Fig 1.
TbLC2 shows a high degree of structural and electrical potential conservation with the C. reinhardtii LC2 protein.
A. Alignment of the sequence-based structural homology model of TbLC2 (orange) with LC2 and LC9 (denoted as CrLC2, purple, and CrLC9, cyan, respectively) from the C. reinhardtii outer dynein arm core subcomplex structure (PDB ID: 7KZN [26]). R100 and R104 (green) in CrLC2, which are highly conserved polar residues, have extensive contacts (red dotted line) with CrLC9. The structures (top) of CrLC2 in B., the structural homology model of TbLC2 in C., and the template-free model of TbLC2 in D., are shown with surface charge density mapped to the electrostatic potential distribution (bottom). The homology model of TbLC2 in panels A and C (orange) was generated by SWISS MODEL and used and modified under the creative commons license CC BY-SA 4.0.
Fig 2.
Overexpressed TbLC2 localized to the axoneme as well as the cytoplasm.
A. Doxycycline induced TbLC2::eGFP overexpressing wild-type cells (WT/LC2 OE cell) visualized using differential interference contrast (DIC) microscopy (left) and under widefield fluorescence microscopy (right). The TbLC2::eGFP (green) localized to the cytoplasm and the cilium. The immunostained paraflagellar rod 2 (PFR2, red), DAPI stained DNA (DAPI, blue), and TbLC2::eGFP overlay (right), suggest that the TbLC2 has ciliary and cytoplasmic localization. B. SEM images showing cilium and cell body morphology of wild-type (WT, left), FLAM3-LC2 KD (middle), and FLAM3-LC2 KD/LC2 OE (right) cells showing short flagellar attachment zones (arrows) and nearly complete ciliary detachment from the cell body in the FLAM3 knocked down cells. The scale bar is 5 μm in each panel. C. LC2 KD (left), FLAM3 KD (center), and FLAM3-LC2 KD (right) cells visualized with DIC microscopy showing short flagellar attachment zones (arrows) and nearly complete ciliary detachment from the cell body. D. TbLC2::eGFP (green) expressing FLAM3-LC2 KD/LC2 OE cells stained with anti-ɑ-tubulin antibody (red) visualized with a confocal microscope. E. TbLC2::eGFP (green) expressing FLAM3-LC2 KD/LC2 OE cells stained with DAPI (blue) and the anti-PFR2 antibody (red) visualized under a widefield fluorescence microscope. The inset shows that TbLC2::eGFP and the paraflagellar rod are separately localized, and the line scan plots shows the normalized fluorescence intensity of TbLC2::eGFP (green) and PFR2 (red) along the line (white) shown in the inset.
Fig 3.
Stable assembly of outer arm dynein into the axoneme does not require TbLC2.
Transmission electron microscopy (TEM) images of wild-type (WT, left), FLAM3-LC2 KD (middle), and FLAM3-LC2 KD/LC2 OE (right) cells show the axoneme (canonical 9+2 microtubule arrangement) and paraflagellar rod (yellow arrowheads). The outer arm (red arrowheads) and inner arm (blue arrowheads) dyneins were intact in all three cell lines. The scale bar is 100 nm, and all micrographs have the same magnification.
Fig 4.
LC2 knocked down cells show growth and motility defects.
A. Growth curves of various RNAi knocked down cells and uninduced cells (green). At 72 hours post-induction (black arrow), we diluted the cells back to the starting cell density (1×106 cells/mL) and allowed them to grow for another 72 hours. We scaled the cell densities from the diluted culture to reflect the total growth. B. Sedimentation curves of various RNAi knocked down cells and uninduced cells (green). In both panels, each data point represents the mean from three independent experiments, and the error bars represent the standard error of the mean (SEM). The legend applies to both panels.
Fig 5.
Optically trapped LC2 KD and FLAM3-LC2 KD cells showed upregulated ciliary beat frequency.
A. Schematic of a T. brucei cell (gray) trapped by a low power (~20 mW) optical tweezer (red). B. Brightfield image of an uninduced FLAM3-LC2 KD cell taken near the coverslip, thus enabling direct visualization of the trap location and aligning the cell in the imaging plane. The identified trap location on a trypanosome cell (red arrowhead) is typical of all cells we trapped. Scale bar = 5 μm. C. Typical maximum projection (intensity inverted) of a movie (2 s) showing a trapped FLAM3 KD cell as it rotates about the trap center (red arrowhead). Scale bar = 5 μm. D. Typical power spectral densities (PSD) of a trapped cell with its cilium beating (FLAM3-LC2 KD/LC2 OE cell, orange), a trapped cell with its cilium not beating (FLAM3-LC2 KD/LC2 OE cell stalled at the bottom of the imaging chamber, purple), and background trap noise (gray, about six orders of magnitude smaller than the PSD of the cell with a beating cilium at the beat frequency). Peaks in the PSD represent the characteristic cell rotation rate (blue arrowhead) and ciliary beat frequency (red arrowhead). Each PSD represents the average of 3 spectra (1 second of data, each). E. Ciliary beat frequency, fω, of multiple cell lines obtained from the higher of the two characteristic frequencies from the PSD analysis. The error bars represent the SEM and *** represents p-values < 0.0001 and ns represents p-values > 0.05 from two-tailed paired t-tests. N = 25, 86, 75, 85, and 96 for uninduced, LC2 KD, FLAM3 KD, FLAM3-LC2 KD, and FLAM3-LC2 KD/LC2 OE cells, respectively.
Fig 6.
LC2 knockdown trypanosome cells have non-processive swimming behaviors.
A. Swimming trajectories of trypanosome cells plotted from an initial position at the origin. Each trajectory represents 10 seconds of tracked data. B. Magnitude of the average velocity (, Materials and methods) of swimming trypanosome cells. C. Mean average swimming speed (
, Materials and methods) measured for multiple cell lines over the entire trajectory of 10 s. Error bars represent the SE of the mean, *** corresponds to p-values < 0.0001, and ** corresponds to p-values < 0.003, two-tailed paired t-tests, in panels A. and B. D. The mean directional ratio (DR, dark lines) calculated over elapsed time and averaged for each strain. Error bars (lightly shaded bands) represent the SE of the mean. E. The percentage of cells that exhibit high persistence (DR > 0.2, solid color), medium persistence (0.05 < DR < 0.2, hatch mark), and low persistence (DR < 0.05, brick pattern), as determined by the last point (as time goes to infinity) directional ratio. We showed paths for and calculated values from N = 50, 75, 53, 73, 47, and 68 for uninduced, LC2 KD, FLAM3 KD, FLAM3-LC2 KD, FLAM3-LC2 KD/LC2 OE, and WT/LC2 OE cells, respectively, in all panels.
Fig 7.
Loss of TbLC2 alters multiple aspects of the ciliary beat waveform.
Representative plots of curvature, κ, normalized to the contour length (L) for A. FLAM3 KD and B. FLAM3-LC2 KD cells plotted as a function of s/L, which is the ratio of the arc length along the cilium (s) to the total ciliary length (L), where s/L = 0 is the base and s/L = 1 is the tip of the cilium. Colors (red to magenta) represent the time evolution of ciliary beat shape over one complete period of the oscillation. C. Schematic showing the radii of curvature corresponding to the maximum positive, r1, and the maximum negative, r2, curvature, where the curvature is 1/the radius of curvature, κ = 1/r, in this frame. We used these radii of curvature to calculate the asymmetry ratio of a ciliary bend during a ciliary beating. This image is a representative frame from a high-speed movie of a FLAM3-LC2 KD cell. Scale bar = 5 μm. D. Mean maximum curvature asymmetry ratio, ARMC, for FLAM3 KD, FLAM3-LC2 KD, and FLAM3-LC2 KD/LC2 OE cells. N = 15 waveforms from 10–12 different cells in each cell line, the error bars represent the SEM, and ** represents a p-value < 0.005, two-tailed paired t-tests. E. Fraction of time cells dwelled in tip-to-base (positive fraction) and base-to-tip (negative fraction) ciliary beating modes. N = 42, 44, and 33 cells for FLAM3 KD, FLAM3-LC2 KD, and FLAM3-LC2 KD/LC2 OE cells, respectively. ** represents a p-value of < 0.001 and ns a p-value > 0.05, two-tailed paired t-tests. F. Cumulative distribution function (CDF) of individual dwell times for various FLAM3 knockdown cells in base-to-tip (left) and tip-to-base (right) ciliary beating modes. The fits (dotted lines) to exponential functions are shown. Forty-five cells of each strain were analyzed with a total number of n = 59, 75, 65 base-to-tip mode and n = 70, 99, and 90 tip-to-base mode switching events for FLAM3 KD, FLAM3-LC2 KD, and FLAM3-LC2 KD/LC2 OE cells, respectively.
Table 1.
Characteristic beating mode dwell times.
Dwell time is the fit parameter ± the standard error of the fit, obtained by fitting exponential functions to the cumulative probability distributions of the dwell times (Fig 7F).