The Zn finger protein Iguana impacts Hedgehog signaling by promoting ciliogenesis

Abstract

Hedgehog signaling is critical for metazoan development and requires cilia for pathway activity. The gene iguana was discovered in zebrafish as required for Hedgehog signaling, and encodes a novel Zn finger protein. Planarians are flatworms with robust regenerative capacities and utilize epidermal cilia for locomotion. RNA interference of Smed-iguana in the planarian Schmidtea mediterranea caused cilia loss and failure to regenerate new cilia, but did not cause defects similar to those observed in hedgehog(RNAi) animals. Smed-iguana gene expression was also similar in pattern to the expression of multiple other ciliogenesis genes, but was not required for expression of these ciliogenesis genes. iguana-defective zebrafish had too few motile cilia in pronephric ducts and in Kupffer's vesicle. Kupffer's vesicle promotes left-right asymmetry and iguana mutant embryos had left-right asymmetry defects. Finally, human Iguana proteins (dZIP1 and dZIP1L) localize to the basal bodies of primary cilia and, together, are required for primary cilia formation. Our results indicate that a critical and broadly conserved function for Iguana is in ciliogenesis and that this function has come to be required for Hedgehog signaling in vertebrates.

Figure 1. Smed-iguana(RNAi) planarians display defects in locomotion and osmoregulation

(A) Smed-iguana(RNAi) animals lost the ability to locomote away from a source of light. Data are mean velocities from animals at different times following initial RNAi ± standard deviations. unc-22 dsRNA was used as the RNAi control. At least five animals per condition were analyzed at each timepoint. (B) Smed-iguana(RNAi) animals were capable of regenerating normally, but displayed bloating and blistering defects. Animals were fed dsRNA-containing food on days 0 and 4, cut on day 5, and cut again on day 14. Animals were photographed 28 or 29 days following initiation of RNAi. Anterior, left. Bar, 0.5 mm. Top, trunk fragments (see cartoon, oriented with anterior up, for amputation sites). Bottom, head fragments. 3/14 head fragments were blistered (not shown, 8/20 tail fragments blistered). 3/21 trunks displayed bloating (not shown, 19/20 tail fragments bloated). Red dotted lines indicate approximate blastema boundary. pr, photoreceptors. px, pharynx. white arrows, bloating. red asterisk, blistering.

Figure 2. Smed-iguana(RNAi) planarians lack head rim cilia

(A) Differential interference contrast (DIC) images of the head rims of control and Smed-iguana(RNAi) animals. Black dotted lines demarcate the single-cell thick epidermis at the edge of the animal. Cilia protrude from this epidermis. Yellow asterisks label the few remaining cilia of a Smed-iguana(RNAi) animal. Animals were examined 14 days after initiation of RNAi and are representatives from the data shown in panel B. (B) The length and number of cilia were determined from unc-22 and Smed-iguana RNAi animals. Cilia lengths were determined for at least seven animals and cilia density for at least five animals per timepoint (the day six iguana(RNAi) timepoint involved four animals). Data are means from animals at different times following initial RNAi ± standard deviations.

Figure 3. Smed-iguana(RNAi) planarians cannot maintain or regenerate cilia

(A) The cilia of intact, fixed planarians were labeled with an anti-acetylated tubulin antibody. unc-22 dsRNA was used as the RNAi control. Top, views of the dorsal animal surface. A stripe of dorsal cilia is visible in the control(RNAi) animal (yellow arrows), but was absent from a Smed-iguana(RNAi) animal (iguana(RNAi). Bottom, the ventral surface of a control(RNAi) animal is covered with cilia, but an iguana(RNAi) animal failed to maintain cilia. The anti-acetylated tubulin antibody also labels numerous sub-epidermal tubules that remained visible as small fluorescent dots in the dorsal and ventral images of iguana(RNAi) animals. 15/15 animals showed a similar phenotype to the representatives shown. Animals were fixed 14 days after initial RNAi feeding. Bar, 100 microns. (B, C) Animals were fed dsRNA on days 0 and 4, decapitated on day 5, and resultant head fragments that were regenerating tails (cartoon) were fixed 7 days later. The tail blastemas of head fragments (circled in red in the cartoon) are depicted. Anterior, up. Bars, 50 microns. (B) Ventral surfaces of tail blastemas on head fragments, labeled with an anti-acetylated tubulin antibody are shown. Control animals displayed coverage of the ventral blastema epidermis with motile cilia whereas an iguana(RNAi) animal did not. 15 control and 15 iguana(RNAi) animals displayed results similar to the representative shown. px, pharynx opening. (C) Ciliated flame bulbs of the planarian excretory system in the tail blastemas of control and iguana(RNAi) head fragments were labeled with an anti-acetylated tubulin antibody and visualized with optical sectioning. Z-stacks of optical sections taken internal to the labeled epidermis are shown; not all flame bulbs can be visualized in these images. Flame bulbs are indicated with a yellow asterisk. The numbers of flame bulbs in the tail blastemas of 13 animal head fragments were determined. The region posterior to the regenerated pharynx was used for counting. Below, mean numbers of flame bulbs per blastema ± standard deviations are shown. Data were determined to be significantly different, P <0.0001, unpaired t-test.

Figure 4. Smed-iguana is expressed in ciliated cell types

(A) Wild-type animals were labeled with Smed-iguana, dnah1, ift88, BBS1, BBS2, BBS9, and rootletin riboprobes in a whole-mount in situ hybridization and similar expression patterns were observed. The animal labeled with the iguana riboprobe is oriented with anterior up and all other animals are oriented with anterior left. (A-C) White arrows, dorsal stripe cilia. Black arrows, peripheral head signal. pr, photoreceptors. px, pharynx. Bar, 100 microns. (B) Intact iguana and control RNAi animals were labeled with riboprobes from cilia genes. There was no detected difference in expression of cilia genes, despite the absence of detectable cilia on iguana(RNAi) animals as determined with anti-acetylated tubulin antibody (α-ac tub) labeling, right. Animals were fixed 14 days after initiation of RNAi. (C) The head regeneration blastemas of trunk fragments are shown (see Figure 1b for depiction of trunk fragment). Animals were fixed 16 days after initiation of RNAi, and following 7 days of regeneration. iguana(RNAi) animals were able to regenerate cells expressing cilia markers, despite absence of ability to regenerate ciliated cells (right, dorsal cilia are labeled with an anti-acetylated tubulin antibody. Bar, 50 microns. Arrow, dorsal cilia stripe).

Figure 5. hedgehog(RNAi) and iguana(RNAi) planarians have different phenotypes

RNAi animals were amputated to remove heads and tails, and were analyzed following eight days of regeneration. (A) Ventral view of the head blastemas of animals labeled with an anti-acetylated tubulin antibody (α-ac tub). iguana(RNAi) animals had greatly reduced or complete absence of cilia (n=27). By contrast, hedgehog(RNAi) and gli2-1(RNAi) animals displayed normal ciliogenesis (n=29). Anterior, up. Bar, 100 microns. (B) Differential interference contrast (DIC) images of tail blastemas in fixed animals. The approximate boundary between the blastema and the preexisting tissue was determined by fluorescence differences between these regions. Values listed are average areas of blastemas determined from multiple images as shown. hedgehog and gli2-1 RNAi caused animals to regenerate grossly aberrant tails that were small (n=17; P<0.0001, unpaired t-test). By contrast, iguana(RNAi) animals regenerated tails of normal size (n=18). Anterior, up. Bar, 100 microns.

Figure 6. iguana mutant zebrafish have defects in ciliogenesis and left-right asymmetry

(A) Top, brightfield images of an igu mutant embryo and a wild-type sibling embryo (labeled at right as ‘WT sib’, see methods), 48 hours post fertilization (hpf). The red line indicates the region of the pronephric duct that was analyzed for cilia. Anterior, left. Dorsal, up. Bar, 200 microns. Bottom, animals were labeled at 48 hpf with an anti-acetylated tubulin antibody. Bar, 20 microns. Right, the numbers of cilia within the posterior-most 100 microns of the pronephric duct were determined. The numbers of cilia in igu morphants and control morpholino-injected embryos were also determined. Data are means ± standard deviations. Asterisk, data were determined to be significantly different, P <0.0001, unpaired t-test. (B) Animals were fixed at 28 hpf and labeled with a riboprobe from the cardiac myosin light chain 2 gene (cmlc2). Anterior, up; dorsal view. Bar, 100 microns. Asterisks indicate the labeled heart. A control morpholino-injected embryo displayed heart development on the left side of the embryo and an iguana morpholino-injected embryo displayed heart development on the right side. Right, percentages of embryos displaying left, medial, or right heart development are shown. 44 control morphants (control MO), 55 igu morphants (igu MO), 49 wild-type siblings of igu^tm79a/tm79a embryos (WT sib, see methods), and 9 igu mutant embryos were analyzed. (C) Cilia in Kupffer's vesicle from control morpholino, iguana morpholino, and smoothened (smo) morpholino-injected embryos were labeled with an anti-acetylated tubulin antibody (red). Nuclei are labeled with DAPI (blue). White dotted circle, approximate boundary Kupffer's vesicle. Embryos were fixed at the six somite stage. Bar, 10 microns. Right, mean numbers of cilia in Kupffer's vesicle ± standard deviations. Asterisk, data were determined to be significantly different, P <0.0001, unpaired t-test.

Figure 7. The human Iguana-like DZIP1 and DZIP1L proteins localize to basal bodies and are required for primary ciliogenesis

(A) GFP-tagged human Iguana-like proteins dZIP1 and dZIP1L localized to the basal bodies of cilia. An anti-centrin antibody labeled the basal bodies and an anti-acetylated tubulin antibody labeled primary cilia. hTERT-RPE1 cells were used to visualize primary cilia. (E) RNAi of dZIP1 and dZIP1L resulted in a defect in formation of primary cilia. 295 control, 282 double RNAi, 84 dZIP1, and 99 dZIP1L RNAi cells were analyzed. Only cells for which there was a clear presence or absence of a cilia were counted. Differences between the double RNAi and control cells were significant (P< 0.0001)