Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Qilin Is Essential for Cilia Assembly and Normal Kidney Development in Zebrafish

  • Jade Li,

    Affiliation Department of Genetics, Yale University School of Medicine, New Haven, Connecticut, United States of America

  • Zhaoxia Sun

    Affiliation Department of Genetics, Yale University School of Medicine, New Haven, Connecticut, United States of America

Qilin Is Essential for Cilia Assembly and Normal Kidney Development in Zebrafish

  • Jade Li, 
  • Zhaoxia Sun


Defects in the cilium, a once thought vestigial organelle, have recently been implicated in many human diseases, including a number of cystic kidney diseases such as polycystic kidney disease (PKD), Bardet Bieldl Syndrome, and Meckel-Gruber Syndrome. In a forward genetic screen, qilin was identified as a novel gene important in the pathogenesis of kidney cysts in zebrafish. In this paper we characterized qilinhi3959A mutant's phenotypes in detail, investigated cilia formation in this mutant and performed structural and functional analysis of the Qilin protein. Results reveal Qilin's essential role in cilia assembly and maintenance in multiple organs, including the kidney, the lateral line organ, and the outer segment of the photoreceptor cell. In addition, rescue experiments suggest that defective pronephric cilia correlate with the formation of kidney cysts in qilinhi3959A mutants. Further, genetic analysis suggests that qilin interacts with multiple intraflagellar transport (IFT) complex B genes, which is supported by the striking phenotypic similarities between qilinhi3959A and IFT complex B mutants. Finally, through deletion analysis we provide evidence that the well-conserved N-terminus and the coiled-coil domain of Qilin are both essential and sufficient for its function. Taken all the observations together, we propose that Qilin acts in a similar role as IFT complex B proteins in cilia assembly, maintenance and kidney development in zebrafish.


The cilium, an organelle projecting from the cell surface, had long been believed to be vestigial in vertebrates; however, in the past decade it has been shown to play a critical role in both vertebrate physiology and development. Defects in cilia are being linked to an increasing list of human diseases, including polycystic kidney disease (PKD), Bardet-Biedl Syndrome (BBS) and Joubert syndrome, as well as developmental defects such as situs inversus and polydactyl [1][11]. Despite the growing awareness of the functional importance of cilia, our understanding of the regulation of cilia biogenesis and maintenance remains incomplete.

In a forward genetic screen in zebrafish, a group of cystic kidney mutants were identified [12]. Consistent with a central of cilia in PKD pathogenesis, three of the identified genes encode components of complex B of intraflagellar transport (IFT) particles. First identified in the green algae Chlamydomonas, IFT particles are multi-protein complexes believed to carry cargos essential for cilia biogenesis, maintenance and signaling [13], [14]. They are composed of complex A and complex B subunits, with complex A more associated with retrograde transport [15], [16] and complex B involved in anterograde transport [13], [17]. Interestingly, despite phenotypic similarities, some mutants isolated in the screen exhibited cilia biogenesis defects while others were able to assemble cilia [12]. Qilin (also later named as Cluap1 as its encoded protein was identified as a Clusterin-associated protein, [18]) was a novel gene identified in this screen, and qilinhi3959A mutants develop kidney cysts but are capable of cilia assembly [12]. However, although Qilin was not among the IFT particle components biochemically purified from Chlamydomonas, subsequent studies involving Qilin homologues in other organisms suggest a link between Qilin and cilia. In C. elegans, the Qilin homologue DYF-3 was observed to move along the cilia at exactly the same biphasic anterograde rate as those reported for known IFT particles and IFT motors in C. elegans, providing strong evidence that DYF-3 is part of the IFT machinery [19], [20]. In addition, dyf-3 mutants develop truncated sensory cilia, suggesting that dyf-3 plays a role in cilia formation or maintenance in C. elegans [19], [21]. A connection between Qilin and cilia is also implicated by the identification of its human homologue as part of the human cilia proteome, as well as the observation that its homologue in Chlamydmonas is highly upregulated during flagella regeneration [22], [23].

Although implicated in cilia biogenesis, the precise role of Qilin in cilia formation, maintenance and embryonic development, particularly in vertebrates, remains unclear. In this study, we characterized the zebrafish qilinhi3595A mutant in detail. We show that qilin is a maternally supplied gene and the maternal contribution masks the essential function of Qilin in cilia biogenesis and maintenance during early development in zebrafish. Further, in addition to almost identical morphological phenotypes, qilinhi3595A mutants display similar cilia biogenesis defects as IFT complex B mutants. Moreover, we provide evidence that qilin genetically interacts with multiple IFT B complex genes. Together, these results suggest that Qilin functions in the same pathway as IFT B complex genes in cilia biogenesis. Finally, through deletion analysis we show that the N-terminus together with the coiled-coil domain of the Qilin protein is both necessary and sufficient for Qilin's function.


  • qilin is essential for cilia formation and maintenance in zebrafish
  • qilin functions in similar processes as intraflagellar transport (IFT) genes
  • N-terminal and coiled-coil domains of Qilin are essential and sufficient for its functions

Materials and Methods

Zebrafish husbandry

Standard protocols were used for maintaining zebrafish colonies. Embryos were obtained through natural spawning. All lines were maintained in the TAB background.

All zebrafish works have been conducted according to protocols approved by Institutional Animal care and Use Committee (IACUC) of Yale University (Protocol number: 2009–10778).


RNA was extracted from embryos at different developmental stages using Trizol reagent (Invitrogen) according to manufacturer's instructions. Total RNA was reverse-transcribed using an oligo-dT primer and the Superscript II RT-PCR system (Invitrogen). Subsequent Qilin-specific PCR was performed with the following primers: 5′-TACAACTAAAACGGTGACAGT-3′ and 5′-AACCCTCTCAAACTCACAAATTAAC-3′.

Genotyping qilinhi3959A mutants

To genotype progeny from qilinhi3959A carrier in-crosses, three primers were used. One primer is specific to the proviral insertion, close to the insertion site (5′-ACTTGTGGTCTCGCTGTTC-3′). The rest two (5′-GTGACGAACACAGCAACAGACG-3′ and 5′-CCAGTAAACACACAACTGTCACC-3′) are specific for genomic regions flanking the proviral insertion. In the absence of the proviral insertion, amplification would take place between the genomic pair. In contrast, the presence of the 10 kb proviral insertion will disrupt the amplification between the genomic pair. Instead, amplification would occur between the proviral-specific primer and one of the genomic primers.

Histological analysis

Embryos were fixed in Bouin's fixative overnight at room temperature, washed three times in PBS with 0.1% Tween-20 (PBT), embedded in JB-4 resin (Polysciences) following manufacturer's instructions and cut at 4 µm with a microtome. Slides were then stained with hematoxylin and eosin.

Generation of Qilin constructs

The qilin coding sequence was PCR amplified from a zebrafish cDNA pool and cloned into the pCS2+ vector. pCS2-qilin tagged with eGFP on the N-terminus end was generated via PCR cloning. Different qilin deletion constructs were generated via PCR cloning.

In situ hybridization

Embryos were fixed in diluted formalin (1∶2.7 in PBT) at room temperature for an hour or at 4°C overnight. Digoxigenin-UTP labeled RNAs synthesized in vitro were used as probes. Alkaline phosphatase-coupled anti-digoxigenin (Roche) was used to localize hybridized probes. NBT/BCIP (Roche) was used as the chromogenic substrate to produce blue precipitates.

Assay for rescue activity

mRNA was synthesized in vitro using the mMESSAGE mMACHINE kit (Ambion) following the manufacturer's instructions. 174 pg mRNA of eGFP-qilin, as well as other deletion constructs were injected into zebrafish embryos. Embryos were scored for body curvature at 2 dpf (days post-fertilization), and for pronephric cysts at 4–5 dpf.


Embryos were anaesthetized with MESAB and fixed in Dent's fixative (80% methanol and 20% DMSO). The following antibodies were used: mouse monoclonal anti-ϒ tubulin antibody (1∶200 dilution, Sigma T5326), mouse monoclonal anti-acetylated tubulin antibody (1∶5000 dilution, Sigma clone 6-11b-1), rabbit polyclonal anti-Scorpion antibody (1∶2000 dilution, [24]) and rabbit polyclonal anti-Cdh17 (1∶200, [24]). Secondary antibodies from Jackson Immuno Research Laboratories Inc were used at 1∶200 dilutions. DAPI (Invitrogen D3571) was used at 1∶10,000 dilution.

Morpholino Oligos

Morpholino oligonucleotides were purchased from Gene Tools and injected into zebrafish embryos at the one- to four-cell stages. 5′-CATGATTGCTGTCCTTTAATCCAGT-3′ was used to block the translation of qilin. Previously described morpholino oligo was used to block the translation of ift172 [25]. 5′-GGAGGTAATAGTGTGTGTCTACGTG-3′ was used to block the translation of ift27 and 5′- GGACGTAATACTGTCTGTGTACCTG -3′ was used as the mismatch control.

Statistical Analysis

Microsoft Excel was used to derive standard deviation and to perform student's t-tests.


qilinhi3959A mutants develop kidney cysts and ventrally curved body axis

qilinhi3959A was identified in an insertional mutagenesis screen for cystic kidney in zebrafish [12]. Similar to other group II mutants identified in this screen, qilinhi3959A mutants develop ventrally curved bodies, visible at the end of 1 dpf, followed by bilateral kidney cysts, detectable by 2 dpf (Fig. 1A, B). In addition, by 5 dpf, we observed pericardial edema in 40–60% of qilinhi3959A mutants (data not shown).

Figure 1. qilinhi3959A mutant develops body curvature and kidney cysts.

(A, B) Side view of embryos at 3 dpf. Inset in B is a zoomed in view of the kidney cyst in the mutant. (CD) Cross section through the glomerular-tubular region of embryos at 5 dpf. Red arrow in C points to the fused glomeruli, while red dotted lines in D outline the cysts. (EF) Cross sections of the duct of embryos at 5 dpf. Solid green lines outline the duct. Green arrows point to the duct. (GH) Side view of the pronephric duct in whole-mount embryos at 5 dpf stained with Cdh-17. Yellow dotted lines outline the duct. WT: wild type; MUT: mutant. Scale bars in C–F: 20 µm, in G and H; 5 µm.

To analyze kidney cysts in greater detail, we examined the kidney in qilinhi3959A mutants using histological analysis. Cross sections of qilinhi3959A mutant embryos at 5 dpf revealed large, bilateral cysts in the glomerular-tubular region not seen in wild type sibling embryos (Fig. 1C, 1D). In addition, mutant kidney ducts are grossly enlarged compared to their wild type siblings (Fig. 1E, 1F). Kidney duct dilation in qilinhi3959A mutants was further confirmed by immunofluorescent staining of whole-mount embryos using a kidney epithelium-specific marker, Cdh17 (Fig. 1G, 1H).

Disruption of qilin is responsible for phenotypes observed in hi3959A mutants

qilin is a novel gene with twelve exons (Fig. 2A). In hi3959A mutants, a proviral insertion is located in the 5′ UTR of the gene (Fig. 2A). RT-PCR using 2 dpf embryos revealed that full-length qilin transcript is absent in hi3959A mutants (Fig. 2B).

Figure 2. hi3959A is a zygotic null allele of qilin.

(A) Graphic representation of qilin gene. Blue squares represent exons. Red triangle indicates the proviral insertion site. Arrows indicate the direction and location of the genotyping primers used. (B) qilin transcript in embryos at 5 dpf shown by RT-PCR of cDNA from hi3959A embryos (lane 2) and wild type siblings (lane 1). Lane 3 and 4 are beta-actin loading controls for hi3959A (Lane 4) and wild type siblings (Lane 3). (C) Genotyping PCR of embryos from hi3959A carrier in-crosses injected with eGFP-qilin mRNA. All embryos genotyped exhibited wild type phenotypes. Lower band is specific for the mutant allele, while the upper band is specific for the wild type allele. From left to right, the three embryos are heterozygous, homozygous mutant and homozygous wild type, respectively. (D) Graph displaying percent embryos with the curved body phenotype in hi3959A carrier-in crosses that are uninjected (n = 3, with an average of 140 embryos per experiment) and injected with eGFP-qilin mRNA (n = 3, with an average of 120 embryos per experiment). * p< 0.05.

To provide further support for our hypothesis that disruption of qilin resulted in the body curvature and kidney cyst phenotype observed in hi3959A mutants, we designed morpholino oligo against the AUG translational start site of qilin. Wild type embryos injected with qilin morpholino displayed similar phenotypes to those observed in the hi3959A mutants, with ventrally curved bodies and kidney cysts (data not shown).

Finally, we performed rescue experiment using mRNA synthesized in vitro. Embryos from hi3959A heterozygous carrier in-crosses were injected with eGFP-qilin mRNA, encoding Qilin tagged with eGFP at the N-terminus. In three different experiments, an average of 4% of the injected embryo displayed the body curvature phenotype, compare to 20% of the uninjected siblings (Fig. 2D). To confirm that Qilin over-expression rescues the qilinhi3959A mutant phenotype, we genotyped phenotypically wild-type embryos from those injected with the eGFP-qilin mRNA. Of seventeen embryos genotyped, three were homozygous for the hi3959A proviral insertion (Fig. 2C). The rescuing capacity of untagged qilin mRNA is similar to that of eGFP-qilin (data not shown). Taken together, these data indicate that the hi3959A mutant phenotypes are due to the lack of wild type Qilin.

qilin is ubiquitously and maternally expressed

To examine the function of qilin during embryonic development, we first analyzed the expression profile of qilin using in situ hybridization. Results showed that qilin mRNA is distributed ubiquitously throughout the embryo across different developmental stages, from as early as the 8-cell stage to 24 hpf (Fig. 3A–F). The specificity of the result is verified by the complete lack of signal in embryos hybridized with a sense control probe (Fig. 3D, F). RT-PCR analysis further verified the presence of qilin transcript at both the 16-cell and the 2 dpf stages in wild-type embryos (Fig. 3G). In zebrafish, zygotic expression initiates when embryos reach approximately the one thousand-cell stage. Therefore these observations suggest that qilin mRNA is maternally supplied.

Figure 3. qilin is expressed maternally and ubiquitously.

(AF) In situ hybridization for qilin on embryos at the 8- cell stage (A), the shield stage (B), the 3–5 somites stage (C), and the 24 hours-post-fertilization (hpf) stage (E). D and F are sense controls at the 3–5 somites stage (D) and the 24 hpf stage (F). (G) qilin transcript shown by RT-PCR from 16 cell (Lane 1) and 2 dpf (Lane 2) wild type embryos. Lane 3 and 4 are elf1a loading controls for 16-cell (Lane 3) and 2 dpf (Lane 4) samples with matching dilutions of cDNA used in PCR reactions.

qilinhi3959A mutants exhibit cilia biogenesis defects in the pronephric duct

To investigate whether Qilin is required for cilia assembly or maintenance, we examined cilia morphology at various developmental time points in the pronephric duct of qilinhi3959A mutants using the anti-Sco/Arl13b antibody, a cilia-specific marker we established in a previous study [24], [26]. In the pronephric duct, there are two populations of ciliated cells: the multi-ciliated cells (MCCs) that produce bundled cilia, and the single-ciliated cells (SCCs) that display only a single cilium per cell [27], [28]. Both populations are present in the anterior to middle regions of the pronephric duct, while only the SCC population is present in the posterior region of the duct. On 1 dpf, cilia in the anterior pronephric duct of qilinhi3959A mutants are already visibly defective (Fig. 4A, B): while individual cilia from SCCs are still present in the mutants, cilia bundles from MCCs are absent. Consistently, cilia of the posterior pronephric duct, which contains SCCs but not MCCs, appear unaffected in qilinhi3959A mutants (Fig. 4C, D). In comparison, by 5 dpf, the entire pronephric ducts of qilinhi3959A mutants lack cilia, whereas the ducts of wild type embryos have abundant cilia (Fig. 4E, F). Cilia displayed by SCCs that were initially present in the mutants are no longer detectable (Fig. 4E, F). Because bundled cilia never form in qilinhi3959A mutant, Qilin appears to be necessary for the assembly of cilia in MCCs. Meanwhile, since single cilia develop normally but become defective over time, Qilin seems to be required for the maintenance, not initial biogenesis, of cilia in SCCs.

Figure 4. Pronephric cilia in qilin hi3959A mutants are defective.

(AD) Epifluorescent projections showing the pronephric cilia in a wild-type sibling (A, C) and a mutant (B, D) at the 1 dpf; in both anterior (A, B) and posterior (C,D) portions of the duct. Yellow arrowheads in A and B point to rows of basal bodies in MCCs. (E, F) Epifluorescent projections showing the pronephric cilia in a wild-type embryo (E) and a mutant (F) at 5 dpf in the posterior portion of the duct. All embryos were stained with anti-γ-tubulin (red), anti-Sco (green), and DAPI (blue).

Formation of single cilia in the pronephric duct precedes the formation of cilia bundles in multi-ciliated cells

The differential cilia phenotypes in the kidney duct of qilinhi3959A mutants may suggest a cell-type specific function of Qilin. Alternatively, in light of the maternal contribution of Qilin transcript, this phenotypic difference may simply be caused by differential developmental timing of the formation of single cilia and bundles of multi-cilia in the pronephric duct. To test this hypothesis, we performed a careful time course analysis on the formation of cilia in the pronephric duct. At the 24-somite stage, in none of the wild-type embryos analyzed from three independent experiments can bundled cilia be detected, while individual cilia are clearly visible (Fig. 5A, B). At 24 hpf, single cilia are visible in all analyzed wild-type embryos, but multi-cilia are only detected in 22% of the embryos (Fig. 5A, C). At 30 hpf, the percent of embryos with detected multi-cilia increases to 75% (Fig. 5A, D). These observations suggest that single cilia indeed develop earlier than multi-cilia in the developing pronephric duct, consistent with the hypothesis that maternally contributed qilin mRNA was sufficient to support the formation of single cilia, but not bundled multi-cilia (Fig. 5E).

Figure 5. Single cilia form earlier in development than multi-cilia in the pronephric duct.

(A) Graphical representation of multi-cilia observed in embryos at 24 somite, 24 hpf, and 30 hpf. Single cilia were observed at all time points analyzed. Bars represent percentage of embryos that developed multi cilia in the pronephric ducts. Each bar represents data from three independent experiments with at least 8 embryos each. (BD) Representative images of cilia at 24-somite (B), 24 hpf (C) and 30 hpf (D). Yellow arrowheads point to basal bodies of multi-cilia. (E) A model of how different developmental timing of single cilia and multi-cilia could contribute to the pronephric cilia phenotypes observed in hi3959A mutants.

qilin hi3959A mutants exhibit cilia biogenesis defect in multiple sensory organs

Given our finding that Qilin is important in the assembly and maintenance of both populations of pronephric cilia, we were interested in whether Qilin is required for the assembly of cilia in other ciliated cells. We chose to examine ciliated organs formed later in development to eliminate the possibility of maternally contributed qilin masking its requirement in cilia assembly. The lateral line organ usually displays long cilia in wild-type embryos by 2 dpf (Fig. 6A). In qilinhi3959A mutants, however, there are no visible cilia in this organ (Fig. 6B). Notably, the structure of the lateral line organ structure is not affected in qilinhi3959A mutants, suggesting that the lack of Qilin results in a very specific defect in cilia assembly.

Figure 6. Sensory cilia in qilin hi3959A mutants are defective.

(AB) Epifluorescent projections showing the lateral line organ in a wild type (A) and a mutant embryo (B) at 3 dpf. Embryos were stained with rodamine-phalloidin (red), anti- acetylated tubulin (green) and DAPI (blue). Scale bars: 10 µm. (CD) The absence of the outer segment of photoreceptors in the eye as shown through cross sections of a wild type (C) and a mutant embryo (D) at 5 dpf. WT: wild type; MUT: mutant; RPE: retinal pigment epithelium; OS: outer segment; ONL: the outer nuclear layer; INL: the inner nuclear layer; OPL: the outer plexiform layer. Scale bars: 20 µm.

qilinhi3959A mutants also display cilia biogenesis defects in the eye. In the retina, the outer segments (OSs) of photoreceptor cells are modified cilia, which appear as a lightly stained layer between the retinal pigment epithelium (RPE) and the out nuclear layer (ONL) in histological sections (Fig. 6C). We found that in qilinhi3959A mutants, the entire outer segment is missing and the number of nuclei in the outer nuclear layer, which is comprised of the cell bodies of photoreceptor cells, is much reduced. By contrast, the inner nuclear layer (INL) is intact (Fig. 6D). Together, these results suggest that Qilin is required for cilia biogenesis and maintenance in multiple sensory organs in zebrafish.

qilin genetically interact with ift172 and ift27.

qilinhi3959A mutants display many of the phenotypes reported for IFT complex B mutants, including similar body curvature and kidney cyst formation [12]. On the cellular level, qilinhi3959A mutants also show similar cilia phenotypes to that of IFT complex B mutants, but distinct from other cystic kidney mutants in zebrafish [24], [25], [26]. These observations led us to test whether qilin genetically interacts with genes encoding IFT complex B components. Using a morpholino-based assay we first titrated morpholino oligos against qilin, ift172 and ift27 to their suboptimal dosages and then combined two different morpholinos to test whether they act synergistically by assessing the percentage of embryo developing the body curvature phenotypes (Fig. 7A, 7B). Specifically, when wild-type embryos were injected with 2.0 ng ift172 morpholino together with 6.7 ng control morpholino, or 6.7 ng qilin morpholino together with 2.0 ng control morpholino, only 2% and 6.7% displayed the body curvature. However, when wild-type embryos were injected with 2.0 ng ift172 morpholino together with 6.7 ng qilin morpholino, 46.4% developed body curvature (Fig. 7C).

Figure 7. qilin genetically interacts with ift172 and ift27.

(AB) Side view of a representative uninjected control embryo (A) and a phenotypic embryo injected with qilin and/or ift172, ift27 morpholino (B) at 2 dpf. (C, D) Graph displaying the percent of embryos that develop curved bodies. In C, embryos are either injected with qilin morpholino and the control morpholino; ift172 morpholino and the control morpholino; or qilin morpholino with ift172 morpholino. In D, embryos are either injected with qilin morpholino and the ift27 mismatched-control morpholino; ift27 morpholino and the ift27 mismatched-control morpholino; or qilin morpholino with ift27 morpholino. (EG)Graphical representation of the effectiveness of ift27:GFP in rescuing body curvature (E), kidney cysts (F), and laterality defects (G) observed in ift27 morphants. Embryos are either injected with ift27 morpholino and myc:GFP mRNA (blue bars), or ift27 morpholino with ift27:GFP mRNA (pink bars). Laterality defects in (G) is presented as the percentage of embryos developed hearts positioned on the right or center. N = 3 for all experiments, with at least 40 embryos per experiment per condition. *: p< 0.05, **: p< 0.01. MO: morpholino.

The same assay was performed between qilin and another IFT complex B gene, ift27. We designed a morpholino oligo against the translational initiation site of ift27. At the optimal dosage of 8.0 ng, this oligo causes the development of body curvature, kidney cyst, and laterality defect as shown by the position of the heart. Importantly, all three phenotypes can be rescued by expressing a full length ift27 mRNA in the morphants (Fig. 7 E–G), validating the specificity of the ift27 morpholino. When wild-type embryos were injected with the suboptimal 4.0 ng ift27 morpholino together with 6.7 ng of a general control morpholino, or 6.7 ng qilin morpholino together with 4 ng ift27 mismatch-control morpholino, only 14% and 3.0% of the embryos developed the curved body phenotype, respectively. However when wild-type embryos were injected with 4.0 ng ift27 morpholino together with 6.7 ng qilin morpholino 88.2% of the embryos developed body curvature (Fig. 7D). These results, together with the phenotypic similarities between qilinhi3595A mutants and multiple IFT complex B mutants, support our hypothesis that Qilin functions with the IFT B complex in similar processes.

Structural and functional analysis of the Qilin protein

The Qilin protein (Q7ZVC2.2) is predicted to have a coiled-coil domain and an aspartic acid rich domain (Fig. 8A). The zebrafish Qilin is fairly conserved with its mouse and human homologue with an overall identical percent of 64% and 62% respectively (Table S1). The N-terminus and the coiled-coil domain are especially well conserved, with an identical percent of 80% and 79% respectively. To delineate the functional significance of the structural features of Qilin, we generated a series of deletion constructs and tested their ability to rescue qilinhi3595A mutant phenotypes. To ensure that the deletion proteins were stably expressed, we tagged each of them with eGFP to the N-terminus and verified the presence of eGFP signal in ensuring assays. Importantly, similarly tagged full length Qilin was able to rescue qilinhi3595A mutants (Fig. 2).

Figure 8. The N-terminus together with the coiled-coil domain is sufficient for Qilin

's function. (A) Diagram of Qilin structure. NT: N-terminal region; CC: coiled coil domain; AR: aspartic acid rich region; CT: C-terminal tail. Numbers are amino acid coordinates in the protein. All constructs were tagged with eGFP at the N-terminus. (B) Table summarizing the rescuing ability of different deletion constructs assayed in three independent experiments for each construct, with at least 45 embryos per sample per experiment.

We started with the construct that encodes Qilin without the coiled-coil domain (eGFP-ΔCC). qilinhi3595A heterozygote in-crosses injected with ΔCC mRNA developed curvature and kidney cysts at an average of 18.9% from three experiments, not significantly different from uninjected embryos where an average of 24.7% of the embryos developed curvature and kidney cysts (Fig. 8B), suggesting that the coiled-coil domain is necessary for Qilin's functions. We further tested if the coiled-coil domain itself is sufficient to rescue the qilinhi3595A mutant phenotype by over-expressing the coiled-coil domain (eGFP-CC) in embryos from heterozygote in-crosses. Injecting embryos with the eGFP-CC mRNA does not result in significant rescue of the mutant phenotype, where an average of 23.2% of the embryos from three independent experiments developed body curvature and cysts, compared to an average of 21.6% of the uninjected embryos. Taken these results together, we conclude that the coiled-coil domain is necessary but not sufficient for Qilin's function.

Next we systematically deleted the rest of the regions in Qilin, and results suggest that both Qilin without the aspartic acid rich region (eGFP-ΔAR) and Qilin without the C-terminal tail (eGFP-ΔCT) can still rescue mutant phenotypes, where an average of 3.2% and 1.0% of the embryos injected with the respective mRNA developed curvature and cysts. These results suggest that the aspartic acid rich domain, as well as the C-terminal tail, is dispensable for Qilin's function.

Lastly we tested the rescuing ability of mRNA encoding both the N-terminal region and the coiled-coil domain (eGFP-NT+CC), and we discovered that an average of 3.6% of the heterozygous in-cross embryos injected with this mRNA developed body curvature and cysts, compared to an average of 30.0% of the embryos developed these phenotypes in uninjected siblings. These results, combined with the observation that neither the coiled-coil domain nor the N-terminal domain alone can significantly rescue mutant phenotypes, lead us to conclude that the N-terminal tail and the coiled-coil domain together are sufficient for the function of Qilin.


Qilin in vertebrate cilia biogenesis and maintenance

Qilin is a novel protein we isolated in a previous genetic screen for cystic kidney mutants in zebrafish [12]. Consistent with the critical role of cilia in kidney cyst formation, Qilin was subsequently linked to cilia. In the nematode C. elegans, mutants of qilin's homologue dyf-3 show truncated cilia projections [21]. In addition, dyf-3 transcription is regulated by DAF-19, a transcription factor that regulates multiple ciliary genes in C. elegans [29], [30]. In the green alga Chlamydomonas, Qilin is present in the flagellum proteome and the expression of qilin is highly up regulated during flagellum regeneration[23]. In this study, for the first time in an in vivo vertebrate system, we provide clear evidence that Qilin plays an essential role in both cilia biogenesis and maintenance.

We showed that qilin3959A mutants display cilia biogenesis defects in multiple cell types, including MCCs in the pronephric duct, photoreceptors of the eye, and hair cells in the lateral line organ. Further, the maternal contribution of qilin and the differential development timing of cilia biogenesis allowed us to uncover Qilin's role in cilia maintenance. Specifically, we found that single cilia and bundled cilia in the pronephric duct form at different developmental stages: while single cilia can be detected as early as the 24 somite stage, bundled cilia are not visible in a majority of embryos until 30 hpf. Interestingly, in qilin3959A mutants, single cilia initially assemble normally, but degenerate as embryos develop, and by 5 dpf single cilia are no longer visible in mutant embryos. By contrast, bundled cilia never form. One possible interpretation is that Qilin's function is cell type specific. However, given that qilin transcript is maternally supplied, we postulate that maternally expressed Qilin is able to support the biogenesis of single cilia in the pronephric duct and the gradual degradation of the maternal protein reveals Qilin's role in cilia maintenance. The failure of cilia formation in photoreceptors in the eye and hair cells in the lateral line organ, both of which form single cilia later in development, supports our hypothesis.

Interestingly, the phenotype of the qilin mutant we observed is different from previously observed in a qilin morphant [31]. It is possible that the published oligo was able to block the translation of maternally deposited qilin mRNA, thus revealing more severe phenotypes. Alternatively, off-target effect of the morpholino oligo, which is different from the one used in this study, could contribute to the reported morphant phenotypes. The generation and analysis of maternal zygotic qilin mutants in the future should be able to provide definitive results to distinguish between these two possibilities.

Qilin and IFT

In this study, we show that partial reduction of Qilin synergize with partial reduction of IFT B complex components, suggesting that qilin genetically interacts with IFT B complex genes. Further, despite the similar morphological phenotypes displayed by Group II mutants isolated from our previous mutagenesis screen for cystic mutants, careful analysis revealed distinct ciliary defects in these mutants. For example, In scorpion/arl13b mutants, pronephric cilia are severely shortened and reduced in number at 50 hpf, but a significant number of cilia manage to form by 5 dpf [24]. In seahorse mutants, neither cilia density nor length is significantly different from those in wild type embryos [26]. Interestingly, both ift57 and ift172 mutants show cilia maintenance defect in SCCs of the pronephric duct and cilia biogenesis defect in MCCs of the pronephric duct, photoreceptors and the lateral line organ [25]. The almost identical cilia biogenesis and maintenance defects in qilin and IFT B-complex mutants suggest that Qilin's function is more closely associated with IFT B complex, among other cilia-associated proteins.

This hypothesis is also supported by other studies. Qilin homolog was seen to travel in the cilium in a pattern and velocity precisely as those observed for the anterograde movement of IFT particles and IFT motors in C. elegans [19], [23]. Further, in C. elegans, loss of BBSome leads to the separation of the IFT-B/OSM-3 complex and the IFT-B/OSM-3 complex. Importantly, under this condition, the Qilin homologue travels specifically with the latter, providing strong support for the association of Qilin with IFT B complex [32]. Finally, Qilin was pulled down together with Ift27, Ift57 and Ift172 from zebrafish embryo lysate [33]. Whether Qilin is a core component of IFT B complex or it functions as a peripheral regulator of this complex remains unclear at this stage. Biochemical purification of IFT complexes, as those done in Chlamydomonas [34], in both zebrafish and C. elegans, will provide direct evidence for the role of Qilin in IFT complexes.

Structural features of Qilin protein

Similar to many cilia associated proteins, structure of Qilin reveals little regarding its function. The only recognizable domain in Qilin is a coiled-coil domain followed by an aspartic acid rich region in the middle of the protein. Through serial deletion analysis, we showed that the N-terminal region and the coiled-coil domain are both required for Qilin's function in normal body axis and kidney development, while both the aspartic acid-rich region and the C-terminal tail are dispensable for these functions. This result is consistent with our observation that the N-terminal region and the coiled-coil domain are highly conserved between multiple species, whereas the aspartic acid-rich region and the C-terminal end of the protein are less well conserved. Identification of proteins that directly interact with the N-terminal region and the coiled-coil region of Qilin will provide critical insight to mechanisms underlying Qilin's function.

Supporting Information

Table S1.

Sequence conservation of Qilin protein. The names and accession numbers of Qilin's homologue in C. elegans, Drosophila Melanogaster, Mus musculus, and Homo sapiens are listed. The percentages of protein sequence identity of the different homologues to zebrafish Qilin are also listed.



We thank members of the Sun lab and members of the Yale Center for PKD research for helpful discussions, Nicole Semanchik for superb technical assistance, and SueAnn Mentone for assistance on histology.

Author Contributions

Conceived and designed the experiments: ZS JL. Performed the experiments: ZS JL. Analyzed the data: ZS JL. Contributed reagents/materials/analysis tools: ZS JL. Wrote the paper: ZS JL.


  1. 1. Doherty D (2009) Joubert syndrome: insights into brain development, cilium biology, and complex disease. Semin Pediatr Neurol 16: 143–154.
  2. 2. Pazour GJ, Dickert BL, Vucica Y, Seeley ES, Rosenbaum JL, et al. (2000) Chlamydomonas IFT88 and its mouse homologue, polycystic kidney disease gene tg737, are required for assembly of cilia and flagella. J Cell Biol 151: 709–718.
  3. 3. Murcia NS, Richards WG, Yoder BK, Mucenski ML, Dunlap JR, et al. (2000) The Oak Ridge Polycystic Kidney (orpk) disease gene is required for left-right axis determination. Development 127: 2347–2355.
  4. 4. Haycraft CJ, Swoboda P, Taulman PD, Thomas JH, Yoder BK (2001) The C. elegans homolog of the murine cystic kidney disease gene Tg737 functions in a ciliogenic pathway and is disrupted in osm-5 mutant worms. Development 128: 1493–1505.
  5. 5. Qin H, Rosenbaum JL, Barr MM (2001) An autosomal recessive polycystic kidney disease gene homolog is involved in intraflagellar transport in C. elegans ciliated sensory neurons. Curr Biol 11: 457–461.
  6. 6. Ansley SJ, Badano JL, Blacque OE, Hill J, Hoskins BE, et al. (2003) Basal body dysfunction is a likely cause of pleiotropic Bardet-Biedl syndrome. Nature 425: 628–633.
  7. 7. Badano JL, Ansley SJ, Leitch CC, Lewis RA, Lupski JR, et al. (2003) Identification of a novel Bardet-Biedl syndrome protein, BBS7, that shares structural features with BBS1 and BBS2. Am J Hum Genet 72: 650–658.
  8. 8. Mykytyn K, Nishimura DY, Searby CC, Shastri M, Yen HJ, et al. (2002) Identification of the gene (BBS1) most commonly involved in Bardet-Biedl syndrome, a complex human obesity syndrome. Nat Genet 31: 435–438.
  9. 9. Blacque OE, Reardon MJ, Li C, McCarthy J, Mahjoub MR, et al. (2004) Loss of C. elegans BBS-7 and BBS-8 protein function results in cilia defects and compromised intraflagellar transport. Genes Dev 18: 1630–1642.
  10. 10. Li JB, Gerdes JM, Haycraft CJ, Fan Y, Teslovich TM, et al. (2004) Comparative genomics identifies a flagellar and basal body proteome that includes the BBS5 human disease gene. Cell 117: 541–552.
  11. 11. Eley L, Yates LM, Goodship JA (2005) Cilia and disease. Curr Opin Genet Dev 15: 308–314.
  12. 12. Sun Z, Amsterdam A, Pazour GJ, Cole DG, Miller MS, et al. (2004) A genetic screen in zebrafish identifies cilia genes as a principal cause of cystic kidney. Development 131: 4085–4093.
  13. 13. Kozminski KG, Johnson KA, Forscher P, Rosenbaum JL (1993) A motility in the eukaryotic flagellum unrelated to flagellar beating. Proc Natl Acad Sci U S A 90: 5519–5523.
  14. 14. Pedersen LB, Rosenbaum JL (2008) Intraflagellar transport (IFT) role in ciliary assembly, resorption and signalling. Curr Top Dev Biol 85: 23–61.
  15. 15. Pazour GJ, Wilkerson CG, Witman GB (1998) A dynein light chain is essential for the retrograde particle movement of intraflagellar transport (IFT). J Cell Biol 141: 979–992.
  16. 16. Piperno G, Siuda E, Henderson S, Segil M, Vaananen H, et al. (1998) Distinct mutants of retrograde intraflagellar transport (IFT) share similar morphological and molecular defects. J Cell Biol 143: 1591–1601.
  17. 17. Cole DG, Diener DR, Himelblau AL, Beech PL, Fuster JC, et al. (1998) Chlamydomonas kinesin-II-dependent intraflagellar transport (IFT): IFT particles contain proteins required for ciliary assembly in Caenorhabditis elegans sensory neurons. J Cell Biol 141: 993–1008.
  18. 18. Takahashi M, Lin YM, Nakamura Y, Furukawa Y (2004) Isolation and characterization of a novel gene CLUAP1 whose expression is frequently upregulated in colon cancer. Oncogene 23: 9289–9294.
  19. 19. Ou G, Qin H, Rosenbaum JL, Scholey JM (2005) The PKD protein qilin undergoes intraflagellar transport. Curr Biol 15: R410–411.
  20. 20. Snow JJ, Ou G, Gunnarson AL, Walker MR, Zhou HM, et al. (2004) Two anterograde intraflagellar transport motors cooperate to build sensory cilia on C. elegans neurons. Nat Cell Biol 6: 1109–1113.
  21. 21. Murayama T, Toh Y, Ohshima Y, Koga M (2005) The dyf-3 gene encodes a novel protein required for sensory cilium formation in Caenorhabditis elegans. J Mol Biol 346: 677–687.
  22. 22. Marshall WF (2004) Human cilia proteome contains homolog of zebrafish polycystic kidney disease gene qilin. Curr Biol 14: R913–914.
  23. 23. Stolc V, Samanta MP, Tongprasit W, Marshall WF (2005) Genome-wide transcriptional analysis of flagellar regeneration in Chlamydomonas reinhardtii identifies orthologs of ciliary disease genes. Proc Natl Acad Sci U S A 102: 3703–3707.
  24. 24. Duldulao NA, Lee S, Sun Z (2009) Cilia localization is essential for in vivo functions of the Joubert syndrome protein Arl13b/Scorpion. Development 136: 4033–4042.
  25. 25. Cao Y, Park A, Sun ZIntraflagellar transport proteins are essential for cilia formation and for planar cell polarity. J Am Soc Nephrol 21: 1326–1333.
  26. 26. Kishimoto N, Cao Y, Park A, Sun Z (2008) Cystic kidney gene seahorse regulates cilia-mediated processes and Wnt pathways. Dev Cell 14: 954–961.
  27. 27. Liu Y, Pathak N, Kramer-Zucker A, Drummond IA (2007) Notch signaling controls the differentiation of transporting epithelia and multiciliated cells in the zebrafish pronephros. Development 134: 1111–1122.
  28. 28. Ma M, Jiang YJ (2007) Jagged2a-notch signaling mediates cell fate choice in the zebrafish pronephric duct. PLoS Genet 3: e18.
  29. 29. Swoboda P, Adler HT, Thomas JH (2000) The RFX-type transcription factor DAF-19 regulates sensory neuron cilium formation in C. elegans. Mol Cell 5: 411–421.
  30. 30. Senti G, Swoboda P (2008) Distinct isoforms of the RFX transcription factor DAF-19 regulate ciliogenesis and maintenance of synaptic activity. Mol Biol Cell 19: 5517–5528.
  31. 31. Aanstad P, Santos N, Corbit KC, Scherz PJ, Trinh le A, et al. (2009) The extracellular domain of Smoothened regulates ciliary localization and is required for high-level Hh signaling. Curr Biol 19: 1034–1039.
  32. 32. Ou G, Koga M, Blacque OE, Murayama T, Ohshima Y, et al. (2007) Sensory ciliogenesis in Caenorhabditis elegans: assignment of IFT components into distinct modules based on transport and phenotypic profiles. Mol Biol Cell 18: 1554–1569.
  33. 33. Omori Y, Zhao C, Saras A, Mukhopadhyay S, Kim W, et al. (2008) Elipsa is an early determinant of ciliogenesis that links the IFT particle to membrane-associated small GTPase Rab8. Nat Cell Biol 10: 437–444.
  34. 34. Lucker BF, Miller MS, Dziedzic SA, Blackmarr PT, Cole DGDirect interactions of intraflagellar transport complex B proteins IFT88, IFT52, and IFT46. J Biol Chem 285: 21508–21518.