Toxoplasma gondii actin filaments are tuned for rapid disassembly and turnover

Abstract

The cytoskeletal protein actin plays a critical role in the pathogenicity of Toxoplasma gondii, mediating invasion and egress, cargo transport, and organelle inheritance. Advances in live cell imaging have revealed extensive filamentous actin networks in the Apicomplexan parasite, but there is conflicting data regarding the biochemical and biophysical properties of Toxoplasma actin. Here, we imaged the in vitro assembly of individual Toxoplasma actin filaments in real time, showing that native, unstabilized filaments grow tens of microns in length. Unlike skeletal muscle actin, Toxoplasma filaments intrinsically undergo rapid treadmilling due to a high critical concentration, fast monomer dissociation, and rapid nucleotide exchange. Cryo-EM structures of stabilized and unstabilized filaments show an architecture like skeletal actin, with differences in assembly contacts in the D-loop that explain the dynamic nature of the filament, likely a conserved feature of Apicomplexan actin. This work demonstrates that evolutionary changes at assembly interfaces can tune dynamic properties of actin filaments without disrupting their conserved structure.

Figure 1:. Imaging and analysis of TgAct1 polymerization in vitro.

a, Domain organization of TgAct1 fused to a β-thymosin–HIS tag. After expression and purification from Sf9 cells, the tag was cleaved by chymotrypsin after the terminal phenylalanine producing native, untagged TgAct1. b, Coomassie stained SDS-PAGE gel of (lane 1) protein molecular weight marker; (lane 2) TgAct1–β-thymosin–HIS following HIS purification; (lane 3) TgAct1 after chymotryptic digest; (lane 4) purified TgAct1 after ion-exchange and size exclusion chromatography. c, Schematic of the in vitro polymerization assay. TgAct1 monomers (blue) were induced for polymerization and added to a blocked flow chamber absorbed with NEM-myosin II to maintain a semi-stable attachment for imaging. Epifluorescence microscopy was used to image the TgAct1 filament dynamics by inclusion of 25–50 nM actin chromobody fused to EmeraldFP (EmFP) (green), which interacts transiently with the growing filaments. d, Static epifluorescence microscopy image of TgAct1 filaments in the in vitro polymerization assay showing filament lengths > 60 µm. e, Montage of a single TgAct1 filament that is shrinking from the pointed (–) end and growing from the barbed (+) end in the presence of 16 µM TgAct1 monomers. Blue dashed line indicates the position of the (–) end of the filament at time zero. f, Kymographs for three individual filaments in the presence of 16µM TgAct1 monomers showing disassembly from the pointed (–) end (left side) and assembly from the barbed (+) end (right side). Rates for individual filaments are shown in subunits per second. g, Measuring the filament length of 5 representative filaments over time shows constant growth rate over 1200 sec of imaging. h, Plot of the rate of barbed (+) end growth in subunits/sec, per actin concentration for TgAct1 (blue) compared to skeletal actin (orange). Data shown is an aggregate of three independent preps. The average critical concentration (C_c), determined by the x-intercept of the fitted line, is higher for TgAct1 (6.5 µM) compared to skeletal actin (0.09 µM). i, Plot of barbed (–) end disassembly rate per TgAct1 concentration. The average C_c for the barbed end, determined by the x-intercept of the fitted line, is 73.5 µM. Imaging conditions: 25 mM imidazole, pH 7.4, 50 mM KCl, 2.5 mM MgCl₂, 1 mM EGTA, 2.5 mM MgATP,10 mM DTT, 0.25% methylcellulose, 2.5 mg/mL BSA, 0.5% Pluronic F-127, oxygen scavenging system (0.13 mg/mL glucose oxidase, 50 μg/mL catalase, and 3 mg/mL glucose), 37°C.

Figure 2:. Cellular concentration and kinetic analysis of TgAct1.

a, anti-actin western blot of T. gondii parasite lysate and known amounts of purified TgAct1. The average cellular concentration of TgAct1 (148 ± 7.6 µM) was determined by comparing the relative band intensity of a known number of parasites to those of known amounts of purified TgAct1. Error ± SEM, n=4. b, Comparison of TgAct1 cellular concentrations to other known model systems (79). Cartoons created with BioRender.com. c, Time course of phosphate release from 12 µM TgAct1 (blue) and skeletal actin (orange) after inducing polymerization. d, Plot of ATP hydrolysis rates over a range of TgAct1 concentrations. Conditions for ATPase and ɛATP experiments: 25 mM Imidazole, pH 7.4, 50 mM KCl, 1 mM EGTA, 2 mM MgCl₂, 0.2 mM MgATP,1 mM DTT, 37°C. e, Plot used to determine the dissociation rate constant (k_-ATP) of ATP for TgAct1 (blue) and skeletal actin (orange). Fluorescence decay of ɛATP-bound actin after addition of a large molar excess of MgATP was measured over time. k_-ATP was determined by exponential fit to the data. Conditions: 20 µM actin, 25 mM Imidazole, pH 7.4, 50 mM KCl, 1 mM EGTA, 2 mM MgCl₂,1 mM DTT, 37°C. f, Bar graph of average k_-ATP for TgAct1 (blue, 0.16 ± 0.03) and skeletal actin (orange, 0.003 ± 0.0002). Error ± SD, n=3.

Figure 3:. Comparison of unstabilized TgAct1 filament to skeletal actin filaments.

a, Micrograph of TgAct1 filaments in the presence of 1 mM Mg-ATP. b, Reconstruction of TgAct1 filaments in the presence of 1 mM Mg-ATP; protomers in shades of blue. The circle indicates the location of the D-loop in one protomer. c, Overlay of D-loops from unstabilized TgAct1 filament model (blue) and the skeletal actin filament 8d13 (orange). d/e, View of the D-loop (ribbon and sticks) and binding pocket (surface) from TgAct1 (d, blue) and skeletal actin (e, orange). D-loop residues within 5 Å of the pocket are shown as sticks. f/g, View of the D-loop (surface) and binding pocket (ribbon and sticks) from TgAct1 (f, blue) and skeletal actin (g, orange). Residues within 5 Å of the D-loop are shown as sticks, except for TgAct1 F376, which is shown for illustrative purposes only. Dotted lines indicate the locations of D-loop residues 41–44 (TgAct1) and 40–43 (skeletal actin). h/i, Surface representation of a protomer of TgAct1 (h, blue) and skeletal actin (i, orange), with the shaded region showing the buried surface area for each protomer.

Figure 4:. Comparison of stabilized and unstabilized TgAct1 filaments to actin filaments from other species.

a, The D-loops from unstabilized TgAct1 (gray), chicken skeletal actin (orange, PDB ID 8d13), P. falciparum Act1 + jasplakinolide (magenta, PDB ID 6tu4), and P. falciparum Act2 (purple, PDB ID 8ccn). b, Reconstruction of TgAct1 filaments in the presence of 33 µM jasplakinolide and 0.1 mM Mg-ATP; protomers in shades of blue. The circle indicates the location of the D-loop in one protomer. c, Overlay of volume from the D-loops of TgAct1 + jasplakinolide (blue) and unstabilized TgAct1 (gray); C, CA, and N backbone atoms from residues 36–55 in each model shown and selected residues shown as sticks. d, The D-loops from TgAct1 + jasplakinolide (blue), chicken skeletal actin (orange, PDB ID 8d13), P. falciparum Act1 + jasplakinolide (magenta, PDB ID 6tu4), and P. falciparum Act2 + jasplakinolide (purple, PDB ID 8cco). e, Overlay of the D-loop (ribbon) and binding pocket (surface) from unstabilized TgAct1 filaments (gray) and TgAct1 + jasplakinolide filaments (blue). Pro42 and Ile44 shown as sticks. Orange arrows indicate the direction of motion.

Figure 5:. The Filament Properties of Toxoplasma Actin versus Skeletal Actin.

A low barbed-end assembly rate and a high pointed-end disassembly rate leads to a high critical concentration for TgAct1 relative to skeletal actin - an effect mediated in part by changes within the D-loop of TgAct1. Paired with the ability to rapidly exchange nucleotide, actin filaments in T. gondii rapidly treadmill at concentrations where skeletal actin filaments elongate. Change in rates of TgAct1 assembly, disassembly and nucleotide exchange compared to skeletal muscle actin are indicated in grey text.