Diverse type of cilia are present in zebrafish.

(A) Cilia in kupffer’s vesicle (KV) of a 10-somite stage zebrafish larvae. (B–H) Confocal images showing cilia in different type of cells at 4 dpf as indicated. The position of these cells were indicated in the top diagram. (I–J) Confocal images showing cilia in the sperm (flagellum) or spermatocyte of adult zebrafish. All the cilia were visualized with anti-acetyleated tubulin. The photoreceptor double cones were stained with zpr-1 antibody in panel H, and nuclei were stained with DAPI in panel I. Immunostaining with anti-SYCP3 labelled the synaptonemal complexes of primary spermatocytes in panel J. Scale bar = 10 μm.

Rescue of ovl mutants with Tg(hsp70l:ift88-GFP) transgene. (A) Schematic diagram of hsp70l:ift88-GFP construct. (B) Procedure of heat shock experiments for ovl mutants rescue assay. hphs, hour post heat shock. (C) External pheontype of 5 dpf wild type, ovl mutant or ovl mutant larvae carrying Tg(hsp70l:ift88-GFP) transgene. The numbers of larvae investigated were shown on the bottom right. (D) Confocal images showing cilia in different type of organs as indicated. Red channel indicates cilia visualized by anti-acetylated ?-tubulin antibody and fluorescence of Tg(hsp70l:ift88-GFP) is showed in green. Scale bars: 200 ?m in panel C and 10 ?m in panel D.

Rescue of ciliogenesis defects in ovl (ift88) mutants via Tg(hsp70l:ift88-GFP).

(A) Rescue of ciliogenesis defects in ear macula, olfactory placode and pronephric duct. Red channel shows cilia visualized by anti-acetylated α-tubulin antibody. Fluorescence of Tg(hsp70l:ift88-GFP) is shown in green. (B) External phenotypes of adult ovl homozygotes rescued with Tg(hsp70l:ift88-GFP) transgene. Scale bars:10 μm in panel A and 1 cm in panel B.

Intraflagellar transport in different type of cilia. (A?E) Left, Snapshot of Intraflagellar transport videos in different cell types as indicated. Middle (A, B) Snapshot of same cilia at different time points. Arrowheads with the same color indicate the same IFT particle. Right, kymographs illustrating the movement of IFT particles along the axoneme. Horizontal scale bar: 10 ?m and vertical scale bars: 10 s. Representative particle traces are marked with black lines in panels D and E. Yellow arrows denote the cilia used to generating the kymograph. (F) Histograms displaying the velocity of anterograde and retrograde IFT in different type of cilia as indicated. ?Antero.? and ?Retro.? represent anterograde and retrograde transport, respectively. Each plot was fit by a Gaussian distribution. The developmental stages of zebrafish were consistent with (A?E) (G) Summary of IFT velocities in different tissues of zebrafish. Numbers of IFT particles are shown in the brackets. n, number of cilia detected. N, number of zerafish larvae analyzed. hpf, hours post-fertilization. dpf, days post-fertilization. (H) Left, Statistics analysis of cilia length in different tissues of Tg(hsp70l:ift88-GFP) larvae (crista, n=72 cilia from 16 larvae; neuromast, n=31 cilia from 9 larvae; pronephric duct, n=72 cilia from 11 larvae; spinal cord, n=86 cilia from 6 larvae; epidermal cell, n=48 cilia from 18 larvae). Average cilium length: crista, 25.46 ?m; neuromast, 15.2 ?m; pronephric duct, 8.1 ?m; spinal cord, 1.03 ?m; epidermal cell, 0.49 ?m. Right, anterograde and retrograde IFT average velocity in different tissues of zebrafish. (I) Anterograde and retrograde IFT velocities plotted versus cilia length. Linear fit (black line) and coefficient of determination are indicated. (J) Frequency of anterograde and retrograde IFT entering or exiting cilia. (crista, n=47 cilia from 13 larvae; neuromast, n=31 cilia from 13 larvae; pronephric duct, n=43 cilia from 14 larvae; spinal cord, n=27 cilia from 7 larvae; epidermal cell, n=26 cilia from 15 larvae).

Overview of zebrafish larvae treatments and IFT velocity analysis.

(A) Diagram showing the procedure of heat shock treatments of transgenic zebrafish larvae (see Materials and methods for details). (B) Representative kymographs illustrating the movement of IFT particles along the axoneme using automated and manual marking methods. Left, lines represent trajectories fitted by KymographDirect. Right, particle traces are manually marked with lines. (C) Quantification of anterograde and retrograde average velocities in 4dpf crista cilia using two speed calculation methods. (D) Summary of anterograde and retrograde average velocities in 4dpf crista cilia using two speed calculation methods. Numbers of IFT particles are shown in the brackets. n, number of cilia. N, number of zerafish larvae. (E) Quantification of anterograde and retrograde velocities along the entire length of crista cilia. n=11 cilia from 15 larvae.

Generation of Tg (βactin:tdTomato-ift43) transgene for IFT imaging.

(A) Schematic diagram showing βactin:tdTomato-ift43 transgenic construct. (B–C) Left, Snapshot of Intraflagellar transport videos in cilia of ear crista and pronephric duct epithelial cells. Right, kymographs illustrating the movement of IFT particles along the axoneme. Horizontal scale bar = 10 μm. Vertical scale bar = 10 s. (D) Summary of the anterograde and retrograde velocities of IFT measured in embryos expressing Tg(βactin:tdTomato-ift43). Numbers of IFT particles are shown in the brackets. n, number of cilia. N, number of zerafish larvae.

IFT velocity in cilia of different lengths within crista.

(A) Histograms displaying the velocity of anterograde and retrograde IFT in crista cilia of different lengths as indicated. (B and C) Dot graphs showing the velocity of anterograde and retrograde IFT in crista cilia of different lengths. Statistical significance is based on one-way ANOVA, Bonferroni’s multiple comparisons test. n.s., not significant, p＞0.05. (D) Summary of the anterograde and retrograde velocities in crista cilia of different lengths. Numbers of IFT particles are shown in the brackets. n=46 crista cilia from 14 4dpf larvae. ＜10 μm, n=12; 10–18 μm, n=22, ＞18 μm, n=12.

Comparison of cilia length and IFT volecity at different developmental stages.

(A, B) Representative images showing cilia in the pronephric duct labeled with anti-acetylated tubulin in 30 hpf and 48 hpf zebrafish larvae. (C) Histograms displaying the velocity of anterograde and retrograde IFT in pronephric duct single cilia of 30 hpf and 48 hpf zebrafish larvae. (D) Length of pronephric duct single cilia of 30 hpf and 48 hpf zebrafish larvae (30 hpf, n=94; 48 hpf, n=97, Unpaired two-sided Student’s t-test). (E) Summary of anterograde and retrograde average velocities in pronephric duct single cilia. (F, G) Snapshot of Tg(hsp70l: ift88-GFP) larvae IFT videos in epidermal cells of 30 hpf and 4 dpf zebrafish larvae. (H) Histograms displaying the velocity of anterograde and retrograde IFT in epidermal cells cilia. (I) Length of epidermal cells cilia (30 hpf, n=40; 4 dpf, n=36, Unpaired two-sided Student’s t-test) (J) Summary of anterograde and retrograde average velocities in epidermal cells cilia. ‘Antero.’ and ‘Retro.’ represent anterograde and retrograde transport, respectively. Each plot was fit by a Gaussian distribution. Scale bars = 5 μm. Numbers of IFT particles are shown in the brackets. n, number of cilia. N, number of zerafish larvae. n.s., not significant, p＞0.05.

Alterations in motor proteins, BBSome proteins, or tubulin modifications have minimal effects on IFT. (A) Left: Snapshot of IFT videos in crista cilia of 4dpf wild type or mutant larvae as indicated. Right: Kymographs showing IFT particle movement along axoneme visualized with Ift88:GFP. Horizontal scale bar: 10 ?m and vertical scale bar:10 s. (B) Histograms showing anterograde and retrograde IFT velocity in crista cilia of control or mutant larvae. (C) Genomic structure and sequences of wild type and ttll3 mutant allele. PAM sequence of sgRNA target are indicated in blue. (D) Protein domain of Ttll3 in wild type and ttll3 mutants. (E) Confocal images showing crista cilia in 4dpf wild type (wt) or maternal-zygotic (MZ) ttll3 mutant visualized with anti-monoglycylated tubulin antibody (green) and anti-acetylated ?-tubulin antibody (red). (F) Histograms depicting IFT velocity in crista cilia of control and ttll3 mutants. Top, anterograde IFT. Bottom, retrograde IFT.(G) Histograms illustrating IFT velocity in crista cilia of ccp5 or ttll6 morphants. Scale bars: 10 ?m in panel A and 5 ?m in panel E. *p?0.5; *** p?0.001. The average IFT velocity and the number of samples (N) are shown in Supplementary file 1.

IFT in the cilia of neuromast hair cells of different zebrafish mutants.

(A) (Left) Snapshot of IFT videos in neuromast cilia of 4 dpf control, kif17 or bbs4 mutants. (Right) Kymographs showing IFT particle movement along axoneme. (B) Histograms showing anterograde and retrograde IFT velocity in neuromast cilia in control and mutants. Horizontal Scale bars = 10 μm. Vertical scale bar = 10 s.

Loss of tubulin glycylation in ttll3 mutants.

(A) Immunostaining with anti-monoglycylated tubulin antibody in 4 dpf ttll3 mutants showing complete absence of monoglycylation modification in cilia of neuromast, olfactory placode and pronephric duct. (B) Immunostaining with anti-polyglycylated tubulin antibody revealed complete elimination of polyglycylation modification in multicilia of olfactory placode and pronephric duct. Red channel shows cilia visualized with anti-acetylated α-tubulin antibody. (C, D) Snapshots of high-speed video showing cilia in pronephric duct of 5 dpf wild type (C) and ttll3 (D) mutant. Bottom images show kymographs of cilia beating. The yellow dashed line indicates the boundary of pronephric duct, and the red line represents the path used to generate the kymograph. (E) Beating frequency of cilia in pronephric duct of wild type and ttll3 mutant (wt, n=11; ttll3, n=10, Unpaired two-sided Student’s t-test). (F) Statistical results showing fertilization rate of wild type and ttll3 mutant (n=7, Unpaired two-sided Student’s t-test). (G, H) Sperm movement trajectory of wild type (G) and ttll3 (H) mutant within 2 s. Scale bar = 5 μm in panel A and B, and 10 μm in panel C, D, G and H.

Validation of the efficiency of ttll6 and ccp5 morpholinos. (A, B, E, F) External phenotypes of control, ttll6 and ccp5 morphant embryos at 2 dpf. (C) Agarose gel electrophoresis of RT-PCR amplicons using primer pairs as indicated in panel (D). The PCR product size in the ttll6 morphants was significantly larger than that in the control (con.). (D) Schematic diagram of ttll6 sp MO target sites (orange) and RT-PCR primer positions. Sequencing results indicated that ttll6 morphant exhibited incorrect splicing, with intron12 being incorrectly included in the mature mRNA. (G) Agarose gel analysis of RT-PCR amplicons using primer pairs as indicated in panel (H). The PCR product size in the ccp5 morphant was significantly larger than that in the control (con.). (H) Schematic diagram illustrating the splice donor sites targeted by antisense morpholinos (orange). Injection of ccp5 MO resulted in the production of aberrant transcripts, wherein intron 5 was mis-spliced into the mature mRNA. Scale bar = 500 ?m. From top to bottom, the size of bands in the panel C and panel G is 8000bp, 5000bp, 3000bp, 2000bp, 1000bp, 750bp, 500bp, 250bp and 100bp respectively.Validation of the efficiency of ttll6 and ccp5 morpholinos. (A, B, E, F) External phenotypes of control, ttll6 and ccp5 morphant embryos at 2 dpf. (C) Agarose gel electrophoresis of RT-PCR amplicons using primer pairs as indicated in panel (D). The PCR product size in the ttll6 morphants was significantly larger than that in the control (con.). (D) Schematic diagram of ttll6 sp MO target sites (orange) and RT-PCR primer positions. Sequencing results indicated that ttll6 morphant exhibited incorrect splicing, with intron12 being incorrectly included in the mature mRNA. (G) Agarose gel analysis of RT-PCR amplicons using primer pairs as indicated in panel (H). The PCR product size in the ccp5 morphant was significantly larger than that in the control (con.). (H) Schematic diagram illustrating the splice donor sites targeted by antisense morpholinos (orange). Injection of ccp5 MO resulted in the production of aberrant transcripts, wherein intron 5 was mis-spliced into the mature mRNA. Scale bar = 500 ?m. From top to bottom, the size of bands in the panel C and panel G is 8000bp, 5000bp, 3000bp, 2000bp, 1000bp, 750bp, 500bp, 250bp and 100bp respectively.

Generation of ATP reporter transgenic line.

(A) Schematic diagram of βactin: arl13b-mRuby-iATPSnFR1.0 transgenic construct. (B) Confocal images showing cilia of crista and epidermal cell in 4dpf transgenic embryos. (C) Statistical results of relative fluorescence intensity (ratio of red fluorescence intensity vs green fluorescence intensity) in cilia of crista and epidermal cells (crista, n=22 cilia from 17 larvae; epidermal cell, n=21 cilia from 13 larvae. Mann-Whitney test. n.s., not significant, p＞0.05). Scale bar = 5 μm. Time-lapse images were continuously collected at an exposure time of 200ms. The display rate was 20 frames per second. Scale bar: 10 μm. The elapsed time is displayed in the top left corner.

Increased size of IFT fluorescent particles in crista cilia. (A) Representative STED images of crista (top) and spinal cord (bottom) in 4dpf Tg(hsp70l: ift88-GFP) larva. Cilia was stained with anti-monoglycylated tubulin (magenta), and IFT88-GFP particles were counterstained with anti-GFP antibody (green). Enlarged views of the boxed region are displayed on the right. (B) Dot plots showing the number of IFT particles per cilia in crista recorded by spinning disk and STED. (Spinning disk, n=15 cilia from 6 larvae; STED, n=14 cilia from 8 larvae.Two-tailed Mann-Whitney test) (C) Statistical analysis showing IFT particles size in the cilia of ear crista and spinal cord. (Crista, n=44 particles from 7 larvae; Spinal cord, n=35 particles from 5 larvae; Unpaired two-sided Student?s t-test) (D) External phenotypes of 2 dpf zebrafish larvae injected with higher and lower dose of ift88 morpholinos. (E) Cilia length quantification of control and ift88 morphants. (ctr, n=62 cilia from 18 larvae; ift88 MO, n=88 cilia from 17 larvae; Two-tailed Mann-Whitney test) (F) STED images showing IFT particles in crista cilia of 3dpf control or ift88 morphants. Enlarged views of the boxed region are displayed on the right. (G) Dot plots showing the number of IFT particles per cilia in control and ift88 morphants. (ctr, n=9 cilia from 5 larvae; ift88 MO, n=11 cilia from 6 larvae; Unpaired two-sided Student?s t-test) (H) Statistical analysis showing IFT particles size of crista cilia in control or ift88 morphants. (ctr, n=141 particles from 11 larvae; ift88 MO, n=88 particles from 13 larvae; Two-tailed Mann-Whitney test) (I) Mean intensity of IFT particles was plotted together with IFT particle size. Linear fit (black line) and coefficient of determination are indicated. (J) Left, Snapshot of IFT videos in crista cilia of 3dpf control (top) or ift88 morphant (bottom) carrying Tg(hsp70l: ift88-GFP). Right, Kymographs showing movement of IFT particles along axoneme. Horizontal scale bar: 10 ?m and vertical scale bar:10 s. (K) Histograms showing IFT velocity in the crista cilia of 3dpf control or ift88 morphants. (L) Model illustrating IFT with different train sizes in long and short cilia. Scale bars: 0.2 ?m in panel A and F, and 500 ?m in panel I. **p?0.01; *** p?0.001. BB, basal body; TZ, transition zone; AX, axoneme.