Alveolar proteins stabilize cortical microtubules in Toxoplasma gondii

Abstract

Single-celled protists use elaborate cytoskeletal structures, including arrays of microtubules at the cell periphery, to maintain polarity and rigidity. The obligate intracellular parasite Toxoplasma gondii has unusually stable cortical microtubules beneath the alveoli, a network of flattened membrane vesicles that subtends the plasmalemma. However, anchoring of microtubules along alveolar membranes is not understood. Here, we show that GAPM1a, an integral membrane protein of the alveoli, plays a role in maintaining microtubule stability. Degradation of GAPM1a causes cortical microtubule disorganisation and subsequent depolymerisation. These changes in the cytoskeleton lead to parasites becoming shorter and rounder, which is accompanied by a decrease in cellular volume. Extended GAPM1a depletion leads to severe defects in division, reminiscent of the effect of disrupting other alveolar proteins. We suggest that GAPM proteins link the cortical microtubules to the alveoli and are required to maintain the shape and rigidity of apicomplexan zoites.

Fig. 1

GAPM proteins localise to the IMC and are enriched at ring-like structures. a Cladogram of the GAPM proteins conserved among apicomplexans. Predicted membrane topology for each clade is illustrated. b Table of GAPM genes, their transcriptional expression, and their importance for tachyzoite-stage T. gondii and blood-stage P. berghei. c GAPM proteins were C-terminally tagged with YFP or mCherry and visualised by SR-SIM in relation to acetylated tubulin. Projections of z-stacks of dividing cells are shown. GAPM1b-YFP and GAPM2b-YFP were expressed at a low level and therefore visualised indirectly with antibodies against GFP. All proteins were localised to the IMC in both mature and dividing cells. d Endogenously tagged GAPM proteins co-localise throughout the IMC and at ring structures (detail). Detail scale bar is 100 nm. e Live-cell SR-SIM of GAPM1a-YFP. Z-stacks were acquired every 7 min and projected to show fixed ring structures within the IMC that do not change in size or position over the course of 90 min. Images are representative of three independent experiments. Detail scale bar is 500 nm. Unspecified scale bars are 5 μm

Fig. 2

Conditional depletion of GAPM1a results in severe replication defects. a Strategy to experimentally determine protein topology. b Localisation of the mCherry-labelled GFP nanobody (GFPnano) in strains expressing different GFP/YFP-tagged proteins. The nanobody relocalises to sites of cytosolic GFP (H2B-GFP and OMP-GFP), but not to the luminal GFP of SOD2-GFP, as expected. Relocalisation of the nanobody in parasites expressing either GAPM1a-YFP or GAPM3-YFP indicates that the C-termini of these proteins are cytosolic. c Diagram of the auxin-induced degron (AID) system used to regulate GAPM1a expression. d Immunoblot showing rapid GAPM1a-AID depletion following treatment with IAA. GAPM1a-AID was visualised with an anti-Ty antibody, and anti-aldolase used as a loading control. Not treated (NT). e GAPM1a-AID depletion blocks plaque formation. Because of a reduction in plaque size for the untreated GAPM1a-AID strain, plaques were observed at 7- and 9-days post infection. Results are representative of three independent experiments. f Time course of morphological changes following the GAPM1a depletion after the addition of IAA. Parasites were stained with anti-GAP45 (cyan) and anti-IMC1 (magenta) antibodies and DNA visualised by Hoechst staining (blue). g Morphological changes were quantified in treated GAPM1A-AID parasites. After 4 h of IAA treatment, almost 50 % of vacuoles displayed morphological abnormalities, and by 18 h there were no normal vacuoles. Over 100 vacuoles were scored at each timepoint. Graph represents mean ± SD for n = 2 independent experiments. h TEM images of GAPM1a-AID parasites untreated or treated for 6 or 18 h with IAA. Boxes show detailed view of the IMC. Scale bars are 2 μm, and 500 nm within the detail. Unspecified scale bars are 5 μm. Source data are provided as a Source Data file

Fig. 3

Acute depletion of GAPM1a does not inhibit 2D gliding, invasion, or egress. a 2D gliding motility assayed by trail deposition, visualised with anti-SAG1 using the parental line or GAPM1a-AID parasites after treatment with IAA for 2 or 4 h. Graph represents mean ± SD for n = 3 independent experiments; ns p > 0.05, Student’s t test. Scale bar is 10 μm. b Depletion of GAPM1a for up to 4 h had no effect on invasion. All IAA treatments were performed on intracellular parasites, prior to mechanical release and incubation with fresh monolayers for 1 h. The percentage of parasites that were intracellular was quantified. Each point represents the average invasion from one well. Graph represents mean ± SD for n = 3 independent experiments; ns p > 0.05, Student’s t test. c Zaprinast-induced egress measured as the percent of host cells permeabilized over time. The number of DAPI-positive host-cell nuclei was calculated by automated image analysis. Results are the mean of three wells and representative of two independent experiments. d Video microscopy of zaprinast-induced egress; representative frames are shown. Parasites depleted of GAPM1a-AID overnight lysed the host cells, despite remaining immobile and unable to reinvade for the course of the experiment. Scale bars are 10 µm. Source data are provided as a Source Data file

Fig. 4

Depletion of GAPM1a results in cortical microtubule defects. a Colocalization of GAPM1a-AID (green) and cortical microtubules (mTagRFP-T_TubA1, magenta) by live SR-SIM following different treatments with IAA or vehicle (NT). Dotted lines indicate path of the profile plotted for the relative intensity of GAPM1a-AID (green) and mTagRFP_TubA1 (magenta). Scale bar is 5 μm. b GAPM1a-AID parasites treated for the indicated time with IAA were fixed and stained for acetylated tubulin (Tub-Ac, magenta) and GAP45 (cyan); GAPM1a-AID fluorescence is visible in green. Aberrant cortical microtubule arrangements (closed arrowhead) and depolymerisation (open arrowhead) are indicated. Daughter cell cortical arrays observed after 18 h of IAA treatment (asterisks) are highlighted. Scale bar is 5 μm. c Prevalence of cortical microtubule defects quantified over time from SR-SIM images. Graph represents mean ± SD for n = number of parasites indicated on each bar. d Acetylated microtubules (magenta) in extracellular parasites treated with IAA for 2 or 4 h prior to fixing and staining. Images are representative of two independent experiments. e Live SR-SIM of non-treated (NT) and IAA-treated GAPM1a-AID/ mTagRFP-T_TubA1 parasites. Gaps in the microtubule array are indicated (arrows). Scale bar is 5 μm. f Distance between adjacent microtubules in e. Each dot represents the distance measured between adjacent microtubules from at least 50 cells in three independent experiments. Graph represents mean ± SD for n = 137 (NT), 279 (2 h), 175 (4 h) measurements; p values from two-sample Kolmogorov-Smirnov (KS) test. g Coronal TEM sections showing the arrangement of cortical microtubules (arrowheads) following treatment with vehicle or IAA for 6 h. Scale bar is 500 nm. h Distance between cortical microtubules observed in g. Each dot represents a single distance measured as above from two independent experiments. Graph represents the mean ± SD for n = 124 (NT), 169 (6 h); p value from two-sample KS test. Source data are provided as a Source Data file

Fig. 5

Depletion of GAPM1a causes rapid changes in vacuolar organisation, cell morphology, and 3D motility. a Live video microscopy of intracellular parasites co-expressing GAPM1a-AID and mTagRFP-T_TubA1. Acute depletion of GAPM1a-AID led to the loss of the characteristic rosette organisation. Graph represents mean ± SD for n = 16 vacuoles per condition; **p < 0.005, ****p < 0.00005, Student’s t test with Holm–Sidak correction. b Length of individual, non-dividing parasites was measured every 14 min. Graph represents mean ± SD for n = 21 parasites per condition, * p < 0.05, ***p < 0.0005, Student’s t test with Holm–Sidak correction. Representative images shown for (a, b). Scale bars are 5 μm. c Comparison of parasite length following extracellular GAPM1a depletion or microtubule depolymerisation after incubating TKO parasites for 4 h at 4° C. Box plots for n = 561 (NT, 37 °C), 348 (NT, 4°C), 326 (4 h), 616 (TKO, 37 °C), 266 (TKO 4 °C) parasites per condition aggregated from three independent experiments. d Parasite volume estimated from over 8000 parasites per sample by Coulter counter. Asterisk indicates treatments performed before mechanical release of parasites from host cells. Box plots for n = 6 (37 °C, 4 h*), 7 (37 °C, NT*), 8 (37 °C, NT and 37 °C,TKO), 7 (4 °C, TKO) biological replicates; p values from two-tailed Student’s t test. e Buoyant mass of parasites was quantified using SMR. Box plots for n = 479 (NT), 509 (4 h), 442 (TKO 37^oC), 455 (TKO 4^oC) aggregated from three independent experiments; p values from two-tailed Student’s t test. f Maximum intensity projections of GAPM1a-AID parasite 3D motility either untreated or pre-treated for 4 h with IAA. Scale bar is 100 μm. g Mean displacement measured for at least 150 parasite per experiment. Graph shows mean ± SD for n = 5 biological replicates; p values from two-tailed Student’s t test. All box plots represent median and 25th and 75th percentiles and whiskers are at 10th and 90th percentiles. Source data are provided as a Source Data file

Fig. 6

The number of cortical microtubules is correlated to parasite size. Analysis of the number of cortical microtubules and surface area of various apicomplexan species and life cycle stages from the published literature. Values and references shown in Supplementary Table 1