence of oxygen on asexual blood cycle and susceptibility of Plasmodium falciparum to chloroquine: requirement of a standardized in vitro assay

Drug resistance of Plasmodium falciparum, the most deadly human malaria parasite with nearly 500 millions of new clinical cases each year [1], makes malaria control more difficult [2, 3]. There are basically three approaches to the assessment of the antimalarial drug susceptibility of P. falciparum: in vivo assays as defined by the World Health Organization [4], in vitro assays and molecular markers of resistance [5].

In a number of laboratories surveying malaria drug resistance, in vitro tests are performed using the uptake of a radiolabelled nucleic acid precursor [³H]-hypoxanthine [6] as a marker of parasite growth. Others methods can be used: the WHO schizont maturation tests by optical microscopy (Mark III) with pre-dosed plates [7], which was based on the method of Rieckmann et al [8] and of Wernsdorfer [9], a flow cytometric analysis of propidium iodide incorporation into parasite, which permits a stage-specific evaluation of antimalarial compounds [10], and colorimetric assays with the measurement of Histidine Rich Protein II (HRP2) by an enzyme-linked immunosorbent assay (ELISA) [11, 12] and the DELI-microtest (Double-site Enzyme-linked Lactate dehydrogenase Immunosorbent assay) [13, 14].

Many factors can influence the results of the chemosusceptibility tests [15]: the initial parasitaemia, the haematocrit, the time of incubation, the time point when [³H]-hypoxanthine is added, the use of serum substitutes and the gas mixture. Laboratories using isotopic microtest to monitor drug resistance work at different oxygen tensions: 3% O₂ [10], 5% O₂ [11, 12], 10% O₂ [16], in candle jars [13, 14] (which corresponds to approximately 15% O₂ [17]) and 21% O₂ [15] (in CO₂ incubators). WHO recommends the use of a candle jar in their in vitro microtests (Mark III). But all have adopted the same threshold for the resistance to antimalarial compounds under different oxygen tensions. The aim of this study was to evaluate the influence of oxygen on the asexual blood cycle and the in vitro chemosusceptibility of P. falciparum to chloroquine in order to contribute to the establishment of a standardized in vitro assay protocol.

In order to determine the effect of oxygen on asexual blood stages of P. falciparum, synchronous cultures of 3D7 were exposed to 5, 10 and 21% O₂ during two cycles of parasites. As shown in Figure 1A, there were no significant difference between parasitaemia or parasite stage distribution under 5 and 10% O₂ during all the experiment. On the contrary, from 35 h to the rest of experimental points, a significant difference between parasite stages was observed (Figure 2), for example at 40 h, there were 42% of rings and 44.9% of schizonts under 5% O₂ and 21.7% of rings and 67.6% of schizonts under 21% O₂ (p < 0.001, Chi-square test). After the complete reinvasion around 60 h, parasitaemia were the same in the three experimental conditions and no morphologic alteration were observed by light microscopy (Figures 3A and 3B). These results have shown that parasite exposure under 21% O₂ had no lethal effect on parasites, but increased the length of schizogony. Moreover, only mature trophozoïtes and schizonts were susceptible to an effect of exposure to 21% O₂ (Figure 4).

The delay was also observed with W2 clone (Figure 1B), with a briefer schizogony (8 hours for W2 and 11 hours 3D7) and a higher multiplication index for W2 (% of infected red blood cells by rings after reinvasion/% of infected red blood cells by rings at the beginning of the previous cycle), (8 for W2 and 3 for 3D7).

No significant difference were observed in tritiated hypoxanthine uptake between 10 and 21% O₂ with the two clones 3D7 and W2 (Figures 5A and 5B), (p = 0.087 and p = 0.76 respectively). The maximum of incorporation (around 2,500 cpm with 3D7 and 4,500 cpm with W2) was achieved at 50 hours with a steady-state until 60 hours for 3D7 and W2 in the two experimental conditions. 3D7 and W2 chloroquine IC₅₀ were evaluated under 5, 10 and 21% O₂ (three experiments). No significant difference were observed with 3D7, the IC₅₀ were 15.8 nM [10.8–20.7], 14.1 nM [12.4–16.9], 17.6 nM [14.1–21.1] under 5% O₂, 10% O₂ and 21% O₂ respectively. On the contrary, with W2 clone, IC₅₀ under 21% O₂ was significantly lower than IC₅₀ under 5% O₂ and 10% O₂, 83.3 nM [15–168], 299 [228–369] and 277 [192–322] respectively (p < 0.05, Student's T test).

One hundred and thirty six Plasmodium falciparum isolates were used to evaluate chloroquine susceptibility at 10% O₂ and 21% O₂. Three tests at 10% O₂ and 13 tests at 21% O₂ were not interpretable, because of lack of significant difference in cpm between control-wells and higher chloroquine concentration wells. Chloroquine susceptibility of 120 isolates was finally tested at 10 and 21% O₂ (Figure 6A). The differences between the chloroquine IC₅₀ of the 120 isolates at 10% O₂ and at 21% O₂ were statistically significant (p < 0.0001, Wilcoxon signed-rank test), means of 173.5 nM with a standard deviation of 168.4 and 121.5 nM with a standard deviation of 106.7 respectively. The median and interquartile (Q25% & Q75%) values of the chloroquine IC₅₀ at 10% O₂ and IC₅₀ at 21% O₂ of the 120 isolates were respectively 127 nM, 30 nM, 258 nM at 10% O₂ and 85.9 nM, 36 nM, 168 nM at 21% O₂. In particular of interest, 52.5% of isolates had chloroquine IC₅₀ > 100 nM at 10% O₂ although 42.5% of the same isolates had chloroquine IC₅₀ > 100 nM at 21% O₂. Moreover, among the 63 isolates which had chloroquine IC₅₀ > 100 nM at 10% O₂, 17 had chloroquine IC₅₀ < 100 nM at 21% O₂ (Figure 6B) (i.e. 1/3) although five isolates only had chloroquine IC₅₀ > 100 nM at 21% O₂ among the 57 isolates which had chloroquine IC₅₀ < 100 nM at 10% O₂ (Table 1), (Chi-square test for paired samples, p < 0.025).

There were no significant difference between the observed control cpm at 10 (mean 12,607, standard deviation 13,425) and 21% O₂ (mean 13,302, standard deviation 14,662).

The first works about oxygen effects on P. falciparum asynchronous cultures [17] had shown that microaerophilic environment allowed an optimal development of parasites. Their growth was impossible in strict anaerobic conditions. P. falciparum possesses a functional mitochondrial respiratory chain with oxygen consumption [25]. It has been shown that there is some protector effect of CO₂ at high oxygen concentration [17] through the medium pH which stability (between 7.2 et 7.45) is required for parasite growth [26]. The analysis of the different parasites stages distribution according to the time and the oxygen concentration has allowed us to reveal a possible slow down of cellular cycle without morphologic alteration of parasites under 21% O₂.

In the present study, the exposure to 21% O₂ did not change parasitaemia growth rate after complete reinvasion. Previous studies [27] had shown the same absence of oxygen effect on synchronous cultures parasitaemia during four days. Here, it has been shown that the exposure of 3D7 and W2 P. falciparum clones to 21% O₂ had parasitostatic effect by lengthening the schizogony. Mature stages had a particular susceptibility to high oxygen concentration. Moreover, these results justified to test drug susceptibility after a 60 hours period of incubation with [³H]-hypoxanthine. At the end of trophozoïtes stage, DNA replication begins and is followed by a succession of S phases during the schizogony [28]. Under 21% O₂, the nucleic acids synthesis decreased from 30 hours for the two P. falciparum clones comparing to 5% O₂ exposure. Under 5% O₂, a low [³H]-hypoxanthine incorporation took place for the first twenty hours, probably corresponding to RNA synthesis in rings, followed by an increase between 20 and 30 hours in possible relation with the beginning of DNA replication at late trophozoïtes stage [29]. Between 30 and 40 hours, during schizogony, DNA synthesis increased and the peak was achieved at 50 hours [30].

In the present study, the chloroquine IC₅₀ at 10% O₂ were significantly higher than those at 21% O₂. A previous study did not show oxygen dependent effects of chloroquine on P. falciparum in culture [31], but in that experiment only four strains were tested. In the present work, among the 63 isolates which had chloroquine IC₅₀ > 100 nM at 10% O₂, 17 had chloroquine IC₅₀ < 100 nM at 21% O₂. The effect of gas mixture on the results of chloroquine chemosusceptibility should led different laboratories involved in malaria resistance survey to adapt a resistance threshold for each gas mixture or to use the same conditions to perform isotopic microtests.

Several factors influencing the results of the chemosusceptibility tests (the initial parasitaemia, the haematocrit, the time of incubation, the time point when [³H]-hypoxanthine is added, the use of serum substitutes) have already been investigated by Basco [15, 32]. The present data suggest the importance of the gas mixture on isotopic microtest results for chloroquine. Further studies are needed to evaluate gas mixture impact on isolates susceptibility to other antimalarial compounds and their correlation with molecular markers of resistance and in vivo evaluation of drug efficacy. Other investigations about the preparation of drug solutions, the storage of pre-dosed culture plates are required before in vitro drug sensitivity assay can become a standardized tool for laboratories to validate the threshold for resistance in respect to the clinical responses and molecular markers.

The author(s) declare that they have no competing interests.

The views and opinions are those of the authors and do not purport to represent those of the French Ministry of Defense.