Dynein axonemal intermediate chain 2 is required for formation of the left–right body axis and kidney in medaka

Developmental Biology 347 (2010) 53–61

Contents lists available at ScienceDirect

Developmental Biology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / d e v e l o p m e n t a l b i o l o g y

Dynein axonemal intermediate chain 2 is required for formation of the left–right body axis and kidney in medaka Yusuke Nagao a, Jinglei Cheng b, Keiichiro Kamura c, Ryoko Seki d, Aya Maeda a, Daichi Nihei c, Sumito Koshida c, Yuko Wakamatsu a,d, Toyoshi Fujimoto b, Masahiko Hibi a,d,e, Hisashi Hashimoto a,d,⁎ a

Department of Biological Sciences, Graduate School of Science, Nagoya University, Nagoya, Japan Department of Anatomy and Molecular Cell Biology, Graduate School of Medicine, Nagoya University, Nagoya, Japan Department of Biological Sciences, Graduate School of Science, the University of Tokyo, Tokyo, Japan d Bioscience and Biotechnology Center, Nagoya University, Nagoya, Japan e Laboratory for Vertebrate Axis Formation, RIKEN Center for Developmental Biology, Kobe, Japan b c

a r t i c l e

i n f o

Article history: Received for publication 17 March 2010 Revised 14 July 2010 Accepted 3 August 2010 Available online 11 August 2010 Keywords: Ciliary disease Dynein arms Kidney cyst Nodal flow Left–right asymmetry Medaka

a b s t r a c t Ciliary defects lead to various diseases, such as primary ciliary dyskinesia (PCD) and polycystic kidney disease (PKD). We isolated a medaka mutant mii, which exhibits defects in the left–right (LR) polarity of organs, and found that mii encodes dynein axonemal intermediate chain 2a (dnai2a). Ortholog mutations were recently reported to cause PCD in humans. mii mutant embryos exhibited loss of nodal flow in Kupffer's Vesicle (KV), which is equivalent to the mammalian node, and abnormal expression of the left-specific gene. KV cilia in the mii mutant were defective in their outer dynein arms (ODAs), indicating that Dnai2a is required for ODA formation in KV cilia. While the mii mutant retained motility of the renal cilia and failed to show PKD, the loss of dnai2a and another dnai2 ortholog dnai2b led to PKD. These findings demonstrate that Dnai2 proteins control LR polarity and kidney formation through regulation of ciliary motility. © 2010 Elsevier Inc. All rights reserved.

Introduction Primary ciliary dyskinesia (PCD; OMIM accession 242650) is a group of autosomal-recessive genetic disorders of the motile cilia which affects approximately one in 20,000–60,000 humans. Motile cilia dysfunction results in pleiotrophic phenotypes in diverse organs of PCD patients (Zariwala et al., 2007). For instance, multiple motile cilia covering the upper and lower respiratory tracts are known to be required for mucociliary clearance, and thus their dysfunction leads to recurrent respiratory infections and inflammation of the respiratory tract in PCD patients. Male PCD patients often manifest reduced fertility, which can be accounted for by the dysmotility of the sperm flagella. These motile cilia in respiratory systems and sperm exhibit a 9 + 2 axonemal orientation with dynein arms on each peripheral microtubule doublet, and exhibit a beating activity (Satir and Christensen, 2007). A portion of PCD patients exhibits situs inversus (heterotaxy), which is referred to as Kartagener syndrome (OMIM accession 244400). The abnormal organ laterality seen in PCD patients is caused by monocilia dysfunction in the embryonic node. The nodal cilia are ⁎ Corresponding author. Bioscience and Biotechnology Center, Nagoya University, Nagoya, Japan. Fax: +81 52 789 5053. E-mail address: [email protected] (H. Hashimoto). 0012-1606/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ydbio.2010.08.001

solitary and have a 9 + 0 axonememal structure with dynein arms. The nodal cilia are motile and generate a leftward flow across the node (nodal flow) by rotational movement (Okada et al., 2005; Satir and Christensen, 2007). Studies in mouse mutants revealed that the nodal flow is critical for the proper formation of LR body axis (reviewed in (Wagner and Yost, 2000)). Polycystic kidney disease (PKD; OMIM accession 173900), a common genetic disease characterized by distension of the renal tubules (and/or correcting ducts), is now categorized as a ciliary disease (Igarashi and Somlo, 2007). Most PKD patients carry a mutation in PKD1 or PKD2, which encode polycystin 1 and polycystin 2, respectively, which localize at the ciliary membrane. Among the mouse PKD models, the Tg737 (Polaris) mutant, which is defective in a component of the intraflagellar transport (IFT) system and lacks ciliary structures, displays PKD and abnormal LR polarity formation. Although both PKD and PCD are caused by defective cilia, PCD does not usually accompany PKD. Renal cilia in the lumen of the tubules, which consist of 9 + 0 axonemes without dynein arms in mammals, are immotile, and are called “primary cilia” (Satir and Christensen, 2007). Recent advances in the understanding of PKD pathogenesis suggest that immotile primary cilia in the renal tubules function as passive mechanosensors in detecting the fluid flow rate in the tubule and thereby sense the lumen size (reviewed in (Yoder, 2007)). It is thought that abnormality of primary cilia mechanoreceptor function

54

Y. Nagao et al. / Developmental Biology 347 (2010) 53–61

results in impaired regulation of the renal lumen size and eventually leads to PKD. Fish, such as medaka (Oryzias latipes) and zebrafish (Danio rerio), are emerging model animals for study of developmental genetics (Furutani-Seiki et al., 2004; Ishikawa, 2000; Kasahara et al., 2007; Wittbrodt et al., 2002). Accumulating data have revealed that the mechanisms underlying LR polarity formation are conserved between fish and other vertebrates (Essner et al., 2002; Levin, 2005). In fish, initial breaking of symmetry is achieved in Kupffer's vesicle (KV), which is functionally equivalent to the mouse node, where the nodal flow is generated by rotational movement of the cilia with 9 + 0 axonemes and dynein arms, as is the case in mammals (Essner et al., 2005; Okada et al., 2005). In contrast to the mammalian kidney, the maintenance of the renal lumen size requires active beating of the renal cilia, which produces the driving force for intratubular urine flow; loss or reduced motility of the cilia causes PKD in medaka and zebrafish (Kramer-Zucker et al., 2005; Omran et al., 2008). Consistent with this, the renal cilia in fish consist of 9 + 2 axonemes with dynein arms, which are features of motile cilia. The species-specific difference in phenotypes associated with defective renal ciliary motility was recently shown in mutants of Kintoun/PF13, which function in the pre-assembly of axonemal dyneins (Omran et al., 2008). The Medaka kintoun mutant exhibits organ laterality defects and PKD, while human patients having a mutation in the ortholog gene exhibit defective LR polarity formation, but not the PKD phenotype (Omran et al., 2008). We carried out mutant screening in medaka and isolated a mutant, mii (mirror image of the internal organs), which displayed defects in organ laterality. By positional cloning, we found that the mii locus encodes a medaka ortholog (Dnai2a) of dynein axonemal intermediate chain 2 (Dnai2), which is a component of the outer dynein arms (ODAs). We demonstrated that dnai2a is expressed in the KV and that mii mutants display loss of nodal flow and randomization of the LR organ polarity. We have found that dnai2b, another dnai2 ortholog, is expressed in the prospective kidney, and that inhibition of Dnai2a and Dnai2b resulted in PKD. The human DNAI2 gene has been recently identified as a causative gene of PCD (Loges et al., 2008). Our findings reveal a conserved role for Dnai2 in ciliary motility and nodal flowmediated LR polarity formation, and the involvement of Dnai2a and Dnai2b in the motility of the renal cilia as well as kidney formation in medaka. We demonstrate that the mii mutant is a medaka disease model for human PCD.

individuals in three embryos per group according to the method described previously (Hashimoto et al., 2009). Positional cloning The mii heterozygous fish were crossed with wild type HNI fish. We obtained 150 embryos of their F2 offspring displaying reversed internal organs. Bulked segregant analysis with the M-marker system (Kimura et al., 2004) using these putative homozygotes mapped the mii locus on the medaka LG 17 (data not shown). By further recombination analysis, the mii locus was narrowed down to the 480 kb region between the flanking markers MF01SSB034M20 and AU169106. Detailed information on the markers used is available on request. Antisense-mediated knockdown and rescue by synthetic RNA injection Antisense gripNA (Active motif) for dnai2a and dnai2b was designed to prevent dnai2 mRNA from being spliced at exon4/intron4 donor site and morpholino for dnai2b at exon1/intron1. The gripNA sequence is: 5'-CTGACCCACCAACATCCC-3' for dnai2a (grip-dnai2a), 5'-TTGACCCACCAATGTCCC-3' for dnai2b (grip-dnai2b). The morpholino sequence is: 5'-ACATTTTCAGGTTAACTCACCTGTT-3' for dnai2b (MO-dnai2b). The gripNA and morpholino were resuspended and diluted with HPLC water. A gripNA 5'-GGCACTGACCTTCAGGCT-3' (recognizes the gene which is unrelated to dnai2a/b) was used as a control for the knockdown experiments described in Fig. 5. Wild type dnai2a and dnai2b mRNAs were synthesized from a cloned cDNA with the open reading frame in pCS2 vector as previously described (Hashimoto et al., 2004). Wild type OR embryos and offspring from crossed heterozygous mii were used for knockdown and for rescue, respectively. gripNA or synthetic RNA was microinjected in one blastomere at the 1 or 2-cell stage embryos. Transmission electron microscopy Embryos were dechorionated with hatching enzyme and incubated until the 6somite stage. Kidneys were obtained from OR fish and mii homozygous fish, which were 3–4 months old. Samples were fixed with 2% glutaraldehyde, 2% paraformaldehyde, 0.2% tannic acid, and 10% sucrose in 0.1 M cacodylate buffer (pH7.4) at 4 °C overnight, followed by postfixation with 1% OsO4 and 10% sucrose in 0.1 M cacodylate buffer (pH7.4) for 1 h and dehydrated with ethanol. Then the samples were stained with 2% uranyl acetate in ethanol for 30 min and embedded in Quetol812 resin. Detection of nodal flow To detect nodal flow, fluorescent beads were injected into Kupffer's vesicle of 2somite embryos as previously described (Hojo et al., 2007), and the movement of the fluorescent beads was observed as an indicator of nodal flow. Sample embryos were prepared by artificial insemination with homozygous eggs and sperm. Histological analysis Paraffin sections were prepared according to the method previously described (Fedorova et al., 2008). Briefly, fish were fixed with Bouin's solution, dehydrated with an alcohol series, and then embedded in paraffin. The samples were transversely sectioned at a thickness of 6 μm and stained with hematoxylin and eosin.

Materials and methods Observation of renal cilia Medaka strains The orange-red variety (OR) of the medaka fish Oryzias latipes was used as a wild type strain. Mutagenesis screen was carried out using the see-through II (STII) strain (Wakamatsu et al., 2001). The HNI-I inbred strain (Hyodo-Taguchi, 1996) was used for crossing in genetic mapping. The mii mutant STII was repeatedly crossed with the OR for more than 5 times to change the genetic background. The mii heterozygous fish was identified by the production of homozygous offspring in pair mating, and used to maintain the mutant line.

Kidneys were excised from OR fish and mii homozygous fish. In Yamamoto's Ringer solution (128 mM NaCl, 5 mM KCl, 1.4 mM CaCl2 and 2.4 mM NaHCO3, pH7.3), kidneys were finely cut up with a scalpel and then the tissues were additionally broken up with trypsinization and voltex mixing. Samples were mounted on a slide glass and observed under microscopy.

Results

Whole mount in situ hybridization

Medaka mii mutant displays abnormal left–right patterning

In situ hybridization was performed as previously described (Hashimoto et al., 2004). A digoxygenin-labeled riboprobe was made from a cloned template in pDrive (Qiagen) or pCRII-TOPO (Invitrogen) using SP6 or T7 polymerase after restriction enzyme digestion. The embryos were treated with Proteinase K (10 μg/ml) in PBS for 5 min at 30 °C before hybridization. Detection signals were developed in BM purple (Roche).

We carried out N-ethyl-N-nitrosourea (ENU)-mediated mutagenesis to screen medaka mutants with defects in embryonic and adult organogenesis (the detailed information will be published elsewhere). Two mutations exhibited abnormal organ laterality, and they turned out to be non-allelic. We named one mutant ‘mii’ (mirror image of the internal organs) after the mutant phenotype and further analyzed it in this study. mii mutants were found to be viable and grew to adulthood, although they exhibited a severely wavy trunk and did not spawn by natural mating (Fig. 1A and B). In the wild type medaka, the heart ventricle is positioned on the right side and the gallbladder and liver are on the left side. This laterality was retained at a high rate (n = 164/165)

Immunohistochemistry Immunohistochemistry was performed as described previously (Hashimoto et al., 2009). The cilia were visualized with anti-acetylated-α-tubulin antibody (Sigma) and goat anti-mouse Alexa 488 (Invitrogen). Images were obtained with a laser-scanning inverted microscope (LSM 700, Carl Zeiss). At least 15 KV cilia length was measured for each

Y. Nagao et al. / Developmental Biology 347 (2010) 53–61

in the OR strain (Fig. 1C and D), indicating that the OR genetic background did not perturb normal LR polarity formation. In contrast, approximately 14% of the embryos obtained from mii heterozygous parents had reverse-positioned organs (Fig. 1E, F and Table 1). Although most of these embryos displayed a mirror image pattern for both the heart ventricle and gallbladder positions (situs inversus; 9.7%), some of the embryos exhibited discordant positioning of these organs (heterotaxia; 2.3% only heart looping, 1.5% only gallbladder). We next examined the expression pattern of southpaw (also called nodal-related 3), which is one of the asymmetric genes expressed only on the left side in the lateral plate mesoderm (LPM) during early left– right patterning (Long et al., 2003; Soroldoni et al., 2007). Whole mount in situ hybridization (WISH) analysis showed that southpaw was asymmetrically expressed in the left LPM of wild type embryos at the 6–9 somite stage (Fig. 1G). In contrast, approximately one quarter of the embryos from mii heterozygous parents had aberrant expression of southpaw in the LPM: left-sided (normal, Fig. 1H, 74.0%), right-sided (Fig. 1I, 4.1%), bilateral (Fig. 1J, 17.8%) and absent (Fig. 1K, 4.1%) (Table 2), indicating that the mii mutation is recessive. The randomized localization of southpaw expression in LPM may account for the aberrant organ laterality in the mii mutant, as Southpaw plays an essential role in LR polarity formation (Long et al., 2003). A smaller number of embryos from the mii homozygotes exhibited abnormal organ laterality, compared with those showing the aberrant southpaw expression (13.6 vs. 26.0 %). The data suggest that the organ laterality was recovered in approximately one half of the mii homozygotes. dnai2a is a strong candidate gene for the mii mutant Positional cloning placed the mii locus in a 480-kbp region on linkage group (LG) 17. Search in the medaka genome database revealed that there are a few candidate genes that are known to be related to the LR patterning. Among them, we considered dnai2a (accession AB546260) as a strong candidate for the mii mutant locus since Dnai2 is known to be associated with cilia function. dnai2 (also known as oda6 or IC2) encodes a dynein axonemal intermediate chain, which is a component of the ODAs in Chlamydomonas flagella (Mitchell and Kang, 1991). Sequencing analysis identified a single nucleotide substitution between the mii homozygote and wild type dnai2a cDNAs: a transversion from C to T at nucleotide position 478 in exon 4 introduces a premature stop codon at amino acid 160 (Fig. 2A). Like the case of the human DNAI2 protein (Pennarun et al., 2000), medaka Dnai2a protein has WD40 repeats, which are lost in the mii mutant Dnai2a. The data suggest that mii mutation causes a loss-of-function of Dnai2a. Loss of the intact DNAI2a protein in the mii mutant causes defects in left–right asymmetry Knockdown experiments with gripNA antisense oligonucleotides revealed that the silencing of dnai2a resulted in a phenotype similar to the mii mutants (Table 1). Approximately one half of the knockdown embryos (5 dpf) had reversed organs; both heart looping and the gallbladder in 31.4%, only heart looping in 2.9%, and only the gallbladder in 14.3%. In addition, most of the knockdown embryos (5 hpf) exhibited abnormal expression of southpaw in the LPM: left-sided expression in 16.7%, right-sided expression in 5.5%, bilateral expression in 63.0%, and no expression in 14.8% (Table 2). The data are consistent with the mii mutant phenotypes, and the organ laterality was recovered in approximately one half of the dnai2a-defective embryos. Next, we carried out rescue experiments by injecting wild type dnai2a mRNA into the embryos obtained from crossing with heterozygous fish (Tables 1 and 2). The organ laterality and the expression of southpaw in the left LPM was recovered in the injected embryos. These results led us to conclude that the mutation in dnai2a is responsible for the mii phenotype.

55

dnai2a mRNA is expressed in ciliated organs WISH revealed that dnai2a is expressed in KV (Fig. 2B) and notochord (data not shown) at the 2-somite stage, and in otic vesicle (Supplemenentary Fig. 1) at stage 28. The expression was detected in the pronephric tubules and ducts of 5 dph larvae (Fig. 2C). In all these organs, motile cilia have been suggested to play important roles in a variety of species. The exclusive expression of dnai2a in these organs implies an involvement of Dnai2a in the motile cilia function in medaka, although we did not detect any otolith abnormalities in mii mutant (Supplementary Fig. 1). mii lacks nodal flow in KV Previous studies in medaka and zebrafish have demonstrated that unidirectional nodal flow breaks the LR symmetry of the embryo in the KV and the nodal flow plays an initial role in establishing proper organ laterality, similar to the mouse (Essner et al., 2005; Hojo et al., 2007; Wagner and Yost, 2000). We visualized the fluid flow in the KV by the injection of fluorescent beads at an early somite stage. In wild type KV, the beads rotated in a counterclockwise manner when observed from the dorsal side, and eventually moved leftward (n = 3, Fig. 3A and Supplementary Movie 1). In contrast, the movement of the beads was completely abolished in mii KV, (n = 3, Fig. 3B and Supplementary Movie 2), indicating that Dnai2a is required for the generation of the nodal flow. It has been reported that mutations in the dnai2a ortholog genes DNAI2 and oda6 deprive the cilia and flagella of the ODAs in humans and Chlamydomonas, respectively (Loges et al., 2008; Mitchell and Kang, 1991). To address whether mii have defects in ODAs in the KV cilia, we examined the ultrastructure of the KV cilia. Whereas the cilia in the KV of wild type embryo exhibited a 9 + 0 axonemal orientation with the ODAs (n = 3, Fig. 3C), the KV cilia in mii mutant embryo had 9 + 0 microtubule doublets, but lacked the ODA (n = 3, Fig. 3D). Furthermore, we examined cilia morphology in KV using antiacetylated-α-tubulin antibody. No significant difference in the cilia length and localization between wild type and mii was observed (Fig. 3E and F, length = 4.64 ± 0.63 in wild type, 4.71 ± 0.68 in mii mutant). These results indicate that Dnai2a is required for the formation of the ODAs in KV cilia. These data reveal that the motility of KV cilia, which requires Dnai2a, generates the unidirectional nodal flow that is required for the LR polarity formation. mii does not result in any PKD lesions Previous studies in medaka and zebrafish have shown that the cilia on renal tubule epithelial cells consist of 9+2 axonemes with the ODAs and are motile; defects in ciliary motility results in PKD (Kramer-Zucker et al., 2005; Omran et al., 2008). To assess whether mii mutant develops PKD, we carried out histological sectioning of the kidney in the mii mutant (Fig. 4A and B). No lumen of the renal tubules was markedly expanded in the mii mutant kidney. The mii mutant kidney did not exhibit any apparent histological abnormality in the renal tubular epithelia. Consistent with this, the ODA formation with 9+2 axonemes in renal cilia was not affected in the mii mutants (n=5, Fig. 4C and D). Furthermore, the beating motion of renal cilia on the surface of tubular epithelia in the mii mutant kidney was indistinguishable from that in the wild type kidney (Supplementary Movie 3 and 4). These findings suggest that the mutation of Dnai2a alone does not affect the motility of renal cilia or the formation of kidney in medaka. Another dnai2 ortholog dnai2b compensates for the loss of dnai2a in medaka Although dnai2a mRNA is expressed in the pronephros and later in the mesonephros (data not shown), loss of Dnai2a in the mii mutant did not cause defects in the ODA formation of renal cilia or cilia

56

Y. Nagao et al. / Developmental Biology 347 (2010) 53–61

Fig. 1. Medaka mii mutant. (A, B) Appearance of adult medaka. Compared to the wild type fish (A), the mii mutant (B) has a shortened body axis because of wavy trunk. Scale bar, 10 mm. (C–F) The position of the heart ventricle and gallbladder in embryos at 7 days post fertilization (dpf). (C, D) In the wild type embryo, the heart ventricle and gallbladder are observed on the left when viewed ventrally and dorsally, respectively. (E, F) Some of the mii mutant embryos exhibit situs inversus. Embryos were dechorionated before being pictured. (C, E) Ventral views. (D, F) Dorsal views. hv, heart ventricle; gb, gallbladder. Scale bar, 200 μm. (G–K) Expression patterns of southpaw at the 9-somite stage. (G) southpaw is normally expressed in the left LPM in the wild type embryo. (H–K) The laterality of southpaw expression in LPM is randomized in the mii mutant embryos; (H) normally left-sided, (I) right-sided, (J) bilateral, and (K) absent.

Y. Nagao et al. / Developmental Biology 347 (2010) 53–61 Table 1 The rate of internal organ laterality. Normal

Reversed organs hl and gb

Embryos from heterozygous parents 86.4% Knockdown embryos of dnai2a* 51.4% Knockdown embryos of dnai2b* 100% Rescued embryos by wild type 95.5% dnai2a mRNA ** Rescued embryos by wild type 96.5% dnai2b mRNA**

9.7% 31.4% 0% 1.8% 2.3%

hl

n gb

2.3% 1.5% 2.9% 14.3% 0% 0% 1.8% 0.9% 1.2%

0%

390 35 39 106 82

The heart looping and the position of gallbladder at 5 day post fertilization were scored. hl, heart looping; gb, gallbladder. *Antisense oligonucleotide (grip-dnai2a or grip-dnai2b) was injected in wild type OR embryos, **synthetic mRNA was injected in embryos from heterozygous parents.

Table 2 The rate of southpaw expression.

Embryos from heterozygous parents Knockdown embryos* Rescued embryos by wild type dnai2a mRNA** Rescued embryos by wild type dnai2b mRNA**

Left

Right

Bilateral

Non

n

74.0% 16.7% 91.1%

4.1% 5.5% 4.4%

17.8% 63.0% 4.4%

4.1% 14.8% 0%

73 54 45

90.6%

1.9%

5.7%

1.9%

53

Expression of southpaw in the lateral plate mesoderm at the 9-somite stage was scored. *Antisense oligonucleotide (grip-dnai2a) was injected in wild type OR embryos, **Synthetic mRNA was injected in embryos from heterozygous parents.

motility. We hypothesize that there is another ortholog gene of dnai2, which functions redundantly with dnai2a in the medaka kidney. We searched the medaka genome database by tBLASTN using the amino acid sequence of medaka Dnai2a, and found a putative ortholog Dnai2b (accession AB546261) containing conserved WD40 repeats on LG 21 (Fig. 5A and B). WISH analysis revealed that dnai2b mRNA is expressed in the pronephros and the mesonephros and in the otic vesicles (Supplementary Fig. 1), but not in the KV (Fig. 5C, D, and data not shown for mesonephros). To test our hypothesis, we attempted to rescue the laterality defects of the mii mutants by injecting wild type dnai2b mRNA into embryos from mii heterozygous parents. Whereas defective organ laterality was observed in approximately one-eighth of the uninjected control embryos (data not shown, Table 1 in other experiments), most of the injected embryos displayed normal situs of the internal organs

57

examined (heart and gallbladder) (Table 1). Consistent with this, the dnai2b mRNA injection suppressed the aberrant expression of southpaw in the LPM of the mii mutants. These results suggest that Dnai2b protein can indeed replace the function of Dnai2a in the KV cilia. In order to examine whether Dnai2b protein actually plays a compensatory role for Dnai2a in the renal cilia of medaka, we next conducted knockdown experiments of dnai2b. Antisense oligonucleotide against dnai2b (dnai2b-gripNA) did not perturb their LR patterning (Supplementary Table), indicating that dnai2b-gripNA has no detectable cross-reactivity to dnai2a transcripts. The dnai2b-gripNA was injected into embryos from the mii heterozygous parents, and putative mii homozygotes showing abnormalities in organ situs were selected at 5 dpf as double lossof-function embryos of dnai2a and dnai2b, fixed at the 9 dpf larval stage and subjected to paraffin sectioning. We also conducted knockdown experiments of dnai2a and dnai2b by injecting antisense oligonucleotides into wild type OR embryos. We selected embryos displaying abnormal organ situs, an indication of strong Dnai2a defects, and further analyzed their kidney structure at the 9 dpf larval stage. In both experiments, a portion of these knockdown larvae exhibited distention of the pronephric tubules or glomus (17.9% and 36.4%, respectively, Fig. 5E–I and Supplementary Table). Furthermore, some of the dnai2b-knockdown larvae displayed pronephric cysts, although they were less severe than those in the double loss-offunction larvae of dnai2a and dnai2b. These results suggest that Dnai2b functions as a complementary molecule of Dnai2a and a more important component of ODAs in the pronephros and, presumably, later in the mesonephros to generate intratubular flow, which is required for the formation and maintenance of the kidney.

Discussion Recent advances in molecular genetics have revealed that not only the motile cilia, but also the immotile cilia, play important roles in organogenesis and tissue physiology (Bisgrove and Yost, 2006; Fliegauf et al., 2007). Medaka mutants have been suggested to be good models for studies of human genetic diseases, including ciliopathy, since a conserved genetic pathway controls the development, function, and maintenance of organs between medaka and mammals. In the present study, we have characterized medaka mii, which is a mutant of dnai2a, and elucidated the roles of Dnai2 in ciliary motility to establish the LR body axis and to prevent medaka kidney from becoming cystic.

Fig. 2. Identification of a point mutation in the dnai2a gene of the mii mutant genome and detection of dnai2a expression in the KV, renal tubules and ducts. (A) A single nucleotide substitution (478C N T) was identified in exon 4 of the medaka dnai2a gene, which is located within the candidate region on LG 17 mapped by positional cloning. The medaka dnai2a gene consists of at least 12 exons. Exons are shown as solid boxes. The mii mutation (478C N T) leads to a premature stop codon, resulting in truncation of the C-terminal half including the WD40 repeats (see Fig. 5). Yellow boxes represent the exonic domain encoding the WD40 repeats. (B, C) dnai2a expression in wild type embryos. (B) At the 2-somite stage, dnai2a was expressed in KV. Dotted circle shows the position of KV. (C) Strong staining was detected in the renal (pronephric) tubules and ducts of 5 days posthatching (dph) larvae.

58

Y. Nagao et al. / Developmental Biology 347 (2010) 53–61

Role of Dnai2 in motile cilia Recent findings have advanced our understanding that disruption of ciliogenic and cilia-associated genes give rise to pleiotropic defects in organ function and maintenance in vertebrates, suggesting that cilia play important roles in a variety of organs (Bisgrove and Yost, 2006). The phenotypic pleiotropy in ciliopathy may be explained by the diverse functions of cilia. In mammals, at least three types of cilia can be found:

type-1, multi- or single-motile cilia with 9 + 2 axonemes and the ODAs exhibiting rhythmic beating in respiratory tract cells and sperm; type-2, a single motile cilium with 9 + 0 axonemes and the ODAs displaying rotation of cells on the node; type-3, a single immotile cilium with 9 + 0 axonemes lacking the ODAs on renal tubular cells and olfactory cells. Previous studies have shown that the peripheral doublet microtubules and the ODAs are critical for the initiation and propagation of ciliary bending, while the central pair/radial spokes serve to regulate the beat frequency and wave form. The ODAs on the peripheral axonemes are critical for ciliary motility (Satir and Christensen, 2007). Our results confirmed that the cilia in the KV epithelial cells of medaka have 9 + 0 axonemes with the ODA (type-2), similar to those in the mammalian node. The mii mutant embryos exhibited loss of the ODAs in the KV cilia and exhibited immotile cilia in the KV. Chlamydomonas oda6, which is a mutant of the dnai2 ortholog IC69 (IC2), has immotile flagella lacking ODAs (Mitchell and Kang, 1991). In human PCD patients carrying DNAI2 mutations, the ODAs are lost in respiratory cilia (Loges et al., 2008). These lines of evidence reveal that Dnai2 is indispensable for the ODA complex assembly across species, and that Dnai2 is required for the ciliary motility. Role of Dnai2 in LR patterning We found that most of the homozygous mii mutants and dnai2aknockdown embryos exhibited an abnormal expression of southpaw in the LPM, but only half of them had abnormal positioning of the internal organs (Fig. 1, Tables 1 and 2). A similar situation is also reported in human PCD patients having a mutation in DNAI2 (Zariwala et al., 2007). There are two possibilities to explain this difference. First, the majority of Dnai2a-defective embryos exhibited bilateral expression of southpaw in the LPM. In this situation, the laterality of the heart and gallbladder may become random, as observed in the zebrafish embryos defective in the Nodal inhibitor Charon (Hashimoto et al., 2004). Thus, only half of these embryos exhibit abnormal laterality and this may be related to the fewer phenotypic embryos caused by the loss of Dnai2a. Alternatively, Nodal signal-mediated formation of LR polarity is regulated by a self-enhancing and lateral inhibition system, and is highly robust (Nakamura et al., 2006). Ectopic activation or suppression of the Nodal signal on one side affects the inhibition and activation of Nodal expression on the other side in mice (Nakamura et al., 2006). It is possible that dnai2a-defective embryos with bilateral and absent southpaw expression might fix unilateral southpaw expression at a later developmental stage, which is sufficient to lead to either normal situs or situs inversus. Consistent with this, most of the dnai2a embryos did exhibit either normal situs or situs inversus, and only a minor population had discordant laterality of the heart and gallbladder (Fig. 1, and Tables 1, 2). Although human DNAI2-defecitve PCD patients are known to exhibit abnormal organ laterality (Loges et al., 2008), the underlying mechanisms have not been elucidated. In this study, we show that in medaka, dnai2a is expressed in the KV epithelia (Fig. 2) and that Dnai2a is required both for ODA formation in the KV cilia and generating leftward nodal flow (Fig. 3). Our data suggest that DNAI2 functions in the nodal cilia to generate nodal flow in humans. Our Fig. 3. Detection of nodal flow in KV. (A, B) Videomicroscopies of nodal flow. Fluorescent beads were injected into the KV of the wild type (A) and mii mutant (B) embryos at the 2somite stage. (A1-A4) In wild type KV, a fluorescent bead moves to the left (arrow) in images captured in intervals of one second. (A5) The trajectory of a single bead is shown (n = 3, see Supplementary Movie 1). (B1–B5) In mii mutant KV, no moving beads were observed. All three beads pointed out by arrowheads remained motionless (n = 3, see Supplementary Movie 2). (A1–5, B1–5) Dorsal views anterior to the top. (C, D) Transmission electron micrographs of KV cilia. (C) Cross-section of cilia in wild type KV shows a 9 + 0 axonemes with outer dynein arms (ODAs), indicated by yellow arrowheads (n = 3). (D) ODAs are completely absent in KV cilia of the mii mutant (n = 3). Scale bar, 50 nm. (E, F) Confocal images of cilia phenotype in KV. Six-somite stage embryos of wild type (E) and mii mutant were labeled with antibody against anti-acetylated tubulin. The dotted circle shows the position of KV. Scale bar, 20 μm.

Y. Nagao et al. / Developmental Biology 347 (2010) 53–61

59

Fig. 4. The kidney phenotype in the mii mutant. (A, B) Histology of medaka kidneys at 3-months posthatching. The tubular lumen was not expanded and was indistinguishable in the wild type (A) and mii mutant (B). gl, glomus; rt, renal tubule. No PKD lesion was found in the mii mutant kidney. Scale bar, 100 μm. (C, D) Transmission electron micrographs of crosssection of renal cilia. In both the wild type (C, n = 5) and mii mutant (D, n = 5), renal cilia have a normal 9 + 2 architecture with clear ODAs (yellow arrowhead). Scale bar, 50 nm.

findings, thus, provide a basic insight into understanding the pathogenesis of LR polarity defects in human DNAI2-defective PCD. It has been known that LR polarity in the node or the KV can influence on bilaterally-symmetric somite formation during later embryogenesis (Kawakami et al., 2005; Meyers and Martin, 1999; Vermot et al., 2005; Vilhais-Neto et al., 2010). We examined if symmetry of segmentation was affected in dnai2a knockdown embryos. Microscopic observation and WISH with myod (a marker for somitogenesis) detected no asymmetrical segmentation at the posterior end of somite stripes in dnai2a knockdown embryos (data not shown and Supplementary Fig. 2). The results indicate that loss of Dnai2 function in the KV does not perturb symmetrical somitogenesis. Loss of Dnai2a in the medaka mii mutant does not cause PKD Active beating of cilia on the renal tubular lumen is required for fluid propulsion and its loss causes PKD in fish (Kramer-Zucker et al., 2005). Since DNAI2 is required for ciliary motility in a variety of human organs having either 9+2 or 9+0 motile cilia (Loges et al., 2008), we speculated that medaka mii mutant would have defects in renal ciliary motility and develop PKD. Omran et al. have shown that medaka kintoun mutant, which is defective in ciliary motility, shows PKD in addition to heterotaxy. Kintoun is a cytoplasmic protein which plays an essential role in assembly of dynein arms to build outer and inner dynein arm complexes in the cilia and sperm flagella. Although the previous report showed that Kintoun interacts with Dnai2, our results revealed that loss of Dnai2a did not lead to defects in the ODA formation and in beating motion of renal cilia; neither mii nor dnai2a-knockdown medaka developed PKD. To resolve this discrepancy, we hypothesized that other molecules might compensate for the loss of Dnai2a in the renal epithelium. Dnai2b is the most probable player to have a complementary role for Dnai2a. Medaka Dnai2b possesses a strong sequence similarity to Dnai2a within the functional domains, including the WD40 repeats, implying functional redundancy of these two molecules. Indeed, the injection dnai2b mRNA into mii mutant embryos rescued the laterality defects. Medaka dnai2b is

expressed in the pronephric and mesonephric tubular epithelia, but not in the KV-lining epithelium. Hence, we hypothesized that ciliary motility defects in the mii mutant renal tubules are masked by the complementary activities of Dnai2b. Consistent with this, inhibition of both Dnai2a and Dnai2b together led to cyst formation in the kidney, as seen in the medaka PKD mutant pc (Hashimoto et al., 2009). The low penetrance of knockdown larvae showing the PKD phenotype might be due to insufficient or transient activities of the antisense morpholino or gripNA oligonucleotides in medaka. However, our data clearly suggest that Dnai2a and Dnai2b function redundantly in the motile cilia in the renal epithelium to suppress the development of a cystic kidney. In addition, our result that the sole inhibition of Dnai2b led to a slight but evident distention of pronephric tubule/duct and glomus suggests that Dnai2b is more important in maintaining the lumen size of pronephros than Dnai2a. Our findings support the hypothesis that the ciliary motility in the renal tubules plays an important role in the formation and maintenance of the kidney in fish. Although some dnai2a- and dani2b-knockdown embryos showed renal cyst formation (Fig. 5), we could not detect apparent abnormalities in the ODA structure in the renal cilia (data not shown). This is possibly due to low penetrance of the ODA defects in the knockdown embryos. Alternatively, subtle changes in the structure and function of the ODA might sufficiently result in the renal cyst formation. Future studies will clarify this issue. Our data revealed the role of Dnai2a and Dnai2b in preventing the kidney from forming cysts. Roles of Dnai2 in other organs Defects in Dnai2 function result in different phenotypic consequences between human and medaka. Mutant medaka survived to adulthood, but displayed a severe wavy malformation in the trunk region, unlike human DNAI2-defective PCD patients (Fig. 1). On the other hand, human DNAI2-defective PCD is characterized by other defects, such as recurrent respiratory tract infection and male infertility, which were not evident in the medaka mii mutant. In particular,

60

Y. Nagao et al. / Developmental Biology 347 (2010) 53–61

Fig. 5. Determination that dnai2b in medaka possesses a complementary activity in KV and kidney. (A) A phylogenic tree of Dnai2 proteins. Full-length amino acid sequences were obtained from the Ensembl Genome Browser or from previous reports (Mitchell and Kang, 1991; Wilkerson et al., 1995), and used for comparison with Clustal W. The result of the Dnai1 and Dnai2 proteins is presented. (B) Homology of two medaka Dnai2 proteins. The homology in the WD40 domain (yellow) and flanking regions (red) is represented as a percentage. (C, D) WISH of dnai2b. (C) Expression of dnai2b mRNA was undetectable in the KV of 2-somite embryos. The dotted circle shows the position of KV. (D) dnai2b mRNA was detected in the renal (pronephric) tubules and ducts of 5 dph larvae. (E–H) Histology of the pronephros of 9 dpf larvae. No notable pronephric cyst was observed in some of the dnai2a- and dnai2b-defective larvae(E). Severe tubular cysts(F), weak tubular cysts(G) or glomic cysts (H) were occasionally detected in the dnai2a- and dnai2b-defective larvae (injected with grip-dnai2a and grip-dnai2b). The arrows indicate the pronephric glomera. The arrowheads indicate normal pronephric tubules. Tubular cysts and glomic cysts are indicated by stars and diamonds, respectively. nc, notochord; Scale bar, 50 μm. (I) The rate of cyst formation by knockdown of dnai2a and/or dnai2b (see Supplementary Table). ++, severely cystic; +, weakly cystic. CO represents the embryos injected with control gripNA. KD represents the antisense oligonucleotide-mediated knockdown. ainjected with gripdnai2b; binjected with grip-dnai2b and MO-dnai2b. MT represents mii mutation. Hyphen (−) represents having unmodified gene function.

although the mii homozygotes did not spawn by natural mating, their sperm exhibited normal motility and were actually capable of fertilizing mii homozygous eggs. Kobayashi et al. identified a spontaneous medaka mutant, jaodori, as having a transposon insertion into the first exon of the dnai2a gene, and this jaodori mutant displayed indistinguishable phenotypes from the mii mutants (personal communication). Thus, the phenotypic difference between humans and medaka may not be due to allelic specificity, but rather, the distinct roles of Dnai2 in different

species. It should be noted, however, that the difference in sperm phenotypes may be attributed to the existence of the dnai2b and its complementary role in sperm, which is unique to medaka. Kobayashi et al., indeed, showed that dnai2b was expressed in medaka testis, suggesting that Dnai2b functions complementarily to Dnai2a in sperm and thereby suppresses the motility defect (personal communication). It was recently reported that ciliary motility is required for formation of bipolar-positioned otoliths in zebrafish (Colantonio et al., 2009). There

Y. Nagao et al. / Developmental Biology 347 (2010) 53–61

was no notable otolith abnormality in medaka mii or dnai2a and dnai2b double-knockdown embryos (Supplementary Fig. 1), although both dnai2a and dnai2b were expressed in the otic vesicles and thus might be involved in generation of cilia movement in the inner-ear. There results imply that Dnai2 is dispensable for in medaka otolith formation. In summary, medaka mii mutant provides a valuable disease model for PCD. Studying the function of Dnai2 in medaka may lead to better understanding of pathogenesis of human PCDs and to the development of therapeutic strategies against human PCD and PKD. Supplementary materials related to this article can be found online at doi:10.1016/j.ydbio.2010.08.001.

Acknowledgments We thank Dr. Daisuke Kobayashi and Dr. Hiroyuki Takeda for sharing unpublished data; Ms. Chikako Inoue for fish care and technical assistance; and the laboratory members for helpful discussions. Pacific Edit reviewed the manuscript prior to submission. Grant Sponsor: Grant-in-Aid for Scientific Research; Grant number 14209009 (Y.W.), 17770190 (H. H.).

References Bisgrove, B.W., Yost, H.J., 2006. The roles of cilia in developmental disorders and disease. Development 133, 4131–4143. Colantonio, J.R., Vermot, J., Wu, D., Langenbacher, A.D., Fraser, S., Chen, J.N., Hill, K.L., 2009. The dynein regulatory complex is required for ciliary motility and otolith biogenesis in the inner ear. Nature 457, 205–209. Essner, J.J., Amack, J.D., Nyholm, M.K., Harris, E.B., Yost, H.J., 2005. Kupffer's vesicle is a ciliated organ of asymmetry in the zebrafish embryo that initiates left–right development of the brain, heart and gut. Development 132, 1247–1260. Essner, J.J., Vogan, K.J., Wagner, M.K., Tabin, C.J., Yost, H.J., Brueckner, M., 2002. Conserved function for embryonic nodal cilia. Nature 418, 37–38. Fedorova, S., Miyamoto, R., Harada, T., Isogai, S., Hashimoto, H., Ozato, K., Wakamatsu, Y., 2008. Renal glomerulogenesis in medaka fish, Oryzias latipes. Dev. Dyn. 237, 2342–2352. Fliegauf, M., Benzing, T., Omran, H., 2007. When cilia go bad: cilia defects and ciliopathies. Nat. Rev. Mol. Cell Biol. 8, 880–893. Furutani-Seiki, M., Sasado, T., Morinaga, C., Suwa, H., Niwa, K., Yoda, H., Deguchi, T., Hirose, Y., Yasuoka, A., Henrich, T., Watanabe, T., Iwanami, N., Kitagawa, D., Saito, K., Asaka, S., Osakada, M., Kunimatsu, S., Momoi, A., Elmasri, H., Winkler, C., Ramialison, M., Loosli, F., Quiring, R., Carl, M., Grabher, C., Winkler, S., Del Bene, F., Shinomiya, A., Kota, Y., Yamanaka, T., Okamoto, Y., Takahashi, K., Todo, T., Abe, K., Takahama, Y., Tanaka, M., Mitani, H., Katada, T., Nishina, H., Nakajima, N., Wittbrodt, J., Kondoh, H., 2004. A systematic genome-wide screen for mutations affecting organogenesis in Medaka, Oryzias latipes. Mech. Dev. 121, 647–658. Hashimoto, H., Miyamoto, R., Watanabe, N., Shiba, D., Ozato, K., Inoue, C., Kubo, Y., Koga, A., Jindo, T., Narita, T., Naruse, K., Ohishi, K., Nogata, K., Shin, I.T., Asakawa, S., Shimizu, N., Miyamoto, T., Mochizuki, T., Yokoyama, T., Hori, H., Takeda, H., Kohara, Y., Wakamatsu, Y., 2009. Polycystic kidney disease in the medaka (Oryzias latipes) pc mutant caused by a mutation in the Gli-Similar3 (glis3) gene. PLoS ONE 4, e6299. Hashimoto, H., Rebagliati, M., Ahmad, N., Muraoka, O., Kurokawa, T., Hibi, M., Suzuki, T., 2004. The Cerberus/Dan-family protein Charon is a negative regulator of Nodal signaling during left–right patterning in zebrafish. Development 131, 1741–1753. Hojo, M., Takashima, S., Kobayashi, D., Sumeragi, A., Shimada, A., Tsukahara, T., Yokoi, H., Narita, T., Jindo, T., Kage, T., Kitagawa, T., Kimura, T., Sekimizu, K., Miyake, A., Setiamarga, D., Murakami, R., Tsuda, S., Ooki, S., Kakihara, K., Naruse, K., Takeda, H., 2007. Right-elevated expression of charon is regulated by fluid flow in medaka Kupffer's vesicle. Dev. Growth Differ. 49, 395–405. Hyodo-Taguchi, Y., 1996. Inbred strains of the medaka, Oryzias latipes. Fish Biol. J. Medaka 8, 11–14. Igarashi, P., Somlo, S., 2007. Polycystic kidney disease. J. Am. Soc. Nephrol. 18, 1371–1373. Ishikawa, Y., 2000. Medakafish as a model system for vertebrate developmental genetics. Bioessays 22, 487–495.

61

Kasahara, M., Naruse, K., Sasaki, S., Nakatani, Y., Qu, W., Ahsan, B., Yamada, T., Nagayasu, Y., Doi, K., Kasai, Y., Jindo, T., Kobayashi, D., Shimada, A., Toyoda, A., Kuroki, Y., Fujiyama, A., Sasaki, T., Shimizu, A., Asakawa, S., Shimizu, N., Hashimoto, S., Yang, J., Lee, Y., Matsushima, K., Sugano, S., Sakaizumi, M., Narita, T., Ohishi, K., Haga, S., Ohta, F., Nomoto, H., Nogata, K., Morishita, T., Endo, T., Shin, I.T., Takeda, H., Morishita, S., Kohara, Y., 2007. The medaka draft genome and insights into vertebrate genome evolution. Nature 447, 714–719. Kawakami, Y., Raya, A., Raya, R.M., Rodriguez-Esteban, C., Belmonte, J.C., 2005. Retinoic acid signalling links left–right asymmetric patterning and bilaterally symmetric somitogenesis in the zebrafish embryo. Nature 435, 165–171. Kimura, T., Jindo, T., Narita, T., Naruse, K., Kobayashi, D., Shin, I.T., Kitagawa, T., Sakaguchi, T., Mitani, H., Shima, A., Kohara, Y., Takeda, H., 2004. Large-scale isolation of ESTs from medaka embryos and its application to medaka developmental genetics. Mech. Dev. 121, 915–932. Kramer-Zucker, A.G., Olale, F., Haycraft, C.J., Yoder, B.K., Schier, A.F., Drummond, I.A., 2005. Cilia-driven fluid flow in the zebrafish pronephros, brain and Kupffer's vesicle is required for normal organogenesis. Development 132, 1907–1921. Levin, M., 2005. Left–right asymmetry in embryonic development: a comprehensive review. Mech. Dev. 122, 3–25. Loges, N.T., Olbrich, H., Fenske, L., Mussaffi, H., Horvath, J., Fliegauf, M., Kuhl, H., Baktai, G., Peterffy, E., Chodhari, R., Chung, E.M., Rutman, A., O'Callaghan, C., Blau, H., Tiszlavicz, L., Voelkel, K., Witt, M., Zietkiewicz, E., Neesen, J., Reinhardt, R., Mitchison, H.M., Omran, H., 2008. DNAI2 mutations cause primary ciliary dyskinesia with defects in the outer dynein arm. Am. J. Hum. Genet. 83, 547–558. Long, S., Ahmad, N., Rebagliati, M., 2003. The zebrafish nodal-related gene southpaw is required for visceral and diencephalic left–right asymmetry. Development 130, 2303–2316. Meyers, E.N., Martin, G.R., 1999. Differences in left–right axis pathways in mouse and chick: functions of FGF8 and SHH. Science 285, 403–406. Mitchell, D.R., Kang, Y., 1991. Identification of oda6 as a Chlamydomonas dynein mutant by rescue with the wild-type gene. J. Cell Biol. 113, 835–842. Nakamura, T., Mine, N., Nakaguchi, E., Mochizuki, A., Yamamoto, M., Yashiro, K., Meno, C., Hamada, H., 2006. Generation of robust left–right asymmetry in the mouse embryo requires a self-enhancement and lateral-inhibition system. Dev. Cell 11, 495–504. Okada, Y., Takeda, S., Tanaka, Y., Belmonte, J.C., Hirokawa, N., 2005. Mechanism of nodal flow: a conserved symmetry breaking event in left–right axis determination. Cell 121, 633–644. Omran, H., Kobayashi, D., Olbrich, H., Tsukahara, T., Loges, N.T., Hagiwara, H., Zhang, Q., Leblond, G., O'Toole, E., Hara, C., Mizuno, H., Kawano, H., Fliegauf, M., Yagi, T., Koshida, S., Miyawaki, A., Zentgraf, H., Seithe, H., Reinhardt, R., Watanabe, Y., Kamiya, R., Mitchell, D.R., Takeda, H., 2008. Ktu/PF13 is required for cytoplasmic pre-assembly of axonemal dyneins. Nature 456, 611–616. Pennarun, G., Chapelin, C., Escudier, E., Bridoux, A.M., Dastot, F., Cacheux, V., Goossens, M., Amselem, S., Duriez, B., 2000. The human dynein intermediate chain 2 gene (DNAI2): cloning, mapping, expression pattern, and evaluation as a candidate for primary ciliary dyskinesia. Hum. Genet. 107, 642–649. Satir, P., Christensen, S.T., 2007. Overview of structure and function of mammalian cilia. Annu. Rev. Physiol. 69, 377–400. Soroldoni, D., Bajoghli, B., Aghaallaei, N., Czerny, T., 2007. Dynamic expression pattern of Nodal-related genes during left–right development in medaka. Gene Expr. Patterns 7, 93–101. Vermot, J., Gallego Llamas, J., Fraulob, V., Niederreither, K., Chambon, P., Dolle, P., 2005. Retinoic acid controls the bilateral symmetry of somite formation in the mouse embryo. Science 308, 563–566. Vilhais-Neto, G.C., Maruhashi, M., Smith, K.T., Vasseur-Cognet, M., Peterson, A.S., Workman, J.L., Pourquie, O., 2010. Rere controls retinoic acid signalling and somite bilateral symmetry. Nature 463, 953–957. Wagner, M.K., Yost, H.J., 2000. Left–right development: the roles of nodal cilia. Curr. Biol. 10, R149–R151. Wakamatsu, Y., Pristyazhnyuk, S., Kinoshita, M., Tanaka, M., Ozato, K., 2001. The seethrough medaka: a fish model that is transparent throughout life. Proc. Natl Acad. Sci. USA 98, 10046–10050. Wilkerson, C.G., King, S.M., Koutoulis, A., Pazour, G.J., Witman, G.B., 1995. The 78, 000 M (r) intermediate chain of Chlamydomonas outer arm dynein isa WD-repeat protein required for arm assembly. J. Cell Biol. 129, 169–178. Wittbrodt, J., Shima, A., Schartl, M., 2002. Medaka–a model organism from the far East. Nat. Rev. Genet. 3, 53–64. Yoder, B.K., 2007. Role of primary cilia in the pathogenesis of polycystic kidney disease. J. Am. Soc. Nephrol. 18, 1381–1388. Zariwala, M.A., Knowles, M.R., Omran, H., 2007. Genetic defects in ciliary structure and function. Annu. Rev. Physiol. 69, 423–450.