Plasmodium myosin A drives parasite invasion by an atypical force generating mechanism

Abstract

Plasmodium parasites are obligate intracellular protozoa and causative agents of malaria, responsible for half a million deaths each year. The lifecycle progression of the parasite is reliant on cell motility, a process driven by myosin A, an unconventional single-headed class XIV molecular motor. Here we demonstrate that myosin A from Plasmodium falciparum (PfMyoA) is critical for red blood cell invasion. Further, using a combination of X-ray crystallography, kinetics, and in vitro motility assays, we elucidate the non-canonical interactions that drive this motor's function. We show that PfMyoA motor properties are tuned by heavy chain phosphorylation (Ser19), with unphosphorylated PfMyoA exhibiting enhanced ensemble force generation at the expense of speed. Regulated phosphorylation may therefore optimize PfMyoA for enhanced force generation during parasite invasion or for fast motility during dissemination. The three PfMyoA crystallographic structures presented here provide a blueprint for discovery of specific inhibitors designed to prevent parasite infection.

Fig. 1

PfMyoA is critical for red blood cell invasion by merozoites. a Schematic showing replacement of the wild-type myoa locus with the full-length mutant locus by single crossover recombination in the myoa HR region using a T2A skip peptide (yellow box) to couple genomic integration to neomycin selection. b Treatment of two MyoA-cKO clones (B9 and H6) or WT with rapamycin (RAP) shows an almost complete invasion block (~90%) compared to DMSO treatment (−). This is comparable to the defect with known invasion inhibitors heparin and Cytochalasin D (CytoD) (90-95%). Parasitaemia was measured in the following cycle (cycle 1) by flow cytometry as the percentage of red blood cells (RBCs) that were DNA-positive by staining with SYBR Green I. Mean of three biological replicates (except for WT+heparin: two biological replicates), each three technical replicates, ± standard deviation (S.D.) of biological replicates. Data from each biological replicate were normalized to the DMSO-treated sample for each parasite line. Significance assessed using parametric t-test (paired, two-tailed). c Genotyping PCR of WT or cKO parasites following DMSO (−) or RAP (+) treatment detecting the wild-type (WT, red half-arrow), unexcised (UN, blue half-arrow) and excised (EX, green half-arrow) pfmyoa loci. d Immunofluorescence analysis of cKO parasites following DMSO or RAP treatment. FLAG-tag is detectable in DMSO-treated schizonts fixed ~48 h post-treatment, colocalising with motor complex protein GAP45, while after RAP treatment GFP but not FLAG is detectable, and the signal is restricted to the cytosol, consistent with a non-functional truncated MyoA. Scale bar 1 μm. Image stacks were deconvolved using the EpiDEMIC plugin for Icy, with a z-step size of 200 nm. e Western blot of WT and cKO parasites following DMSO (−) or RAP (+) treatment. Parasites were lysed ~40 h post-treatment, before the end of cycle 0. In DMSO-treated cKO parasites, the original FLAG-tag is detectable, but following RAP-treatment, only GFP is detectable. PfAct1 was used as a loading control. Representative blot shown from three biological replicates

Fig. 2

Structural states and motor cycle of PfMyoA. a The crystallographic structures of the three states of the motor cycle of PfMyoA are represented along the motor cycle: the Rigor-like and the Post-rigor (PR) states reveal how the motor detaches from F-actin upon ATP binding; the Pre-powerstroke (PPS) state corresponds to the state in which hydrolysis occurs and that rebinds to F-actin to trigger the powerstroke. The Rigor-like state is the conformation the motor adopts at the end of the powerstroke when hydrolysis products have been released. To appreciate the movement of the converter and how it can be amplified by the rest of the lever arm, the IQ region and the two LCs (PfELC and MTIP, Supplementary Fig. 1) are represented schematically in continuity of the last helix of the converter. PfMyoA displays the four canonical subdomains which are the hallmark of the myosin superfamily: N-terminus (N-term) (gray), Upper 50 kDa (U50) (marine blue), Lower 50 kDa (L50) (tint) and the converter (green) and central elements (including a beta-sheet) forming the transducer (dark cyan). During the motor cycle, rearrangements in the motor domain are allosterically transmitted through the Relay helix (yellow) and are amplified by the swing of the converter with the rest of the lever arm. b In the center diagram, a zoom of the Rigor-like structural state is presented to show the opposite side of the motor domain compared to a. Note the position of the unique N-terminal extension (purple) which is close to the connectors that direct rearrangements between motor subdomains (Switch-2 (orange), Relay (yellow) and SH1-helix (red)). PfMyoA employs an atypical motor mechanism in which the N-term extension (purple) compensates for non-canonical sequences in subdomain connectors essential for motor function

Fig. 3

Sequence alignment of connectors essential in driving motor conformational changes–lack of canonical residues in PfMyoA. a Sequence comparison of several key elements involved in myosin mechanical transduction and allosteric communication. The consensus is represented on top of each sequence and the PfMyoA sequence is compared to other myosins. This comparison shows that the sequences of the Switch-2, Wedge, and Relay differ from the consensus. Consensus code: a is an aromatic residue, h a hydrophobic residue, x colored in red is an acidic residue. Residues important in allosteric communication in the motor are indicated in rectangles (blue rectangle if the residue is involved in a conserved interface, red rectangle if the residues are involved in interactions specific to PfMyoA). If the position is not conserved in PfMyoA, the residue is colored in dark red. To see how these residues affect the PfMyoA motor rearrangements along the cycle, see Fig. 4 and Supplementary Movie 2. b Switch-2/Wedge interactions in classical Myo2. c Conformational changes of two connectors before (PPS, transparent) and after (Rigor-like, plain) the powerstroke in Myo2. The Relay helix unkinks while the SH1-helix undergoes a piston-like movement using the canonical flexibility found at the fulcrum ^SH2-SH1Gly (pink ball)

Fig. 4

The unconventional mechanism of force production by PfMyoA. a Overall view of the mechanical communication within the PfMyoA motor domain during the powerstroke. In both the PPS and the Rigor-like states, interactions between Switch-2 and the Wedge are maintained (highlighted by dashed blue lines and detailed in b). The Rigor-like structure indicates that the sequential release of hydrolysis products upon the powerstroke triggers displacement of the Wedge which is associated with straightening of the Relay helix and the converter swing. The PfMyoA unconventional powerstroke requires sequence compensation near the Ser691 (blue dashed lines) since this residue is bulkier than the canonical ^SH2-SH1Gly found in classical myosins at this position. Thus, motor domain rearrangements are allowed by changes in the interactions between the Wedge and the Relay and SH1-helix connectors (details in b). Additional interactions (highlighted by dashed red lines) involving the N-term extension (purple) also stabilize the Rigor-like state and compensate for the immobility of the SH1-helix. (Details are shown using the same view in c). b Non-conserved residues are highlighted by a red rectangle. The SH1-helix lacks the conserved glycine ^SH2-SH1Gly at the fulcrum which is replaced by a serine (light pink spheres, S691). A hydrogen bond is formed between S691 and ^RelayQ494 in both the PPS and Rigor-like states. The presence of a less pliant fulcrum requires sequence adaptation in the Wedge and in the Relay and results in the immobility of the SH1-helix during the powerstroke. ^Switch-2V475 establishes hydrophobic interactions with ^SH2H688, helping to stabilize the Rigor-like position of the Switch-2 (red dashed lines). E503 (shown in pink) is a reporter to indicate the kink of the Relay. c In PfMyoA, the converter establishes a network of interactions with the Relay and the SH1-helix. Non-conserved residues are highlighted by a red rectangle. An electrostatic bond between phosphoserine 19 (SEP19) (N-term extension) and K764 (converter), as well as a salt bridge-π interaction between the ^{N-term extension}E6, ^Switch-1R241, ^Switch-2F476 (red dashed lines) stabilize the position of the converter in the Rigor-like conformation. For comparison with Myo2, see Supplementary Fig. 5 and Supplementary Movies 2 and 3

Fig. 5

Functional properties of wild-type (WT) and mutant full-length PfMyoA constructs. a Schematic of constructs used for functional analysis. Ser19 is fully phosphorylated during protein expression in Sf9 cells (Supplementary Fig. 6). b Speed distributions from a representative in vitro motility assay. WT, 3.88 ± 0.54 µm/s; S19A, 2.07 ± 0.28 µm/s; K764E, 1.75 ± 0.22 µm/s; and ∆N, 0.23 ± 0.04 µm/s. Values, mean ± SD (Supplementary Table 3 shows data from multiple protein preparations). c Actin-activated ATPase activity for WT, V_max = 138 ± 4 s⁻¹ and K_m = 30.3 ± 2.3 µM; S19A, V_max = 74.0 ± 2.0 s⁻¹ and K_m = 8.5 ± 1.0 µM; K764E, V_max = 72.9 ± 1.9 s⁻¹ and K_m = 18.2 ± 1.4 µM; and ∆N, V_max = 9.13 ± 0.20 s⁻¹ and K_m = 7.34 ± 0.67 µM. Data from 2 protein preparations and 3 experiments for each construct were fitted to the Michaelis-Menten equation. Error, SE of the fit. d ADP release rates from acto-PfMyoA. WT, 334 ± 36 s⁻¹; S19A, 115.8 ± 10.9 s⁻¹; K764E, 103.5 ± 8.6 s⁻¹; ∆N, 10.32 ± 0.91 s⁻¹. Values, mean ± SD. WT vs. any other construct. p < 0.0001; S19A vs. K764E, NS; S19A or K764E vs. ∆N, p < 0.01 (one way ANOVA followed by a Tukey’s Honest Significant Difference post-hoc test). Data from at least 3 protein preparations of each construct at different temperatures are shown in Supplementary Table 4. e Ensemble force measurements using a utrophin-based loaded in vitro motility assay. A myosin that produces more force requires higher utrophin concentrations to arrest motion: WT, 1.40 ± 0.08 nM; S19A, 2.42 ± 0.17 nM; K764E, 3.04 ± 0.30 nM; ∆N, 10.8 ± 0.8 nM. Error, SE of the fit. Data from two protein preparations and three experiments for each construct. Supplementary Fig. 7b shows these force data and fits extended to higher utrophin concentrations. Supplementary Fig. 7c–e shows ∆N data shown with an expanded y-axis. Skeletal actin was used for all experiments. Temperature, 30 °C. Source data are provided as a Source Data file

Fig. 6

Phosphorylation of PfMyoA tunes its motor properties. Scheme representing how phosphorylation tunes PfMyoA motor properties and how this could optimize the motor for parasite motility or invasion at different stages of the parasite. In highly motile stages, sporozoites move at a speed higher than 2 µm/s. At this stage, phosphorylation of PfMyoA would allow the parasite to move actin at maximal speed but with low ensemble force. In merozoites, which lack continuous motility, but are instead adapted for erythrocyte invasion, dephosphorylation of PfMyoA localized at the invasion junction would result in active motors efficient in invasion. This merozoite PfMyoA motor would spend more of its total cycle time strongly bound to actin, thereby resulting in greater ensemble force output. The apparent duty ratio in the two phosphorylation states was estimated from the rate of ADP release divided by the total ATPase cycle time. The S19A, K764E and ∆N mutants would likely impair speed but not invasion, because their ensemble force is higher than phosphorylated, wild-type PfMyoA. Phosphorylation of the N-term extension of PfMyoA can thus act as a switch to tune motor activity depending on the needs of the parasite (high speed and low force for gliding, or higher force for the low speed invasion process)