Fig 1.
Schematics of the cytoskeleton and cell division in Trypanosoma brucei.
(A) The subpellicular array is a single layer of microtubules (purple) that underlies the plasma membrane. The microtubules are crosslinked to each other by regularly spaced fibrils (green). The growing plus-ends of array microtubules face the cell posterior. (B) Array microtubules appear to consistently polymerize at the posterior end of the cell (purple shading). As the new flagellum is nucleated, new microtubules are inserted between the old and new flagellum to increase cell width. A membrane inflection forms that demarcates the path of cleavage furrow ingression, which follows the helical path of array microtubules. Cleavage furrow ingression initiates at the cell anterior, and nascent posterior end formation remodels the array at the cell midzone to resolve the subpellicular array into two distinct structures to complete cytokinesis.
Fig 2.
PAVE1 localizes to the subpellicular array microtubules of the cell posterior.
(A) Ty1-PAVE1 cells were fixed in methanol and labeled with anti-Ty1 antibody and YL1/2, which recognizes tyrosinated tubulin. The empty arrowheads are denoting pools of tyrosinated α-tubulin at the basal body. Asterisks are marking kinetoplasts. (B) (Left) A subpellicular array sheet from WT 427 control cells immunogold-labeled with 10 nm beads using TAT1 antibody, which recognizes trypanosome α-tubulin. (Right) A subpellicular array sheet immunogold-labeled for Ty1-PAVE1 with 10 nm beads. Empty arrowhead is pointing to the ventral protrusion of a disrupted crosslink, and solid arrows are marking Ty1-PAVE1 beads on preserved inter-microtubule crosslinking fibrils. (C) Quantification of immunogold label distribution along individual microtubules in subpellicular sheets represented as a Superplot [89], which shows each individual measurement in each biological replicate. The replicates are color-coded and the average is illustrated as large circles. TAT1 α-tubulin—Ty1-PAVE1 dorsal:dorsal and centroid:centroid distributions are not significantly different (P = 0.0542 and P = 0.0741, respectively). *P = 0.0146, **P = 0.0032. P-values are calculated from unpaired two-tailed Student’s t-tests using the averages from N = 3 independent biological replicates. 3 subpellicular sheets in both TAT1 control and Ty1-PAVE1 conditions from N = 3 independent biological replicates were quantified. (D) Heat map of α-tubulin and Ty1-PAVE1 immunogold label distribution on individual microtubules. Shading is representative of the average distribution of labeling from (C). D; dorsal, C; centroid, V; ventral.
Fig 3.
PAVE1 is added to and stabilizes the microtubules at the extreme posterior end of the array.
(A) HaloTag-PAVE1 cells were fixed in paraformaldehyde at T = 0 h, T = 1 h, T = 3 h, and T = 8 h after the addition of TMR-conjugated HaloTag ligand to cell culture in the capping assay. Asterisk represents background labeling of the endomembrane system. Empty arrowhead is pointing to the earliest addition of HaloTag-PAVE1, represented by TMR labeling, at the extreme cell posterior. (B) Fluorescence intensity quantification of HaloTag ligand labeling along the ventral edge of the cell body at T = 0 h, T = 1 h, T = 3 h, and T = 8 h after TMR substrate addition. Fluorescence intensity was normalized to 1 at each time point in order to illustrate the shape of the curve. 50 1N1K cells at each time point from N = 3 independent biological replicates were measured. (C) Fluorescence intensity quantification of JF646-HaloTag ligand labeling as in (B). Inset displays the distance in μm at which the maximum intensity of JF646 ligand was located. 50 1N1K cells at each time point from N = 3 independent biological replicates were measured. One-way ANOVA was calculated using the averages from N = 3 independent biological replicates. ***P<0.001. (D) PAVE1 RNAi was induced and cells were harvested, fixed in methanol, and stained with anti-α-tubulin after T = 0 h, T = 4 h, and T = 6 h of PAVE1 RNAi induction. 1N2K cells were selected for analysis as they completed cytokinesis prior to PAVE1 depletion. (E) Superplot of the measured distances between the posterior kinetoplast to outer posterior edge of the subpellicular array, and the posterior edge of the nucleus to the posterior kinetoplast in control and PAVE1 RNAi cells. 50 1N2K cells were measured at each time point from N = 3 independent biological replicates. One-way ANOVAs were calculated using the averages from each replicate. ****P<0.0001, n.s. P = 0.5653.
Fig 4.
Immunoprecipitation of mNeonGreen-PAVE1 reveals two interacting partners.
(A) Representative image of a live T. brucei cells expressing mNeonGreen-PAVE1 from both endogenous loci. (B) dET-mNeonGreen-PAVE1 cells were harvested, lysed, and sonicated to solubilize cytoskeletal proteins. Clarified supernatant was batch-bound with magnetic camelid nanobodies against mNeonGreen. Beads were washed 6X and bound proteins were analyzed by mass spectrometry. Gel is a representative silver-stained SDS-PAGE gel of the elution from the mNeonGreen immunoprecipitation in control and mNeonGreen-PAVE1 samples. Both lanes are 50% of input material. Arrows indicate the two major unique protein bands present in the mNeonGreen-PAVE1 sample not present in WT 427 controls. (C) The two potential interacting partners of PAVE1 identified in the mNeonGreen-PAVE1 immunoprecipitation were endogenously tagged with the HA epitope tag. The cytoskeleton was extracted and cells were processed for immunofluorescence to co-localize each protein with Ty1-PAVE1 using anti-Ty1 and anti-HA antibody. (D) Protein domain schematics including the predicted molecular weight (kDa) and pI for PAVE1, PAVE2, and TbAIR9. PAVE1 and PAVE2 have no recognizable domain homology to any known protein. However, they contain a series of predicted coiled coils [58]. TbAIR9 lacks the basic serine-rich microtubule binding site found in Arabidopsis thaliana AIR9, but does contain the conserved leucine-rich repeats, glycogen binding domain, and A9 IgG-like folds. Inset illustrates the localization pattern of PAVE1, PAVE2, and TbAIR9 across the array.
Fig 5.
(A) PAVE2 RNAi was induced for 2 d, after which control and PAVE2-depleted cells were fixed with paraformaldehyde and examined by epifluorescent microscopy. (B) Quantification of DNA states in control and PAVE2 RNAi cells after 1, 2, and 4 d of RNAi induction. Bars represent the mean and the error bars are the S.D. 300 cells in both control and PAVE2 RNAi conditions from N = 3 independent biological replicates were counted at each time point. One-way ANOVAs were calculated using the averages from each replicate. ****P<0.0001, ***P<0.0005, **P<0.005, *P<0.05, n.s. P>0.05. (C) Whole-mount transmission electron microscopy of control and PAVE2 RNAi cells after 1 d of PAVE2 depletion. (D) PAVE2 RNAi was induced for 1 d and cells were harvested, fixed in paraformaldehyde, and stained with anti-α-tubulin to demarcate the subpellicular array. Superplot is of the measured distance between the kinetoplast and the outer posterior edge of the array, and the posterior edge of the nucleus to the kinetoplast of 1N1K cells in control and PAVE2 RNAi conditions. 100 1N1K cells were measured for both control and PAVE2 RNAi conditions in N = 3 independent biological replicates. Unpaired two-tailed Student’s t-tests were performed using the averages from N = 3 independent biological replicates. ***P = 0.0006. n.s. P = 0.0753. (E) Control and PAVE1 RNAi cells were fixed in paraformaldehyde and stained after 1 d of PAVE1 depletion. PAVE2 does not localize to the array in the absence of PAVE1. (F) Control and PAVE2 RNAi cells were fixed in paraformaldehyde and stained after 1 d of PAVE2 depletion. PAVE1 does not localize to the array in the absence of PAVE2.
Fig 6.
PAVE1 and PAVE2 form a complex in vitro that binds the microtubule lattice.
(A) (Left) Coomassie-stained SDS-PAGE gel demonstrating the co-purification of mNeonGreen-PAVE1 and Strep-PAVE2 from E. coli cells. E. coli cells co-expressing [MBP-oligoHis10-TEV]-mNeonGreen-PAVE1 and Strep-PAVE2 constructs were harvested, lysed, and the clarified supernatant was batch-bound with Ni-NTA resin. The eluate was treated with TEV protease and allowed to interact with Strep-Tactin XT resin. Proteins were eluted using 50 mM biotin. The Strep eluate lane contains 4 μg of protein. (Right) The PAVE complex was diluted into a low salt buffer containing 50 mM NaCl and ultracentrifuged. Equal fractions of both the supernatant and pellet were separated on an SDS-PAGE gel and stained with Coomassie blue. (B) 5, 50, 100, 200, and 500 nM clarified PAVE complex was incubated with Taxol-stabilized bovine microtubules labeled with the Cy5 fluor for 20 min RT. The PAVE complex-microtubule solution was settled onto a blocked flow cell and unbound protein was washed away. The flow cell was imaged using epifluorescence microscopy. (C) (Left) A PEGylated flow cell was blocked and Taxol-stabilized bovine Cy5-microtubules were allowed to attach to the flow cell surface. 5 nM clarified PAVE complex was added to the flow cell chamber and immediately imaged using TIRF microscopy. Images were taken every 10 sec for 20 min. (Middle) Kymograph of PAVE complex on the microtubule outlined in yellow. (Right) Fluorescence intensity line scan of the microtubule outlined in yellow for the first 4 min 10 sec of TIRF imaging. (D) (Left) 5 nM clarified PAVE complex was mixed with Taxol-stabilized bovine Cy5-microtubules for 10 min RT to pre-seed patches. The PAVE complex-microtubule mixture was flowed into a blocked and PEGylated flow cell and allowed to attach to the flow cell surface. Unbound protein was washed away and the flow cell was immediately imaged using TIRF microscopy. Images were taken every 10 sec for 20 min. (Middle) Kymograph of PAVE complex on the microtubule outlined in yellow. (Right) Fluorescence intensity line scan of the microtubule outlined in yellow for the first 4 min 10 sec of TIRF imaging. (E) Control microtubules and microtubules treated with subtilisin A were collected for western blotting. The post-translational modification of glutamylation is no longer present on subtilisin-treated microtubules, as detected with the antibody GT335, and there is a molecular weight shift showing that the C-terminal tails have been cleaved, as detected with anti-α-tubulin. (F) PAVE complex was incubated with control or subtilisin-treated Taxol-stabilized bovine Cy5-microtubules as in (B) and imaged using epifluorescence. There is no difference in PAVE complex binding to subtilisin-treated microtubules cleaved of their C-terminal tails in comparison to control, indicating that the PAVE complex binds directly to the microtubule lattice.
Fig 7.
TbAIR9 maintains PAVE1 at the cell posterior.
(A) Control and TbAIR9 RNAi cells were fixed in paraformaldehyde after 2 d of TbAIR9 depletion. Images are representative morphologies in control and TbAIR9 RNAi conditions. The epimastigote-like morphotype and elongated cell bodies were common in cells depleted of TbAIR9, as previously reported [35]. Open arrowheads indicate the kinetoplast. (B) Control and TbAIR9 RNAi cells were fixed in paraformaldehyde and stained after 2 d of TbAIR9 depletion. Ty1-PAVE1 signal is no longer maintained at the posterior and ventral edge. (C) The fluorescence signal intensity of Ty1-PAVE1 was measured along the ventral side of control and TbAIR9 RNAi cells after 2 d of TbAIR9 depletion. The signal intensities were normalized to the maximum control intensity, averaged, and plotted with the S.D. 50 1N1K cells in both control and TbAIR9 RNAi conditions from N = 3 independent biological replicates were measured. Ty1-PAVE1 signal is more evenly distributed throughout the array in TbAIR9 depleted cells than in control cells. (D) TbAIR9 RNAi was induced and lysates were collected every 48 h from control and TbAIR9 depleted cells. Lysates were separated by SDS-PAGE and transferred to nitrocellulose for western blotting. Ty1-PAVE1 signal remains stable during TbAIR9 depletion. (E) Subpellicular array sheets of control and TbAIR9 RNAi cells after 2 d of TbAIR9 depletion immunogold labeled against Ty1-PAVE1 with 10 nm beads. Ty1-PAVE1 localizes to the subpellicular microtubules at both the posterior and anterior ends of the array in the absence of TbAIR9. Open arrowheads indicate Ty1-PAVE1 beads on an inter-microtubule crosslink.
Fig 8.
TbAIR9 maintains the localization array-associated proteins at specific subdomains.
(A, B) Identification of array-associated proteins from TrypTag that localize to the middle subdomain (A, Tb927.9.10790) and the anterior subdomain (B, Tb927.11.1840). The schematics illustrate the predicted molecular weight (kDa) and pI for each protein. Tb927.9.10790 does not share homology with any known proteins, but it is predicted to be highly unstructured. Tb927.11.1840 also does not share homology with any known proteins, but it contains predicted coiled coils. Tb927.9.10790 and Tb927.11.1840 were tagged at their C-termini with Ty1 and a cytoskeletal extraction followed by immunofluorescence with anti-Ty1 antibody was performed. Tb927.9.10790 and Tb927.11.1840 remained bound to the middle and anterior subdomains of the following cytoskeletal extractions. (C,D) Cell lines containing 10790-Ty1 (C) and 1840-Ty1 (D) endogenous tags were fixed in paraformaldehyde and stained after 2 d of TbAIR9 depletion. 10790-Ty1 signal is depleted and spread throughout the cell body in the absence of TbAIR9, while 1840-Ty1 signal is detected at the posterior and anterior ends in TbAIR9 depleted cells. (E,F) Quantification of 10790-Ty1 (E) and 1840-Ty1 (F) fluorescent signal intensity along the ventral side of the cell body in TbAIR9 RNAi cells after 2 d of TbAIR9 depletion. The signal intensities were normalized to the maximum control values and plotted with the S.D. 50 1N1K cells in both control and TbAIR9 RNAi conditions from N = 3 independent biological replicates were measured.
Fig 9.
Model of subpellicular array subdomains and potential MAP localizations in T. brucei.
(A) Three primary subdomains of subpellicular array regulation were identified in this work, represented by PAVE1/2, Tb927.9.10790, and Tb927.11.1840. TbAIR9, which localizes to the entire subpellicular array, is required to maintain the integrity of the subdomains. (B) A model for the potential MAP classifications and localizations in T. brucei as suggested by this work. MAP; microtubule associated protein. MT; microtubule. PM; plasma membrane.