Asymmetrically localized proteins stabilize basal bodies against ciliary beating forces

Abstract

Basal bodies are radially symmetric, microtubule-rich structures that nucleate and anchor motile cilia. Ciliary beating produces asymmetric mechanical forces that are resisted by basal bodies. To resist these forces, distinct regions within the basal body ultrastructure and the microtubules themselves must be stable. However, the molecular components that stabilize basal bodies remain poorly defined. Here, we determine that Fop1 functionally interacts with the established basal body stability components Bld10 and Poc1. We find that Fop1 and microtubule glutamylation incorporate into basal bodies at distinct stages of assembly, culminating in their asymmetric enrichment at specific triplet microtubule regions that are predicted to experience the greatest mechanical force from ciliary beating. Both Fop1 and microtubule glutamylation are required to stabilize basal bodies against ciliary beating forces. Our studies reveal that microtubule glutamylation and Bld10, Poc1, and Fop1 stabilize basal bodies against the forces produced by ciliary beating via distinct yet interdependent mechanisms.

Figure 1.

Fop1 is a BB stability protein. (A) Whole-cell BB immunofluorescence (α-centrin). FOP1 knockdown (KD) cells have fewer BBs compared with wild type (WT). The cell’s anterior is defined by the oral apparatus (arrow). Bar, 10 µm. (B) Schematic of a 10-µm segment of BBs along a ciliary row that represents the region used for BB frequency analysis. Bar, 1 µm. (C, top) FOP1 KD cells have fewer BBs than WT cells. BB loss is rescued by the reintroduction of FOP1. Representative fluorescence images of BB rows in WT, FOP1 KD, and FOP1 KD with FOP1 rescue at 30°C and 37°C in cell cycle arrested cells. (bottom) Quantification of BBs per 10 µm. Mean ± SEM; n = 300 rows; *, P < 0.01. (D and E) FOP1 KD causes constricted BBs with defective C tubules (arrows). Electron micrographs of longitudinal (D) and cross-sectional (E) views of FOP1 KD BBs at 37°C. n = 50 BBs for each orientation. Bars, 100 nm.

Figure 2.

Fop1 stabilizes BBs to resist ciliary beating. (A) BB loss in FOP1 KD cells is elevated at increasing temperatures. Representative BB rows (α-centrin) in WT and FOP1 KD cells at 25°C, 30°C, and 37°C. (B) Increased media viscosity using PEO exacerbates the BB loss phenotype in FOP1 KD cells. Representative BB rows in WT and FOP1 KD cells in untreated or PEO media. (C) Inhibition of ciliary beating using NiCl₂ prevents high-temperature–induced BB loss in FOP1 KD cells. Representative BB rows in WT and FOP1 KD cells in untreated or NiCl₂-treated media. Mean ± SEM; n = 200 rows. *, P < 0.01. Bar, 1 µm.

Figure 3.

Fop1 localizes asymmetrically at BBs. (A) Fop1 is enriched at the posterior side of BBs. Inset of Sas6a-labeled (green; Sas6a:GFP) and Fop1-labeled (red; Fop1:mCherry) BBs. Bar, 1 µm. (B) Schematic showing a representative, polarized BB. KF, kinetodesmal fiber; PC, postciliary microtubules. (C) Fop1 is enriched at the posterior side of BBs. (left) Averaged two-color, colocalization of Fop1 (red) relative to centroid localizing, Sas6a (green). (middle) Linescan from a Gaussian fit curve to the mean localization. (right) Schematic representation of Fop1’s BB localization. n = 45 BBs. Bar, 0.5 µm. (D, left) SIM image of Fop1’s enrichment at the posterior (triplet microtubules 8 and 9) and anterior (triplet microtubules 4 and 5) triplet microtubules with an enrichment for posterior microtubules. Averaged SIM image of Fop1-labeled (red; Fop1:mCherry) BBs. n = 328 BBs. Bar, 0.25 µm. (middle) Heatmap illustrating intensity differences. (right) Schematic representation of Fop1 localization. (E) Representative SIM images showing dynamic Fop1 localization at individual BBs. Bar, 0.25 µm. (F, left) Interpolated intensity map of averaged Fop1 SIM images. Linear (middle) and radial (left) linescans. Arrow a in left panel corresponds to the linear linescan. Arrow b corresponds to the radial linescan. Asterisks denote levels distinct from posterior facing peak (*, P < 0.01). (G) Fop1:GFP immuno-EM localization in BB cross-sectional view. (right) Schematic of red dots denote Fop1’s localization based on the relative distribution of 25 gold particles to represent the total quantified gold label. Numbers denote the region where Fop1 was measured. n = 125 gold particles in 21 BBs. Bar, 100 nm.

Figure 4.

Poc1 promotes normal Fop1 protein incorporation into BBs. (A) Poc1 and Fop1 are not required for each other’s localization to BBs. Poc1:mCherry or Fop1:mCherry BB levels do not change in response to knockdown or complete loss of the reciprocal protein, respectively. (B) Fop1 BB levels increase upon Poc1 overexpression (OE). Sas6a overexpression is a negative control. (C) Poc1 BB levels do not change upon Fop1 overexpression. (A–C) Bars, 0.5 µm. n = 150 BBs. (D) Representative Fop1:mCherry image denoting new BB assembly. Quantification of protein incorporation into new BBs is measured using the background-subtracted intensity levels of daughter BBs divided by mother BBs. This measurement is acquired relative to the separation distance of mother and daughter BBs. Bars: 10 µm; (inset) 0.5 µm. (E) The rate of Fop1 incorporation at new BBs decreases in poc1Δ cells. (left) Fop1:mCherry levels relative to BB separation. (right) Quantification of Fop1:mCherry protein incorporation relative to mother-daughter BB separation distance. (F) Poc1:mCherry incorporation into BBs does not change in Fop1 KD cells. (left) Poc1:mCherry levels relative to BB separation. (right) Quantification of Poc1:mCherry protein incorporation relative to mother–daughter BB separation distance. (E and F) Dashed line represents mean distance upon which ciliogenesis occurs (2.6 ± 0.1 µm). Bars, 1 µm. Mean ± SEM; *, P < 0.01.

Figure 5.

Microtubule glutamylation stabilizes BBs differentially from BB stability proteins. (A) SIM imaging reveals that glutamylation is enriched at the posterior and anterior triplet microtubules. (left) Averaged SIM image of BB glutamylation (GT335). n = 110 BBs. (middle) Heatmap representation. (right) Schematic localization of BB glutamylation. (B) Representative SIM images showing variable glutamylation in foci and horseshoe patterns at individual BBs. (C, left) Interpolated intensity heatmap of the mean SIM image reveals enrichment of glutamylation at the BB posterior and anterior faces. Quantification of a linear (middle, a.) and radial linescans (left, b.). (A–C) Bars, 0.25 µm. (D) The timing and rate of incorporation of BB glutamylation and Bld10 is similar to Fop1. Glutamylation levels depict the incorporation of glutamylation relative to BB separation. (right) Quantification of BB stability factor and microtubule glutamylation incorporation relative to the mother–daughter BB separation. Bar, 1 µm. (E) BB microtubule glutamylation is significantly reduced in ttll1,9Δ cells. BBs and cilia stained for glutamylated microtubules (red) and BBs for centrin (green). Bar, 0.5 µm. (F) Quantification of BB glutamylation levels in WT and ttll1,9Δ cells. n = 150 BBs. (G, top) Loss of BB microtubule glutamylation causes BB loss. Representative fluorescence images of BB rows in WT and ttll1,9Δ cells at 30°C and 37°C in cell cycle–arrested conditions. (bottom) BB frequency in WT and ttll1,9Δ cells. n = 300 rows. Bar, 1 µm. (H) Inhibition of ciliary beating prevents temperature-induced BB loss in ttll1,9Δ cells. Representative BB rows in WT and ttll1,9Δ cells in untreated or NiCl₂-treated media. n = 300 rows. (I) Loss of Poc1 or Fop1 increases BB glutamylation levels. Bar, 0.5 µm. n = 150 BBs. (J) Loss of both BB glutamylation and Poc1 increases BB disassembly. Quantification of BB frequency in WT and ttll1,9Δ, poc1Δ cells at 30°C and 37°C. n = 200 BBs. (A–J) Mean ± SEM; *, P < 0.01.