Cellular dissection of malaria parasite invasion of human erythrocytes using viable Plasmodium knowlesi merozoites

Plasmodium knowlesi, a zoonotic parasite causing severe-to-lethal malaria disease in humans, has only recently been adapted to continuous culture with human red blood cells (RBCs). In comparison with the most virulent human malaria, Plasmodium falciparum, there are, however, few cellular tools available to study its biology, in particular direct investigation of RBC invasion by blood-stage P. knowlesi merozoites. This leaves our current understanding of biological differences across pathogenic Plasmodium spp. incomplete. Here, we report a robust method for isolating viable and invasive P. knowlesi merozoites to high purity and yield. Using this approach, we present detailed comparative dissection of merozoite invasion (using a variety of microscopy platforms) and direct assessment of kinetic differences between knowlesi and falciparum merozoites. We go on to assess the inhibitory potential of molecules targeting discrete steps of invasion in either species via a quantitative invasion inhibition assay, identifying a class of polysulfonate polymer able to efficiently inhibit invasion in both, providing a foundation for pan-Plasmodium merozoite inhibitor development. Given the close evolutionary relationship between P. knowlesi and P. vivax, the second leading cause of malaria-related morbidity, this study paves the way for inter-specific dissection of invasion by all three major pathogenic malaria species.


P. knowlesi merozoite invasion assay and growth inhibition assays
To observe the "teardrop" morphology of schizonts, methanol fixed, air-dried smears were Giemsa stained briefly for up to 10 seconds.
To count the different cell populations, an identical volume of CountBright Beads was added to 3x concentration SybrGreen (Thermo Scientific) stained pre-filtration, 3 μm filtrate and double filtrate samples. Samples were stained for 20 mins. Samples were then mixed thoroughly and counted using a BectonDickinson LSR Fortessa II flow-cytometer with a minimum of 2,000 bead counts collected per sample. The different cell populations were Late-stage parasites are large and contain multiple genome copies (FSC high/FITC high), segmented schizonts contain multiple genome copies but are slightly smaller than nonsegmented late-stage parasites as the RBC membrane collapses around the segmented schizont (FSC mid-high/FITC high) and merozoites are small and only contain one copy of the genome (FSC low/FITC low). Countbright beads were gated using a FITC high/FL-2 high gate. Invasion rates were calculated as percentage of RBCs invaded x [(RBCs per µl)/(merozoites per μl)] 7 .
Filtrate, containing merozoites, was added immediately to fresh RBCs pre-aliquoted in a polystyrene 96-well plate in presence or absence of invasion inhibitors as required. For PkIIA, final volume per well was 40 μl (16 µl 1.25% hematocrit in incomplete media, 4 µl 10x concentration invasion inhibitor and 20 µl merozoite filtrate). Plates were shaken at 500 -750 rpm at 37ºC for 30 mins, and then complete media supplemented with heparin or PSS was added to each well to prevent further invasion and parasites allowed to develop for 16-20 hrs in a gassed box at 37 o C. For PkIIA, after the 30-minute invasion period, RBCs were washed twice with incomplete media to remove inhibitors and then returned to culture.
To define the kinetics of invasion, E64-treated schizonts were resuspended and filtered as described above and 30 μl merozoite filtrate was added to 10 μl pre-aliquoted 2% hematocrit RBCs in 96 well plates. The plate was returned to 37 o C and agitated. At regular time points For growth inhibition assays, synchronous P. knowlesi parasites at 0.5-1.0% parasitemia and 2% hematocrit in parasite culture media were added to wells of a 96-well plate in a volume of 100 μl culture. A two-fold series dilution of each invasion inhibitor was performed across the plate. Parasitemia was assessed after 20 hrs.
Parasitemias were determined by staining with SybrGreen and analyzed by flow cytometry.
To determine invasion rates and invasion inhibition by PkIIA, parasitemia was calculated first by gating for singlet RBCs (FSC-A/FSC-W) and then infected FITC-positive cells (FSC high/FITC high). Parasitemia as calculated by PkIIA was expressed as a percentage growth of non-inhibitory control wells and parasitemia as calculated by GIA was expressed as a percentage of untreated control wells.

Immunofluorescence microscopy and electron microscopy
Immunofluorescence assay (IFA) images were acquired with a Nikon plan apo γ 100x/1.45 oil immersion lens on a Nikon eclipse Ti microscope and images processed with NIS Elements (Nikon). 3D-SIM images were acquired using the Zeiss Elyra PS1 system (Carl Zeiss) through a 63× Plan Apochromat 1.4NA oil objective (Carl Zeiss), a 1.6× magnification lens was placed in front of the PCO.edge camera (PCO, Germany) for a pixel size of 64 nm.
A CoolLED pE-300 (CoolLED) was used to locate fluorescent parasites by eye before switching to the SIM mode where 405, 488 and 561 nm lasers were used to capture data, using the appropriate filter cubes. The structured illumination pattern was generated by passing the 405, 488 and 561 nm lasers respectively through a 23, 28 and 34 µm diffraction grating. The illumination pattern was laterally shifted through 5 phases, before passing through z to capture a full stack; this was repeated through the 3 rotations to ensure homogeneous resolution enhancement. Images were visually inspected to ensure minimal bleaching occurred during acquisitions. Laser powers were set to 5 % for the 405 nm channel and 2 % for 488 and 561 nm channels, with acquisition frame rates of 100, 200 and 100 ms respectively. This raw data was then reconstructed, using the Auto noise filter and setting the output to Raw Scale, all other settings were kept as default. The reconstructed pixel sizes for the x, y plane were 32 nm.  Table S1: Growth inhibition of heparin-like compounds against P. knowlesi YH1 parasites compared to published P. falciparum growth inhibition a in vitro.