FIGURE SUMMARY
Title

Characterization of primary cilia during the differentiation of retinal ganglion cells in the zebrafish

Authors
Lepanto, P., Davison, C., Casanova, G., Badano, J.L., Zolessi, F.R.
Source
Full text @ Neural Dev.

Main features of apical primary cilia in the early differentiating retinal neuroepithelium. The embryonic zebrafish retinal neuroepithelium was analyzed using confocal microscopy and TEM. a-b 26 hpf embryos expressing Arl13b-GFP (localized to primary cilia) (a) or double transgenic 35 hpf embryos expressing Arl13b-GFP and atoh7:gap-RFP (expressed in progenitors during the last cell cycle and in RGC neuroblasts) (b) were fixed and analyzed in toto using confocal microscopy. A 3D maximum intensity projection of a 3 µm-thick confocal stack is shown. c 26 hpf embryos expressing Arl13b-GFP were immunolabeled with anti-acetylated tubulin antibody. A maximum intensity projection of a 3 µm-thick stack of the apical region of the neuroepithelium is shown. The arrowheads show primary cilia with Arl13b-GFP and acetylated tubulin labeling. d Single confocal plane with a detail of the stack shown in b. It is possible to observe cells bearing a primary cilium (double arrowhead) and expressing low levels of gap-RFP (full arrowheads). e-h TEM micrographs showing examples of apical primary cilia, either with a complete (e, bracket), incomplete (f, bracket - asterisk) or absent ciliary pocket (g and h, asterisks). Primary cilia in close contact with RPE cells were also observed (h). The basal body is indicated with a white arrowhead. i Cross section of an apically localized primary cilium. j Morphological parameters of apical primary cilia of 35 hpf embryos obtained from measurements performed on TEM micrographs. Measured features are summarized in the upper diagrams, and values (mean ± standard deviation) shown in the lower table. k TEM micrograph showing a basal body (white arrowhead) associated with the apical plasma membrane but lacking an axoneme. l Comparison of apically-localized primary cilia length at 26 and 35 hpf. The numbers in brackets represent the number of cilia / embryos measured in each case. (***) p < 0.001, Student’s t test. RPE: retinal pigment epithelium. Scale bars: A-B, 20 µm; C-D, 10 µm; E, 1 µm; F-H, 0.5 µm; I, 0.1 µm; K, 1 µm

Primary cilia emerging from the basolateral membrane of retinal neuroepithelial cells. TEM micrographs of retinas from 26 and 35 hpf embryos. a Low magnification (left) and high magnification serial sections (right) of a retina from a 35 hpf embryo. A primary cilium emerges from the basolateral membrane, as evidenced by the presence of adherent junctions (AJ). b Basal body without an axoneme, but closely associated with the basolateral membrane. c Basal body with a short axoneme inside a cytoplasmic vesicle. d Low magnification (left) and high magnification serial sections (right) of a retina from a 26 hpf embryo showing a primary cilium emerging from the basolateral membrane. White arrowhead: basal body; RPE: retinal pigment epithelium. Scale bars: A-D, 1 µm

Primary cilia first appear in RGC neuroblasts during apical process retraction and in an apical position. a-b Blastomeres from double transgenic embryos (atoh7:gap-RFP/Arl13b-GFP) were transplanted into wild type hosts. The resulting embryos were imaged through time-lapse confocal microscopy from around 30 hpf onwards. Montages from 3D maximum intensity projections of the stacks are shown. a A progenitor is shown, which loses its primary cilium (arrowhead) at t = 4.00 h, previous to entering M phase. b Neuroblast imaged throughout apical process retraction. The full arrowhead denotes the presence and position of the primary cilium, while the double arrowhead denotes the apical tip of the retracting process and the empty arrowhead the region of axonal outgrowth. c Plot of the relative apico-basal position of the neuroepithelium in which primary cilia first appear during the retraction of the apical process. The final position occupied by the neuroblast cell bodies (grey dashed line) and the median value of the data (black line) are also shown. d Plot of the individual values and median time-delay (black line) between the initiation of apical retraction and primary cilia appearance. e TEM micrograph from a 35 hpf embryo, showing a primary cilium at the apical tip of a retracting process. The white arrowhead indicates the basal body. RPE: retinal pigment epithelium. Time is shown in hrs:min. Scale bars: A-B, 10 µm; E, 1 µm

Upon completion of apical retraction, primary cilia may transiently disappear or lose their apical localization. a 3D maximum intensity projections of a region of the retina of a 35 hpf transgenic embryo expressing Arl13b-GFP and immunolabaled with zn8 antibody to identify RGCs. Primary cilia are observed colocalizing with the apical tip of cells finishing retraction. b TEM micrograph from a 35 hpf embryo showing a primary cilium at the tip of an apical process. The white arrowhead indicates the basal body. c-e Blastomeres from transgenic embryos (atoh7:gap-RFP/Arl13b-GFP) were transplanted into wild type hosts. The resulting embryos were imaged by time-lapse confocal microscopy from around 30 hpf onwards. Montages from 3D maximum intensity projections of the stacks are shown. (d) Detail of the cell shown in (c), where initial filopodial activity (arrowheads) previous to dendrite formation is observed; in order to highlight atoh7:gap-RFP signal intensity, a color ramp is used. The full white arrowheads indicate the presence and position of the primary cilium in c and e. f-g TEM micrographs from 35 hpf embryos, showing cells located in the basal region of the neuroepithelium (RGC neuroblasts) with primary cilia both in apical (f) and basal (g) positions. The white arrowhead indicates the basal body. BL: basal lamina. Scale bars: A, 10 µm; B, 1 µm; C-E, 10 µm; F-G, 1 µm

Primary cilia in maturing RGCs. a-b Arl13b-GFP transgenic embryos were fixed and imaged through confocal microscopy as whole-mounts (a, 48 hpf) or as cryosections (b, 5 dpf). Images shown are maximum intensity Z-projections of 1.5 µm (a) and 15 µm (b) confocal microscopy stacks. c-e TEM micrographs from retinas of 48 hpf embryos. In most of the cases we observed membrane-docked basal bodies without visible axonemes (c). Some cells showed basal bodies associated to long axonemes (d) or short axonemes with dilated tips (e). The white arrowhead indicates the basal body. f Comparison between the number of plasma membrane-associated and non-associated centrioles/centrosomes in the RGC layer of 48 hpf embryos (a total of 6 embryos were used in the quantification). The former were further classified in subtypes illustrated in the micrographs C, D and E. GCL: ganglion cell layer; IPL: inner plexiform layer; INL: inner nuclear layer; BL: basal lamina. Scale bars: A-B, 20 µm; C-E, 1 µm

Evaluation of different morpholino oligomers for elipsa and ift88 knock-down. a-c External phenotype of 48 hpf embryos injected with different morpholinos targeting the ciliary proteins ift88 and elipsa; translational-blocking morpholinos: ift88-ATG MO (a), elipsa-ATG MO (b); splice-blocking morpholinos: ift88-SP or elipsa-SP (c). d RT-PCR analysis of elipsa and ift88 mRNA levels in 35 hpf embryos either injected with ift88-SP (6 ng) or elipsa-SP (6 ng) alone or as a combination (6 ng each) (“s”: spliced mRNA form; “ms”: mis-spliced mRNA form). Gapdh mRNA was used as a control. e Evaluation of the external phenotype of 48 hpf embryos injected with different amounts of the combination of splice-blocking morpholinos (elipsa-SP/ift88-SP MOs), where “malformed” refers to morphological alterations not compatible with a classic “ciliary phenotype”. f Main characteristics of the external phenotype of double morphants at the dose used in the rest of the study. The black rectangle marks the position of the otic vesicle in the low magnification images, magnified in the insets. The arrowheads indicate embryos with enlarged brain ventricles. g Quantification of the percentage of atoh7:gap-GFP embryos injected with different amounts of control MO or elipsa-SP/ift88-SP MOs displaying reduced size of the RGC layer (“retina phenotype”), at 48 hpf. Scale bars: A-C, 500 µm; F, 200 µm

Effective reduction of primary cilia length in zebrafish embryos upon elipsa and ift88 knock-down. a Confocal images of different ciliated organs from 48 hpf embryos. Cilia were labeled with an anti-acetylated tubulin antibody and F-actin with TRITC-phalloidin. b Kupffer’s vesicle of eight-somite stage embryos, where basal bodies were labeled with an anti-γ-tubulin antibody. c Comparison of primary ciliary length in Kupffer’s vesicle. The experiments were performed twice, with similar results; only the results from one of the experiments are shown. d Comparison of the length of apical primary cilia in the retina of 35 hpf morphant and control embryos. Measurements were made on TEM micrographs. In C and D the numbers in brackets represent the number of cilia and embryos analyzed in each condition. (***) p < 0.001, Mann-Whitney test. Scale bars: A-B, 10 µm

Embryos with impaired cilia have smaller eyes and a reduced RGC layer. a Confocal images of 48 hpf and 60 hpf retinas from embryos injected with control morpholino or elipsa-SP/ift88-SP MOs. RGCs were labeled with zn8 antibody. b and d Comparison of retinal volumes in 48 hpf (b) and 60 hpf (d) embryos. c and e Comparison of RGC layer volume (based on zn8 stain) in 48 hpf (c) and 60 hpf (e) embryos. The numbers in brackets represent the number of embryos quantified in each case (we quantified one eye per embryo). (***) p < 0.001. Comparisons were made using rank transformation and Student’s t test. nr: neural retina; le: lens; on: optic nerve. Scale bars: A, 50 µm

Cell-autonomous effect of cilia reduction on RGC differentiation. a-b Blastomeres from atoh7:gap-RFP embryos injected with control MO (12 ng) (a) or elipsa-SP/ift88-SP MOs (5/5 ng) (b) were transplanted into atoh7:gap-GFP hosts. The dotted line indicates de apical margin of the RGC layer of the host and the asterisk denotes the donor cells with high RFP expression. c-d Blastomeres from un-injected atoh7:gap-GFP embryos were transplanted into atoh7:gap-RFP hosts injected with elipsa-SP/ift88-SP MOs (6/6 ng). Retinas with small (c) and large (d) clones of GFP positive cells were observed. In C the dotted line indicates de apical margin of the RGC layer formed by the donor GFP positive cells, and the asterisk denotes host cells with high RFP expression. Single arrowheads mark GFP expressing cells that appear before RFP expressing cells (double arrowheads). In all the cases (a-d) the resulting host embryos were imaged by time-lapse confocal microscopy beginning at around 30 hpf. Fluorescence images show montages from 3D maximum intensity projections of the stacks from three different time points. On the left, the external phenotype at 48 hpf of the donor and host embryos is shown in each case. Time is expressed in hrs:min. Scale bars: A-D, 20 µm. e Comparison of the number of mitotic cells (identified by anti-pHistone H3 labeling) in the ventro-nasal region of 24 and 36 hpf retinas, expressed as the number of positive cells per unit of volume. (*) p = 0.0014. f Comparison of the number of apoptotic cells (identified by anti-activated Caspase 3 labeling) per retina in 36 and 48 hpf embryos. (*) p = 0.009. All experiments were performed twice, with similar results; only the results from one of the experiments are shown. The numbers in brackets represent the number of embryos quantified in each case (we quantified one eye per embryo). (ns) non-significant difference. Comparisons were made using rank transformation and Student’s t test

Cilia impaired embryos show a preferential delay in RGCs formation. a Left: average intensity Z-projections of 10 µm-thick confocal stacks of the nasal region of retinas from 48 hpf SoFa1 embryos injected with control MO or elipsa-SP/ift88-SP MOs. The dotted line delimits the area used to measure the fluorescence intensity profile from apical to basal, plotted in the right side for each image. ap: apical; bas: basal; au: arbitrary units. b Diagram indicating the regions in the SoFa1 retina fluorescence profile plots that were selected to determine the relative contribution of each cell type to the total signal in RFP and CFP channels (the GFP signal was only used to determine the extension spanned by amacrine cells in the RFP channel). The boundaries of each region were set at the distance corresponding to 50 % of the maximum fluorescence for each peak. c Plot of the percentage of signal intensity in the gap-RFP and gap-CFP channels corresponding to each cell type in embryos injected with control MO or elipsa-SP/ift88-SP MOs. The number of embryos analyzed in each case is shown in brackets. (***) p < 0.001, (**) p < 0.01, ns: non-significant difference. Comparisons were made using rank transformation and Student’s t test. PR: photoreceptors; AC: amacrine cells; RGC: retinal ganglion cells; BP: bipolar cells. d Confocal images of retinas from wild-type embryos injected with control MO or elipsa-SP/ift88-SP MOs and fixed at 48 and 60 hpf. The boxed regions were magnified in the upper insets, while lower images are orthogonal 3D projections of these insets. The white arrowhead indicates a patch of IPL in the region adjacent to the optic nerve exit. Scale bars: A, 50 µm; D, 30 µm

Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image. Full text @ Neural Dev.