Cfap91-Dependent Stability of the RS2 and RS3 Base Proteins and Adjacent Inner Dynein Arms in Tetrahymena Cilia

Abstract

Motile cilia and eukaryotic flagella are specific cell protrusions that are conserved from protists to humans. They are supported by a skeleton composed of uniquely organized microtubules-nine peripheral doublets and two central singlets (9 × 2 + 2). Microtubules also serve as docking sites for periodically distributed multiprotein ciliary complexes. Radial spokes, the T-shaped ciliary complexes, repeat along the outer doublets as triplets and transduce the regulatory signals from the cilium center to the outer doublet-docked dynein arms. Using the genetic, proteomic, and microscopic approaches, we have shown that lack of Tetrahymena Cfap91 protein affects stable docking/positioning of the radial spoke RS3 and the base of RS2, and adjacent inner dynein arms, possibly due to the ability of Cfap91 to interact with a molecular ruler protein, Ccdc39. The localization studies confirmed that the level of RS3-specific proteins, Cfap61 and Cfap251, as well as RS2-associated Cfap206, are significantly diminished in Tetrahymena CFAP91-KO cells. Cilia of Tetrahymena cells with knocked-out CFAP91 beat in an uncoordinated manner and their beating frequency is dramatically reduced. Consequently, CFAP91-KO cells swam about a hundred times slower than wild-type cells. We concluded that Tetrahymena Cfap91 localizes at the base of radial spokes RS2 and RS3 and likely plays a role in the radial spoke(s) positioning and stability.

Keywords: CFAP91; Tetrahymena; axoneme; cilia; radial spoke.

Figure 1

Schematic representation of the motile cilium structure. (A) Cilium cross-section. Two central microtubules (grey), together with associated complexes, a connecting bridge and projections (yellow), form a central apparatus. Main outer doublets (grey) complexes: outer dynein arms (ODAs, greenish), inner dynein arms (IDAs, purple), radial spokes (RSs, dark grey), nexin-dynein regulatory complex (N-DRC, orange); (B) Fragment of the outer doublet showing organization of ciliary complexes within a 96-nm outer doublet axonemal unit [9,10]. ODA (7K5B, PDB DOI: 10.2210/pdb7K5B/pdb [11]) and RS (7JTK, PDB DOI: 10.2210/pdb7JTK/pdb [12]) structures were obtained from a Protein Data Bank (https://www.rcsb.org (accessed on 20 November 2022)). Other ciliary complexes were designed using Adobe Photoshop and Blender programs.

Figure 2

Cytokinesis defects in CFAP91-KO cells population. (A–C) Immunofluorescence images showing wild-type (A) and CFAP91-KO mutant (B,C) stained with a mix of anti-α-tubulin (12G10) and polyG antibodies to visualize the entire cilia length. (B) Note that CFAP91-KO cell (likely reverting from doublet stage) has a larger dimension than wild-type cell (A). (C) CFAP91-KO monster cell composed of several subcells formed due to defective cytokinesis.

Figure 3

Knockout of genes encoding radial spoke base or radial spoke docking proteins affect Tetrahymena swimming to a different extent. (A–G) Swimming paths of wild-type (A) and mutant cells: CFAP91-KO (B,C), CFAP91-KO rescued (D), CFAP61-KO (E), CFAP251-KO (F), and CFAP206 (G). All swimming paths were recorded for 3 s (except for (C) where cells were recorded for 60 s) at RT. Cells’ trajectories are indicated by parallel color lines. Bar = 500 μm. (H) Graph showing the comparison of the distances traveled by the wild type (WT) and analyzed mutants for 3 s. Error bars represent standard deviation. The traveled distance with standard deviation and number of measured trajectories (N) are indicated. Calculated p-value (t-test): WT/91-KO 5.67 × 10⁻²⁷², WT/91-res 8.6772 × 10⁻²⁶⁴, WT/61 8.6772 × 10⁻²⁶⁴, WT/251 1.265 × 10⁻²⁷⁷, and WT/206 4.9965 × 10⁻¹⁴⁷.

Figure 4

Lack of Cfap91 drastically reduces cilia beat frequency. (A) Drawings representing subsequent positions of the cilium during the power (red) and recovery stroke (green) of wild-type and CFAP91-KO mutant cells swimming in SPP medium, prepared based on recorded movies. (B) Drawings as in (A) but with indicated numbers of movie frames when the particular cilium position was reached. Cells were recorded in a mix of SPP medium and 1.5% methylcellulose. (C) Examples of kymographs of cilia motility recorded for 0.5 s (450 frames), analyzed using the ImageJ program, and used to calculate the cilia beating frequency. Ciliary beat cycles marked by stars *. (D) Graph showing measurements of cilia beating frequency in wild-type (WT) and CFAP91-KO mutant. Number of analyzed kymographs: WT (5 cells, 36 kymographs), CFAP91-KO (8 cells, 34 kymographs). Calculated p-value (t-test): WT/91-KO 5.29 × 10⁻³⁶ (indicated by four stars ****).

Figure 5

One of the radial spokes is lost in the subset of CFAP91-KO 96-nm ciliary units based on the classical transmission electron microscopy analyses. (A–E) Longitudinal sections of cilia of wild-type (A,B) and CFAP91-KO cells (C–E). A red rectangle box in (E) indicates a region where radial spokes are close to each other (no apparent triplets visible). (A1,A2,C1,C2,E1,E2) Magnified fragments of the axonemes are marked by white rectangle boxes in A, C, and E, respectively. (A2,C2,E2) Reverse contrast images of the images presented as (A1,C1,E1), respectively (A3,C3) Magnified fragments of the wild-type (A3) and mutant (C3) axonemes (marked by a red bracket in A1,C1, respectively) showing two adjacent radial spoke triplets, (F) A graph representing measurements of the distances between radial spokes (as indicated by color lines in A3,C3) in wild-type and CFAP91-KO mutant cilia.

Figure 6

Ciliary localization of Cfap61, Cfap206, and Cfap251 is strongly reduced in cells lacking Cfap91. Immunofluorescence confocal images showing localization of the HA-tagged fusion proteins expressed under the control of respective native promoters in cells with deletion of (A) CFAP61, (B) CFAP206, (C) CFAP251, or (D) CFAP91, and in (E) wild-type cells. Note that only the presence of Cfap91 is required for the ciliary localization of other studied Cfap proteins. (F) Schematic representation of the outer doublet microtubule (light grey) with three radial spokes (dark grey) showing confirmed (Cfap61, Cfap251) and putative (Cfap206 and Cfap91) localization of studied proteins. The proposed model explains why Cfap91 could be a key protein required for Cfap61, Cfap206, and Cfap251 stable docking. Ccdc39 and Ccdc40 (yellow and brown, respectively) are molecular ruler proteins, determining radial spokes position [49], Cfap206 (blue) is required for the stable assembly of RS2 [32], Cfap61 (green) and Cfap251 (red) form part of the RS3 stalk and arch-like structure at RS3 base, respectively [21], Cfap91 (purple) interacts with Ccdc39/Ccdc40 [12] and play a role in the stable assembly of RS2 and RS3-base proteins.

Figure 7

Real-time RT-PCR analyses of the levels of expression of CFAP61, CFAP91, and CFAP251. GCP4 was used as a housekeeping gene.

Figure 8

Proteins building the base of RS2 and RS3 are present in cilia in more than one isoform. (A) The Western blot analyses of ciliary proteins isolated from cells expressing C-terminally 3HA tagged Cfap61, Cfap91, Cfap206, or Cfap251 under the control of the respective native promoters. Stars indicate the position of the band of expected size. Weakly visible bands of lower molecular weight are products of protein degradation. (B) Two-dimensional analyses of axonemal proteins purified from cells expressing -3HA fusion proteins under the control of the respective native promoters. Isoelectric focusing was performed using 7 cm 4–7 or 3–10 ready-strips. The theoretical pI and Mw were calculated using https://web.expasy.org/compute_pi (accessed on 10 October 2022) (Table 3). Black arrows indicate detected isoforms. Whole blots are presented in Figure S4.

Figure 9

Cfap91-GFP interacts with HA-tagged molecular ruler protein, Ccdc39. GFP and Cfap91-GFP proteins were purified from the cytoplasm of GFP or Cfap91-GFP overexpressing cells using magnetic agarose beads-conjugated anti-GFP antibodies and after washings, incubated with a cytoplasmic extract obtained from cells overexpressing HA-Ccdc39 proteins (118 kDa). (A) A Western blot showing that either GFP (26 kDa, first line) or Cfap91-GFP (102 kDa, second line) proteins (both indicated by a red star) were bound to the magnetic beads. (B) A corresponding Western blot showing that HA-Ccdc39 was pulled-down by the GFP-Cfap91 protein but not by GFP alone.