|
Figure 1. Kif9 localization to motile cilia. (A–A″) Kif9–GFP construct localizes to Xenopus multiciliated cells in the axonemes. Kif9–GFP is in green (A), membrane–RFP is in magenta (A′). (B–B″) The Kif9–GFP construct localizes to basal bodies of multiciliated cells in Xenopus. Kif9–GFP in green (B), Centrin4–RFP in magenta (B′). (C–C″) Kif9 antibody (green) confirms localization to Xenopus axonemes. Kif9 in green (C) and acetylated tubulin (ActTub) in magenta (C′). (D–D″) Knockdown of Kif9 with 15 ng of MO reduces the level of Kif9 (green) in the cilium. (E) Quantification of Kif9 staining intensity over the intensity of acetylated tubulin in control and MO-injected frogs. (F–F″) Kif9 immunostaining in human airway multiciliated cells (MCCs). (F) Kif9 immunostaining (green). (F′) DNAI immunostaining (magenta). (F″) Merge of F–F′ with DAPI. Images are representative of 30 different cells imaged from three independent experiments with at least 10 different embryos imaged for each replicate. Scale bars: 10 µm.
|
|
Figure 2. Kif9 is necessary for ciliary beating and ciliary length. (A–C) Still images from bead flow movies subjected to Flowtrace analysis. (A) Control injected, (B) 15 ng Kif9 MO injected (KD), and (C) 15 ng MO with 600 pg Kif9 mRNA (Rescue). A is a still frame of Movie 1, B is a selected still frame from Movie 2, and C is a still frame from Movie 3. Note Movie 3 shows a different flow plane than Movies 1 and 2 hence the different direction of flow in Movie 3, but this was most representative of average flow rescues. Scale bars: 50 µm. (D) Graph of bead velocity averages from collected movies. **P=0.0094 and *P=0.0280 (one-way ANOVA with multiple comparisons and test for normality). (E–F″″) Still frames of ciliary beating in control (E–E″″) and Kif9 KD (F–F″″) multiciliated cells. Arrowheads point to stalled or paralyzed cilia frequently observed in Kif9 KD embryos. Scale bars: 10 µm. (G–I) Representative kymographs of ciliary beating in Control (G), Kif9 KD (H), and Rescue (I) injected embryos. The images represent 15 µm (height) and 10 s (width). (J) Graph of ciliary beat frequency calculated from kymographs of 27 different movies of cilia beating in Hz. (L–N) Representative images of ciliary length marked with membrane–GFP in control (L), Kif9 morpholino (M), and rescue (N) injected embryos. Note: the variability in cilia numbers in these images reflect the normally high variability seen in Xenopus multiciliated cells. Scale bars: 10 µm. (O) Graph of ciliary length from control, morpholino or rescue in µm. ****P<0.0001 and *P=0.0174 (one-way ANOVA with multiple comparisons and test for normality). Images are representative of 30 different cells imaged from three independent replicates with 10 embryos imaged for each replicate.
|
|
Figure 3. Kif9 contributes to central pair proteins and distal tip compartmentalization. (A) Schematic of 9+2 motile ciliary structure with central pair components schematized by tubule association. (B–C′) Confocal images of the localization of the radial spoke head component GFP–Rsph10B (orange) in control (B,B′) and Kif9 MO-injected (C,C′) embryos. (D,E) Magnified views of cilia from area indicated in B′,C′ for control (D) and (E) Kif9 KD cilia showing loss of GFP–Rsph10B (orange) from the distal ends of cilia. Dotted line outlines lack of protein at distal ends. (F) Quantification of GFP–Rsph10B length over total ciliary length in control and Kif9 KD embryos. (G–H′) Confocal images of GFP–Spag16 (orange) and membrane–RFP (gray) localization in Xenopus multiciliated cells in control (G,G′) and Kif9 MO (H,H′) injected embryos. (I,J) Magnified views of distal tips of cilia from area indicated in G′,H′ for control (I) and Kif9 KD (J) cilia showing loss of GFP–Spag16 (orange) from the distal tip of cilia. Dotted line outlining lack of protein at distal ends. (K) Quantification of GFP–Spag16 length over total ciliary length in control and Kif9 KD injected embryos. (L–M′) Confocal images of GFP–Cfap74 localization in control (L,L′) and Kif9 MO (M,M′) injected embryos. (N–O) Magnified views of distal tips of cilia from area indicated in L′,M′ for control (N) and Kif9 KD (O) cilia showing loss of GFP–Cfap74 (orange) from the distal tip of cilia. Dotted line outlining lack of protein at distal ends. (P) Quantification of GFP-Cfap74 length over total ciliary length in Control and Kif9 knockdown embryos. Results in F,K,P are shown as violin plots with the median and quartiles marked by dashed lines. ***P<0.001, ****P<0.0001 (unpaired two-tailed t-test). P, proximal; D, distal. Scale bars: 10 µm.
|
|
Figure 4. Kif9 is necessary for radial spoke, inner dynein arm and outer dynein arm placement in the distal ends of cilia. (A–B′) Confocal images of radial spoke stalk component GFP–Rsph23 (green) localization in control (A,A′) and Kif9 MO (B,B′) injected embryos. (C,D) Magnified views of cilia from area indicated in A′,B′ for control (C) and Kif9 KD (D) cilia showing loss of GFP–Rsph23 from the distal ends of cilia. Dotted line outlining lack of protein at distal ends. (E) Quantification of GFP–Rsph23 length over total ciliary length in control and Kif9 KD embryos. (F–G′) Confocal images of inner dynein arm Dnali1–GFP localization in control (F,F′) and Kif9 MO (G,G′) injected embryos. (H,I) Magnified views of cilia from area indicated in F′,G′ for control (H) and Kif9 KD (I) cilia showing loss of Dnali1–GFP from the distal ends of cilia. Dotted line outlining lack of protein at distal ends. (J) Quantification of GFP–Dnai1 length over total ciliary length in control and Kif9 KD embryos. (K–L′) Confocal images of outer dynein arm GFP–Dnai1 (cyan) localization in control (K,K′) and Kif9 MO (L,L′) injected embryos. (M,N) Magnified views of cilia from area indicated in K′,L′ for control (M) and Kif9 KD (N) cilia showing loss of GFP–Dnai1 (cyan) from the distal ends of cilia. Dotted line outlining lack of protein at distal ends. (O) Quantification of GFP–Dnai1 length over total ciliary length in control and Kif9 KD embryos. Results in E,J,O are shown as violin plots with the median and quartiles marked by dashed lines. **P<0.01, ****P<0.0001 (unpaired two-tailed t-test). P, proximal; D, distal. Scale bars: 10 µm.
|
|
Figure 5. Kif9 contributes to distal tip integrity. (A) Schematic of proximal to distal patterning of the motile cilium. Tip domain schematized with proteins known to localize to that domain, as well as proteins localizing to the rest of the axonemes. (B) Images of immunostaining of Spef1 in mouse tracheal multiciliated cells (MCCs). Spef1 in green, acetylated tubulin (ActTub) in red. (C,C′) Images of immunostaining of human airway multiciliated cells. Spef1 in red, acetylated tubulin in green. (D–E″) Confocal images of RFP–Spef1 (magenta) and Kif9–GFP (green) colocalization in the distal tips of the cilium. (F–H′) Confocal images of RFP–Spef1 (orange) and Membrane–GFP (gray) in control (F,F′), Kif9 MO-injected (G,G′), and rescue (H,H′) embryos. (I) Plot showing RFP–Spef1 fluorescence intensity as a function of axoneme length. Note the higher intensity for RFP–Spef1 at the distal end of control but not Kif9-KD axonemes. Note too that both axoneme length and distal Spef1 enrichment are partially rescued by re-expression of Kif9 (N=45 control axonemes; 45 Kif9-KD axonemes; 42 rescue axonemes; data compiled from 15 embryos across three independent experiments). Scale bars: 1 µm in B; 10 µm in C,C′,F–H′; and 2 µm in D–E″.
|
|
Figure 6. The microtubule plus-end-binding proteins Eb1 and Eb3 remain unaffected in the distal tip upon knockdown of Kif9. (A–B″) Confocal images of control (A–A″) and Kif9 MO-injected embryos (B–B″) with Eb3–GFP (green), RFP–Spef1 (magenta), and Membrane–BFP (blue). (A′,B′) Magnified views of area highlighted in A,B of ciliary tips with Eb3–GFP (green) and RFP–Spef1 (magenta) in control and KD cilia. (A″,B″) Magnified views of area highlighted in A,B of Eb3 alone in control and KD cilia. (C) Length measurements of Eb3 domain in control and Kif9 KD embryos. n.s., not significant (P=0.056) (unpaired two-tailed t-test). (D) Graph of line traces of fluorescence intensities of Eb3 in control and knockdown cilia. (E–F″) Confocal images of control (E–E″) and Kif9 MO-injected embryos (F–F″) with Eb1–GFP (green), RFP–Spef1 (magenta) and Membrane–BFP (blue). (E′,F′) Magnified views of area highlighted in E,F of ciliary tips with Eb1-GFP (green) an RFP–Spef1 (magenta) in Control and Knockdown cilia. (E″–F″) Magnified views of area highlighted in E,F of Eb1 alone in control and KD cilia. (G) Graph of line traces of fluorescence intensities of Eb1 in control and KD cilia. Images are representative of 30 different axonemes quantified across three independent replicates with 10 different embryos imaged for each replicate. Scale bars: 10 µm.
|
|
Figure 7. Motor activity of Xenopus Kif9. (A,A′) Schematic of Xenopus laevis Kif9 constructs made for single-molecule assays. (B–B″) Images from single-molecule microtubule binding assays for full-length Kif9 [Kif9 (FL), green) on Taxol-stabilized microtubules (magenta). (C–C″) Images from single-molecule microtubule-binding assays for a truncated version of Kif9 [Kif9 (1-461), green] on Taxol-stabilized microtubules (magenta). Scale bars: 10 µm. (D–D″) Still frames from live imaging of Kif9 processivity on microtubules in minutes. White arrow follows a Kif9 particle (green) moving along a microtubule (magenta). Scale bars: 2 μm. (E) Kymograph of live imaging of Kif9 (1-461)-mNG in vitro on microtubules. Scale bars: 2 min (vertical) and 5 µm (horizontal). (F) Kymograph of live imaging of Kif9–GFP construct in vivo in Xenopus laevis multiciliated cell axonemes. Images are representative of 30 different cells imaged across three independent replicates with at least 10 experiments imaged for each replicate. Scale bar: 2 μm
|
|
Figure 8. Summary of effect of Kif9 knockdown on proximal to distal patterning of the motile cilium. (A) Schematic of wild-type motile cilia in Xenopus with domains schematized based off findings presented here. (B) Schematic of Kif9 KD cilia and changes to domains observed.
|
|
Fig. S1. A. Gel image of RT-PCR of Kif9 expression in Kif9 morpholino injected embryos at wildtype, 10 ng, 15 ng, and 20 ng doses. B. Graphs of average bead flow across the epidermis of stage 24 Xenopus embryos in Control, Kif9 morpholino, and Rescue injected embryos.
|
|
Fig. S2. A. Raw quantification of C2 bridge central pair component GFP-Spag16 length in Control and Morpholino injected embryos. B. Graphs of fluorescence intensity of GFP-Spag16 across the length of the cilia, normalized to 1. C. Raw quantification of C1 central pair component GFP-Cfap74 length in Control and Morpholino injected embryos. D. Graphs of fluorescence intensity of GFP-Cfap74 across the length of the cilia, normalized to 1. E. Raw quantification of GFP-Rsph10B length in Control and Morpholino injected embryos. F. Graphs of fluorescence intensity of GFP-Rsph10B across the length of the cilia, normalized to 1. G. Raw quantification of radial spoke stalk component Rsph23 length in Control and Morpholino injected embryos. H. Graphs of fluorescence intensity of GFP-Rsph23 across the length of the cilia, normalized to 1. I. Raw quantification of inner dynein arm GFP-Dnali1 length in Control and Morpholino injected embryos. J. Graphs of fluorescence intensity of GFP-Dnali1 across the length of the cilia, normalized to 1. K. Raw quantification of outer dynein arm GFP-Dnai1 length in Control and Morpholino injected embryos. L. Graphs of fluorescence intensity of GFP-Dnai1 across the length of the cilia, normalized to 1.
|
|
Fig. S3. A. Histogram of speed of truncated Kif9 movement along microtubules in vitro. Values binned to 1 nanometer/ second. X-axis = velocity of molecules in nm/sec. Y-value = number of molecules at given speed. B. Histogram of run length of truncated Kif9 (X-axis) movement and number of molecules tracked (Y-axis). C. Dwell time of paused molecules (X-axis) and number of molecules tracked (Y-axis). Kymograph of live imaging of Kif9-GFP in axonemes. C. Kymograph of live imaging of Kif9-GFP in axonemes, arrows pointing to bidirectional trafficking of Kif9-GFP along the axoneme.
|