Compositional and expression analyses of the glideosome during the Plasmodium life cycle reveal an additional myosin light chain required for maximum motility

Abstract

Myosin A (MyoA) is a Class XIV myosin implicated in gliding motility and host cell and tissue invasion by malaria parasites. MyoA is part of a membrane-associated protein complex called the glideosome, which is essential for parasite motility and includes the MyoA light chain myosin tail domain-interacting protein (MTIP) and several glideosome-associated proteins (GAPs). However, most studies of MyoA have focused on single stages of the parasite life cycle. We examined MyoA expression throughout the Plasmodium berghei life cycle in both mammalian and insect hosts. In extracellular ookinetes, sporozoites, and merozoites, MyoA was located at the parasite periphery. In the sexual stages, zygote formation and initial ookinete differentiation precede MyoA synthesis and deposition, which occurred only in the developing protuberance. In developing intracellular asexual blood stages, MyoA was synthesized in mature schizonts and was located at the periphery of segmenting merozoites, where it remained throughout maturation, merozoite egress, and host cell invasion. Besides the known GAPs in the malaria parasite, the complex included GAP40, an additional myosin light chain designated essential light chain (ELC), and several other candidate components. This ELC bound the MyoA neck region adjacent to the MTIP-binding site, and both myosin light chains co-located to the glideosome. Co-expression of MyoA with its two light chains revealed that the presence of both light chains enhances MyoA-dependent actin motility. In conclusion, we have established a system to study the interplay and function of the three glideosome components, enabling the assessment of inhibitors that target this motor complex to block host cell invasion.

Keywords: cell motility; glideosome; invasion; malaria; myosin; myosin light chain; plasmodium.

Figure 1.

Expression of GFP-tagged MyoA through the P. berghei life cycle. A, transcription of myoa analyzed by quantitative RT-PCR normalized against two control genes, hsp70 and arginine-tRNA synthetase, at different stages of the life cycle. The mean normalized myoa expression is shown as a horizontal bar, and S.D. values are indicated by error bars. AS, asynchronous asexual blood stages; Sch, schizonts; NAG, non-activated gametocytes; AG, activated gametocytes; Ook, ookinetes; Spor, sporozoites. B, a plasmid containing the sequence for the 3′-end of the MyoA coding region fused in frame to the gfp sequence, together with a human dhfr-selectable marker, was used to insert the gfp sequence into the endogenous myoa gene. C, successful integration into the P. berghei genome was confirmed by diagnostic PCR on genomic DNA prepared from wild-type parasites and those with integrated plasmid sequence using primers 9 and 10, which amplify a 1.1-kb DNA fragment from the modified myoa locus following correct integration. D, expression of the GFP-tagged MyoA was confirmed by Western blotting of lysates from parasites expressing either the unfused GFP or the MyoA-GFP proteins; as indicated, GFP-specific antibodies reacted with either a 29- or a 120-kDa protein, respectively. E, extracts of schizonts following hypotonic lysis in the soluble fraction (S) and further fractionation of the insoluble material by carbonate buffer into carbonate-soluble (peripheral membrane (PM)) and insoluble (integral membrane (IM)) fractions. Schizonts from parasites expressing either GFP (top row) or MyoA-GFP (bottom row) were used. F, expression of MyoA-GFP at the invasive and motile stages of the parasite life cycle, detected by live fluorescence microscopy. The parasite nuclei are stained with Hoechst 33342, and the ookinete surface is stained with Cy3-conjugated anti-P28 antibody 13.1. In the merged color image, these are blue and red, respectively, and the MyoA-GFP is green. The differential interference-contrast (DIC) bright field images are also shown. Scale bar, 5 μm. G, expression of MyoA-GFP in liver-stage parasites at the early and late cytomere stages. Green, GFP fluorescence; blue, Hoechst 33342 staining. Scale bar, 10 μm. H, temporal profile of MyoA-GFP expression during the six stages of P. berghei ookinete development. MyoA-GFP was detected by live fluorescence microscopy associated with the protuberance that grows out of the spherical body only at stage III, eventually forming the motile ookinete. The parasites were co-stained with Hoechst 33342 and with Cy3-conjugated anti-P28 antibody 13.1 as a marker for the zygote and ookinete surface. The merged (MyoA-GFP (green), P28 (red), and Hoechst (blue)) and DIC bright field images are also shown. Scale bar, 5 μm.

Figure 2.

MyoA-GFP expression and glideosome complex formation during the late stages of intracellular development of P. falciparum in the red blood cell. A, MyoA-GFP expression was detected by live fluorescence microscopy, and the nuclei were detected by staining with DAPI. The merged color image with MyoA-GFP (green) and DAPI (blue) and DIC bright field images are also shown. Schizogony starts at around 30 h postinvasion, and MyoA-GFP is detected from 38 to 40 h in multinucleated forms. Scale bar, 2 μm. B, parasites expressing MyoA-GFP were collected at the indicated time points postinvasion, and lysates were prepared; samples of these lysates (i) and of proteins precipitated from them with a GFP-specific antibody (ii) were fractionated by SDS-PAGE and probed on Western blots with antibodies to GFP, MTIP, GAP45, and GAP50. Molecular mass markers are indicated in kDa.

Figure 3.

The location of MyoA during P. falciparum merozoite invasion. A, MyoA-GFP is located at the periphery of developing intracellular (top row of each pair of images) and free extracellular merozoites (bottom row) and is colocalized with antibodies specific for IMC proteins MTIP, GAP45, and GAP50. In the merged color image, the MyoA-GFP signal is green and antibodies specific for the IMC proteins are red; nuclei are stained blue with DAPI. The DIC image is also shown. B, individual merozoites are captured at different stages of invasion from initial attachment, through early and late invasion to the intracellular ring stage. MyoA-GFP remains peripheral, whereas RON4, initially in the apical rhoptry neck, relocates during invasion. Merged color images with MyoA-GFP (green), RON4 (red), and nuclei (blue) and DIC images are also shown, together with a schematic of each cell pair. Scale bar, 2 μm.

Figure 4.

Identification of GAP40 in the P. falciparum glideosome complex. A, wild-type (lane 1) or GAP45-GFP–expressing (lane 2) P. falciparum 3D7 parasites were incubated with radiolabeled phosphate, and proteins binding to a GFP-specific antibody were immunoprecipitated, resolved on an SDS-polyacrylamide gel, and detected by autoradiography. Bands corresponding to two unidentified proteins are labeled with one or two asterisks. In parallel, the same samples were probed with antibodies to known components of the glideosome complex: MyoA, GAP45, GAP50, and MTIP. B, the immunoprecipitation with anti-GFP antibody resin was repeated with unlabeled schizont lysate from wild-type (lane 1) or GAP45-GFP–expressing (lane 2) P. falciparum 3D7 parasites. Immunoprecipitated proteins were fractionated by SDS-PAGE and detected with SYPRO Ruby, and the indicated bands were excised. C, proteins identified in the three excised bands by tryptic digestion and mass spectrometry. The number of unique peptides identified for each protein is indicated, along with the percentage of the protein sequence that these peptides cover. IP, immunoprecipitation; WB, Western blotting.

Figure 5.

The putative ELC binds to MyoA but not MyoB and is in the P. falciparum glideosome complex together with MTIP. Shown is alignment of Plasmodium ELC homologues (A) and PfELC (B) with the two T. gondii MyoA ELCs. Identical residues are shaded black, and biochemically similar residues are shaded gray. Percentage identities to PfELC are displayed. Gene identifiers are as follows. PfELC, PF3D7_1017500; PrELC, P. reichenowi PRCDC_1016900; PvELC, Plasmodium vivax PVX_001745; PcyELC, Plasmodium cynomolgi PCYB_061180; PkELC, Plasmodium knowlesi PKNH_0601700; PcELC, Plasmodium chabaudi PCHAS_0501900; PyELC, Plasmodium yoelii PY02639; PbELC, P. berghei PBANKA_0501800; TgELC1, T. gondii TGME49_069440; TgELC2, T. gondii TGME49_305050. C, structural homology model of ELC generated using Phyre 2. D, proteins in extracts from wild-type (3D7) schizonts and those expressing either MyoA-GFP or MyoB-GFP were immunoprecipitated with resin-bound antibodies to GFP and then probed by Western blotting with antibodies to GFP and ELC. I, input lysate; U, unbound protein; E, protein eluted from the resin. E, the protein complex containing MTIP also contains ELC. The glideosome complex from lysates of schizonts (S) and merozoites (M) of 3D7 was precipitated with anti-MTIP antibodies and resolved by SDS-PAGE, and the presence of ELC was determined by Western blotting with specific antibodies. A preimmune antibody sample was used to immunoprecipitate from the lysates as a control. IP, immunoprecipitation; WB, Western blotting.

Figure 6.

The putative MyoA ELC binds to the neck region of MyoA. A, purification of recombinant His-PfELC protein. B, far-UV (i) and near-UV (ii) circular dichroism spectra of recombinant protein indicate that it is highly structured, with 32% α-helix, 17% β-sheet, 20% turn. C, PfELC binding to immobilized peptides corresponding to residues 770–787, 786–803, and 801–818 of MyoA was measured by biolayer interferometry, using increasing concentrations of PfELC. The dissociation constant (K_d) with S.E. was calculated from each of the curves. D, the amino acid sequence of the neck region of MyoA, showing the location of the peptides used for analysis, the MTIP-binding site, and the proposed ELC-binding site.

Figure 7.

PfELC expression matches that of known glideosome components in P. falciparum schizonts. A, scheme for modifying the PF3D7_1017500 gene locus by single-crossover homologous recombination to C-terminally modify the ELC protein with a triple-HA tag. B, PCR screening of parental 3D7 parasites and two independently generated clones. PCR product from the wild-type locus was amplified using primers 13 and 14 (WT), whereas a product (I) from the modified locus was amplified using primers 13 and 15. C, Western blot analysis of HA-tagged ELC, MTIP, and GAP45 proteins present during a time course of schizont development. D, immunofluorescent detection of HA-tagged ELC and either MTIP or GAP45 in mature schizonts and merozoites. Merged color images are also shown, with ELC-HA (green) and MTIP or GAP45 (red). Nuclei are stained with DAPI (blue). Scale bar, 2 μm.

Figure 8.

Recombinant MyoA binds actin. A, Coomassie-stained SDS-PAGE showing affinity-purified GST-MyoA, marked with an asterisk. B, an actomyosin event as determined using optical tweezers, with an actin filament held between two beads, a left-hand bead (blue) and a right-hand bead (red), with the MyoA molecule on a third bead. Because of the high sensitivity required in these single-molecule measurements, Brownian motion causes a significant amount of displacement noise, and when MyoA binds to actin, that event is evident by a sudden decrease in the S.D. of the displacement noise (shown in black). Events are scored by thresholding data falling below a fixed value (dotted red/blue line). Note that movement of the left-hand bead and its S.D. are offset by −200 and −30 nm, respectively, for clarity. C, improved visualization of binding events is achieved by moving the beads back and forth using a 10-Hz triangular wave function of 100-nm amplitude. D, displacement histogram generated from 244 acto-MyoA–binding events with a fitted Gaussian distribution. The magnitude of the displacement (nm) is displayed on the x axis, and the number of observations at that value (N_obs) is displayed on the y axis. The average movement caused by the myosin power stroke was measured as 3 nm. E, acto-MyoA event lifetime distributions measured at two Mg·ATP concentrations, 8 μm (blue symbols) and 0.3 μm (red symbols), confirm that the event lifetimes are strongly dependent on [Mg·ATP]. The x axis shows the duration of the binding event (in seconds); the y axis is the number of observations (N_obs) at that value. The lines are single-exponential least-square fits with rate constants 53 s⁻¹ (blue) and 6.9 s⁻¹ (red).

Figure 9.

MTIP/ELC-decorated MyoA shows increased in vitro sliding motility. A, size exclusion chromatography of co-expressed MyoA and MTIP (solid line) and MyoA, MTIP, and ELC (dotted line). B, Coomassie-stained SDS-PAGE of the purified MyoA-light chain complexes. C, motility of nitrocellulose-captured MyoA complexes; D, the corresponding capture of MyoA complexes via an anti-His tag antibody measuring the velocity of actin filament movement in the presence of ATP. A solid line represents a Gaussian fit to each corresponding histogram. E, the actin-activated ATPase of MyoA-MTIP and MyoA-MTIP-PfELC was estimated using an NADH-coupled assay to quantify the rate of ATP hydrolysis.