The ability of tipranavir, lopinavir, ritonavir and saquinavir to inhibit the invasion of erythrocytes by free P. falciparum merozoites was assessed directly in assays using purified merozoites [15]. The only antiretroviral protease inhibitor to demonstrate a clear dose-dependent effect on merozoite invasion in this work was lopinavir (asexual parasite IC50 3 µM), with invasion inhibitory values of 18%, 38% and 73% at 10 µM, 20 µM and 50 µM, respectively (Fig. 1). While tipranavir and saquinavir (asexual parasite IC50 values 63 ± 18 µM and 26 ± 3 µM respectively) demonstrated similar activity to lopinavir at 50 µM (75% and 85% inhibition, respectively) this activity was not dose dependent. Interestingly, while the invasion inhibitory activities of lopinavir, ritonavir (asexual parasite IC50 33 ± 3 µM) and saquinavir were negligible at clinically relevant concentrations (Cmax 15.3, 11.2 and 3.1 µM, respectively [3]), tipranavir demonstrated 75% invasion inhibition at a clinically relevant concentration (Fig. 1A; Cmax 60 to 185 µM [18]).
As merozoite treatment assays are time restricted, due to the short window of merozoite invasion and viability, and the impact of plasmepsin IX and X inhibition on protein processing and downstream egress and invasion has been shown to be time dependent [12, 13], we next sought to assess whether preincubation of developing merozoites within infected erythrocytes with selected antiretroviral protease inhibitors could impact egress and invasion. To do this we treated schizont stage parasites at ~ 46 h post invasion for 8 h, or schizont stage parasites at ~ 42 h post invasion for 12 h. Data derived from 8 h exposure experiments demonstrated that the treatment of schizonts with lopinavir resulted in a dose dependent reduction in ring stage parasites and a reduction of free merozoites. An ~ 20% accumulation of schizonts was also seen at the highest treatment concentration (Fig. 2B). These observations suggest the inhibition of schizont development or merozoite egress, rather than invasion inhibition. However the concentrations of lopinavir required for these effects was higher than clinically relevant levels (Cmax 15.3 µM [3]) and > 20-fold above the concentration required to cause in vitro asexual stage growth inhibition in longer-term (48–72 h) assays (IC50 1–3 µM and [3]; Fig. 1B). Similarly, the reduced invasion associated with tipranavir (Fig. 2A), ritonavir (Fig. 2C), and saquinavir (Fig. 2D) treatment was associated with a loss of free merozoites, an observation that is typically associated with the inhibition of schizont rupture and merozoite egress. However, these phenotypes were typically observed above clinically relevant concentrations (Fig. 2). These data suggest that treatment of parasites with antiretroviral protease inhibitors during the final ∼2–6 hrs (based on treatment 46 hrs after assay population established with an initial parasite age range of 4 hrs) of schizont development does not inhibit parasite invasion directly, but likely impacts schizont rupture at higher concentrations.
Since the essential proteolytic processing events mediated by plasmepsin IX and X may occur earlier in schizont development, we also tested the effect of a longer duration protease inhibitor treatment on synchronized schizonts ∼6–10 hrs (based on treatment 42 hrs after assay population established with an initial parasite age range of 4 hrs). Similar inhibitory profiles were observed for these long treatment experiments as for the shorter (46 h post invasion) assays. However, schizont rupture and merozoite egress inhibition was more pronounced for all treatments (Fig. 2E-H). There was also a notable inverse correlation with the inhibition of ring stages and free merozoites and the presence of schizonts in these long duration treatments. However, this phenotype was most noticeable at treatment concentrations that were generally higher than the concentrations needed to kill asexual parasites (Fig. 2). To assess whether the schizont accumulation seen in treatments was due to merozoite egress inhibition or direct schizont development inhibition, Giemsa-stained thin smears of parasites treated with tipranavir, lopinavir, ritonavir and saquinavir at concentrations where loss of merozoite invasion was evident with extended schizont treatment (Fig. 2E-H) were prepared. Unhealthy schizonts were observed in all treatments (Fig. 3), indicating that the inhibition seen in egress and invasion assays was likely the result of schizont development inhibition rather than merozoite egress or invasion inhibition.
We next attempted to directly assess whether tipranavir, lopinavir, ritonavir or saquinavir can inhibit the processing of the plasmepsin IX [12] and plasmepsin X [12, 13] substrate PfRAP1 and the PfSUB1 substrate PfMSP1. We treated schizont stage parasites (~ 42 h post invasion) with inhibitory concentrations of each drug for 6 h and then harvested cells for Western blot. At concentrations where loss of merozoite invasion was evident with extended schizont treatment (Fig. 2E-H) there was no evidence of full-length PfMSP1 (Fig. 4A) or PfRAP1 (Fig. 4B) accumulation as would be expected if the processing of these proteins was inhibited by the antiretroviral protease inhibitors. However, the treatment of parasites with concentrations of antiretroviral protease inhibitors that inhibited merozoite invasion did appear to reduce the overall expression of PfMSP1 and PfRAP1 when compared to the PfEXP2 loading control (Fig. 4C). This suggested that extended protease inhibitor treatment may impact the overall health and normal development of merozoites.