BASIC RESEARCH www.jasn.org

LRRC50, a Conserved Ciliary Implicated in Polycystic Kidney Disease

Ellen van Rooijen,*† Rachel H. Giles,† Emile E. Voest,† Carina van Rooijen,* Stefan Schulte-Merker,* and Freek J. van Eeden*

*Hubrecht Institute, Developmental Biology and Stem Cell Research, and †Department of Medical Oncology, University Medical Center Utrecht, Utrecht, Netherlands

ABSTRACT Cilia perform essential motile and sensory functions central to many developmental and physiological processes. Disruption of their structure or function can have profound phenotypic consequences, and has been linked to left-right patterning and polycystic kidney disease. In a forward genetic screen for mutations affecting ciliary motility, we isolated zebrafish mutant hu255H. The mutation was found to disrupt an ortholog of the uncharacterized highly conserved human SDS22-like leucine-rich repeat (LRR)-containing protein LRRC50 (16q24.1) and Chlamydomonas Oda7p. Zebrafish lrrc50 is specifically expressed in all ciliated tissues. lrrc50hu255H mutants develop pronephric cysts with an increased proliferative index, severely reduced brush border, and disorganized pronephric cilia manifesting im- paired localized fluid flow consistent with ciliary dysfunction. Electron microscopy analysis revealed ultrastructural irregularities of the arms and misalignments of the outer-doublet on the ciliary , suggesting instability of the ciliary architecture in lrrc50hu255H mutants. The SDS22-like leucine-rich repeats present in Lrrc50 are necessary for proper protein function, since injection of a deletion construct of the first LRR did not rescue the zebrafish mutant phenotype. Subcellular distribution of human LRRC50-EGFP in MDCK and HEK293T cells is diffusely cytoplasmic and concentrated at the mitotic spindle poles and . LRRC50 RNAi knock-down in human proximal tubule HK-2 cells thoroughly recapitulated the zebrafish brush border and cilia phenotype, suggesting conservation of LRRC50 function between both species. In summary, we present the first genetic vertebrate model for lrrc50 function and propose LRRC50 to be a novel candidate for human cystic kidney disease, involved in regulation of -based cilia and actin-based brush border microvilli.

J Am Soc Nephrol 19: 1128–1138, 2008. doi: 10.1681/ASN.2007080917

Cilia are highly complex evolutionarily conserved fluence ciliary beat frequency, IDAs are involved in microtubule-based cellular extensions that perform generation of the ciliary waveform.2,3 essential motile and sensory functions central to Immotile or misassembled cilia have been impli- many developmental and physiological processes.1 cated in many pathologies, including left-right pat- They are nucleated by the centrosome-derived terning defects and polycystic kidney disease and comprise a 9 ϩ 2or9ϩ 0 microtu- bule doublet backbone or . Ciliary motility Received August 20, 2007. Accepted December 31, 2007. depends on the presence of inner (IDAs) and outer dynein arms (ODAs) attached to the peripheral mi- Published online ahead of print. Publication date available at www.jasn.org. crotubule doublets. These dynein arms are multi- subunit protein complexes with ATPase activity Correspondence: Dr. Freek van Eeden, Department of Biomed- ical Science, Sheffield University, Firth Court, Western Banks, that promote sliding between adjacent microtu- Sheffield S10 2TN, UK. Phone: ϩ44-0-1142-222348; Fax: ϩ44-0- bules, and their coordinated activation and inacti- 1142-2765413; E-mail: [email protected] vation generates a ciliary wave. Whereas ODAs in- Copyright ᮊ 2008 by the American Society of Nephrology

1128 ISSN : 1046-6673/1906-1128 J Am Soc Nephrol 19: 1128–1138, 2008 www.jasn.org BASIC RESEARCH

(PKD).1,4–6 Many involved in PKD localize to the liferation in the pronephric tubule (16 of 18 animals) where the basal body, centrosome, and/or renal cilia, where they are in- first cystic expansion occurs (Supplemental Figure 2, B and C). volved in a wide variety of processes including modulating As development progressed, increased proliferation became signaling pathways, cell-cycle control, and planar cell polarity more pronounced and was detected in both lrrc50hu255HϪ/Ϫ (for reviews see references7–10). Cilia are now perceived as im- pronephric tubules and ducts (35 of 35; Figure 1D). These data portant cellular antennae, and mechanisms underlying cilio- indicate pronephric proliferation to be closely associated with genesis and cilia maintenance have recently become a major pronephric cyst development in lrrc50hu255H mutants. Further focus of research, including systematic bioinformatic screens ultrastructural analyses revealed a severely reduced brush bor- to define a comprehensive ciliary proteome or ciliome.4,11–16 der with shorter microvilli in posterior pronephric ducts of 4 In search for novel essential for ciliary motility, we dpf mutants (Figure 1E). Because prominent irregularities of performed a forward genetic N-ethyl-N-nitrosourea (ENU) the pronephric duct brush border microvilli were already ob- screen in zebrafish. Here, we report the isolation of a mutant served in 2 dpf lrrc50hu255H mutants without visible tubular with ciliary dyskinesia that develops proliferative kidney cysts. cysts and in 3 dpf mutants with tubular cysts but without vis- The causative gene encodes the highly conserved leucine-rich ible dilation of the pronephric duct (Supplemental Figure 3), repeat (LRR)-containing protein Lrrc50. Recently, Mitchell we suggest this to represent a primary defect. Homozygous and co-workers17 identified the Chlamydomonas oda7 locus to mutant embryos ultimately die during early larval stages, at encode an lrrc50 ortholog. Oda7p is an axonemal dynein-asso- approximately 8 dpf, as a result of severe edema that is likely to ciated protein, and its loss of function results in the absence of be a consequence of pronephric disease progression. outer row and a reduced flagellar beat frequency. We report the first functional characterization of lrrc50 in both lrrc50hu255H Is Required for Cilia Function in Zebrafish zebrafish and mammalian cell culture systems. Our data sug- Motile cilia in Kupffer’s vesicle (KV) are essential for breaking gest LRRC50 to be a ciliary component in vertebrates. We pro- bilateral symmetry in zebrafish, and in this respect KV is sim- pose LRRC50 to be a novel candidate gene for human cystic ilar to the mouse node.18–21 In lrrc50hu255H mutants, ciliary kidney disease, involved in the regulation of apical cell special- motility in KV was not observed (data not shown). Although izations, including motile cilia and actin-based microvilli. irregularities in ciliary distribution in KV are found (7 of 9), cilia number and length are not strongly affected (Supplemental Figure 4A). To determine whether loss of cilia- RESULTS dependent fluid flow in KV resulted in laterality defects, we performed in situ hybridization analysis for the first asym- To identify recessive mutations affecting ciliary motility, we metrically expressed nodal-related gene southpaw (spaw).22,23 performed a forward genetic ENU mutagenesis screen in ze- We observed randomized expression in lrrc50hu255HϪ/Ϫ at the brafish. F3-generation embryos were screened, and mutant 18 somite stage (Supplemental Figure 4B). In zebrafish, the hu255H (hereafter called lrrc50hu255H) was isolated, based on first morphologic marker for a break in symmetry is the left- the absence of motile cilia in the nose and neural tube (Supple- ward movement of the heart, called jogging. lrrc50hu255H mu- mental Figure 1). tants display a randomized left-right polarity of the heart, as illustrated by altered heart jogging in somewhat more than half lrrc50hu255H Mutants Develop Kidney Cysts and Brush of the embryos (Figure 2A). Sibling embryos manifested a very Border Abnormalities low level of abnormal jogging (7 [1.8%] of 398). Similarly, lrrc50hu255H mutant embryos manifest a pronounced ventral polarity of the visceral organs (liver, pancreas, and gut) is af- body curvature. Approximately 2.5 d postfertilization (dpf), fected in lrrc50hu255HϪ/Ϫ (Supplemental Figure 4C). In hu- fluid-filled cysts became apparent in the pronephric tubule mans and mice, organ left-right polarity disturbances (situs region. As development progressed, pronephric tubular cysts inversus) have been strongly associated with impaired cilia expanded and bulged out behind the pectoral fin, and the pro- function in early embryogenesis,10,24 which support a pro- nephric duct became dilated (Figure 1, A and B). Histologic found cilia defect in the lrrc50hu255H mutants. sections of the mutant pronephros showed the glomerulus to Cilia in lrrc50hu255H mutants appear morphologically differ- be stretched at the midline and glomerular cells to be enlarged ent in organization and distribution, as shown by confocal (Figure 1C, insert). The cuboidal epithelium seen in the wild- analysis of cilia in the anterior and posterior pronephric duct type pronephric tubule was completely flattened (Figure 1C). of 52 hpf embryos immunostained with ␣-acetylated ␣-tubu- To examine whether the observed pronephric phenotypes cor- lin (green) and DAPI (blue; Figure 2B). Because of the densely respond to increased localized proliferation, we performed a packed cilia in the wild-type pronephros, we were unable to time-course bromodeoxyuridine (BrdU) incorporation assay. quantify this defect confidently. We examined the lateral line Before visible cyst development, no obvious difference in pro- organs and found kinocilium number (with median values of liferation rate was observed between lrrc50hu255H mutants and 12.1 versus 8.8 in lrrc50hu255H) and length (median 17.87 versus siblings at 50 h postfertilization (hpf; Supplemental Figure 12.22 ␮minlrrc50hu255H) to be reduced in 7 dpf lrrc50hu255H 2A). In 3-d-old mutants, however, we observed increased pro- mutants (Figure 2C).

J Am Soc Nephrol 19: 1128–1138, 2008 LRRC50 Implicated in PKD 1129 BASIC RESEARCH www.jasn.org

lrrc50-/- PT cyst A sibling PT cyst lrrc50-/- B sibling 3G8 dilated PD PD

PT A P lrrc50-/- 54 hpf 54 hpf sibling PD

PT

6.5 dpf 7 dpf 7 dpf

C sibling lrrc50-/- H&E I

D sibling lrrc50-/- II G II PT I PT G PD

PT PD PD 6.5 dpf BrdU

6.5 dpf 16x 16x

E sibling lrrc50-/-

brush border

4 dpf posterior pronephric duct 5000x posterior pronephric duct 5000x

Figure 1. lrrc50 is involved in the development of cystic kidney disease and brush border abnormalities in zebrafish. (A) lrrc50hu255H mutants are characterized by a curved body axis, the development of pronephric tubular (PT) cysts, and dilated pronephric ducts (PD), as shown at 54 hpf and 7 dpf. (B) Confocal imaging with Cy3-labeled 3G8 (green), staining the lateral part of the PT and the anterior Ϫ Ϫ half of the PD in 6.5 dpf embryos. (C) Cross-sections (hematoxylin and eosin) of a wild-type and lrrc50hu255H / pronephros at 6.5 dpf. Mutants show a grossly distended cyst in place of the pronephric tubule (black arrow), glomerular defects, and dilated pronephric ducts. An enlargement of the glomerulus of a wild-type fish (detail insert I) and a lrrc50hu255H mutant glomerulus with dilated cells (black arrowheads; detail insert II) are depicted in the right panels. (D) BrdU incorporation demonstrates increased proliferation in the pronephric tubule and pronephric duct in lrrc50hu255H mutants at 6.5 dpf. (E) Ultrastructural analysis reveals a severe reduction of the brush border (vertical arrows) in mutant posterior pronephric duct cells at 4 dpf. A, anterior; P, posterior.

Ciliary motility in the lrrc50hu255H Ϫ/Ϫ pronephros was not These data collectively indicate that lrrc50hu255H Ϫ/Ϫ cilia de- observed (data not shown). An established functional assay of fects, as observed in the pronephric ducts and lateral line or- ciliary motility measures pronephric fluid flow in a dye excre- gans, impair normal ciliary function, resulting in ciliary immo- tion test.21,25 Tetramethylrhodamine conjugated dextran tility, left-right polarity defects, a severe pronephric fluid flow (TAMRA) was injected into the heart of 4.5 dpf embryos, and deficiency, and the development of pronephric cysts. excretion of fluorescent dye at the cloaca (end of urinary tract) was monitored. In wild-type animals, TAMRA was filtered by Positional Cloning and Expression of lrrc50hu255H the glomerulus and excreted via the pronephric ducts at the Linkage analysis using 622 meioses positioned the responsible cloaca, generally within 5 min (75%; n ϭ 15) or between 5 and mutation on 7 between single-sequence length 10 min (20%; n ϭ 4) after injection as a result of the cloaca- polymorphism markers z15270 (39 cM) and z8693 (41.5 cM), directed ciliary beat pattern; however, dye excretion in mutant on the Zv4 zebrafish genome assembly. Using this assembly, we embryos was markedly slower (33.3% in 10 to 20 min; n ϭ 6) analyzed further markers until a region that did not show any or completely absent (61.1% Ͼ30 min; n ϭ 11; Figure 2D, recombinations in our mapping panel was identified. Exons white arrows). Glomerular filtration did not seem to be af- from this region were sequenced, and a T/A mutation was fected, because TAMRA dye accumulation was observed in identified, changing a conserved leucine into a stop codon both the pronephric tubule and anterior pronephric duct. (L88X) in a novel Ensembl gene zgc::56169 (Figure 3, A and B).

1130 Journal of the American Society of Nephrology J Am Soc Nephrol 19: 1128–1138, 2008 www.jasn.org BASIC RESEARCH

Figure 2. lrrc50 is required for cilia function in zebrafish. (A) Dysfunctional cilia in KV are implicated in left-right polarity defects, characterized by random heart jogging in lrrc50 mutants. Absolute numbers of embryos scored are depicted in the bars; the bottom panel displays representative frontal images of the different heart jogging phenotypes. (B) Confocal analysis of cilia in the anterior (left) and posterior pronephric duct (right) immunostained with anti-acetylated ␣-tubulin (green) and DAPI (blue). Cilia in lrrc50hu255H mutants appear morphologically disorganized at 52 hpf. *Motor axons, which also stain for acetylated tubulin. (C) Scanning electron micrographs of the lateral line organs (LLO) show a reduction in kinocilia number and length in 7 dpf lrrc50hu255H mutants. (D) For investigation of the fluid excretion via the pronephros, 4.5 dpf embryos were administered an injection of TAMRA. Whereas in wild-type siblings the first TAMRA excretion via the cloaca (*) can be observed within 5 min, in mutants, time to excretion is markedly slower (mutant 1) or completely absent (mutant 2). White arrowheads indicate final position TAMRA. Glomerular filtration does not seem to be affected, since TAMRA dye accumulation can be seen in both the pronephric tubule and anterior pronephric duct. T, time after injection.

Zgc::56169 encodes an ortholog of the uncharacterized human tion of the injected mRNA in lrrc50hu255h mutants and siblings LRR-containing protein LRRC50 of unknown function. LRRs (Figure 3H). are short motifs (20 to 30 residues) thought to provide a struc- The expression profile of zebrafish lrrc50 was determined by tural framework for the formation of protein–protein interac- whole-mount in situ hybridization. At gastrula stages, lrrc50 tions.26,27 The putative LRRs of Lrrc50 are highly conserved was highly expressed in dorsal forerunner cells (Figure 4A), from Chlamydomonas (oda7)toDrosophila and vertebrates which aggregate to form the ciliated KV.19 At the 12-somite (Figure 3C), and their orthology is further supported by phy- stage (14 hpf), expression was restricted to the otic placode, logenetic analysis.17 pronephric duct primordia, and floorplate (Figure 4B). At 20 Injection of 320-pg full-length lrrc50-EGFP mRNA into hpf, lrrc50 was expressed in the pronephric duct, neural tube, one-cell stage lrrc50hu255H embryos rescued the mutant pheno- and nose and diffusely in the brain (Figure 4C), and expression type: Cilia were motile, the body axis was straight, and pro- became restricted to the nose and lateral line organs at 72 hpf nephric cysts did not develop before 4 dpf (Figure 3D). This (Figure 4D). Thus, lrrc50 is expressed in ciliated tissues experiment confirms the lrrc50 L88X mutation to cause the throughout zebrafish development. In lrrc50hu255H mutants observed phenotype. To assess whether the conserved LRRs are lrrc50 mRNA could still be detected (data not shown). necessary for proper protein function, we generated a con- struct deleting the LxxLxL-consensus sequence of the first LRR Characterization and Expression of Human LRRC50 (Figure 3, E and F). Injection of 300-pg ⌬LRR1-lrrc50-EGFP Human LRRC50 is an undescribed gene located on chromo- mRNA did not rescue the mutant phenotype (Figure 3G), some band 16q24.1, and its two predicted transcripts en- whereas Western blot analysis for GFP showed proper transla- code proteins of 637 and 725 amino acids (Figure 5A). Both

J Am Soc Nephrol 19: 1128–1138, 2008 LRRC50 Implicated in PKD 1131 BASIC RESEARCH www.jasn.org

zebrafish lrrc50 (zgc::56169) isoforms feature six N-terminal LRR 0 39 Mc A 72 M 51Z c 6 5. 93 8 1 Z 4 motifs, an LRRcap, a coiled-coil, and a Chr 7 nonconserved proline-rich domain (based ATG hu255H TAA 4.93 Kb on predictions Uniprot/SWISSPROT 28 hu255H LRR 1 LRR 2 LRR 3 [Q8NEP3], PROSITE MotifScan, and zebrafish CIEGLEEYTGLRCLWLECNGIRKIENLENQTELRCLFLHQNLIHTLENLEPLSKLCTLNV 143 C 29 human RIENLEEYTGLRCLWLQSNGIQKIENLEAQTELRCLFLQMNLLRKIENLEPLQKLDALNL 180 SMART5 ; Figure 5B). The first three leu mouse RIENLEEYTGLRCLWLECNGIQRIENLQAQSELRCLFLQVNLLHKIENLEPLQKLDALNL 174 rat RIENLEEYTGLRCLWLECNGIQRIENLQAQSELRCLFLQVNLLHKIENLEPLQKLDALNL 174 B chlamydomonas QIACLEDYVNLKALFLEGNVLETLEGLPPLADLKCLYVQQNCIWKISGLEAVPGLDTLNI 97 LRRs in LRRC50 showed high homology fruitfly CIESLEEYTELKCLWLECNAISEIQGLEKLSKLKCLFLQNNLITKIENLDPCRELDTLNL 107 * **:*. *:.*:*: * : ::.* :.*:**::: * : .:..*:. * :**: to the LRR_SDS22-like subfamily (Fig- LRR 4 LRR 5 wild type zebrafish SNNYIKVIENISS--LSDLSTLQISHNTLENVCDMEELSHCPSISVLDLSHNRISDPALV 201 human SNNYIKTIENLSC--LPVLNTLQMAHNHLETVEDIQHLQECLRLCVLDLSHNKLSDPEIL 238 ure 5C). The sixth LRR (residues 244 to GGGTTTGGAAG mouse SNNYIKTIENLSC--LPVLNTLQMAHNRLETVADIEHLRECLRLCVLDLSHNALSDPEIL 232 rat SNNYIKTIENLSC--LPVLNTLQMAHNRLETVADIEHLRECLQLCVLDLSHNSLSDPEIL 232 270) partially overlapped with the puta- stop chlamydomonas SNNQLTKLEGLAC--CPALRTLIATHNHLVTLDSVAHLAECKALQTLDLQNNELEDPGIV 155 fruitfly SSNHIRKIQNIGTNVLPVLNTLTISSNYLKDSESLSDLIQCKTLSVLDLSNNRIDDILIV 167 *.* : ::.:. . * ** : * * .: .* .* : .***.:* :.* :: tive LRRcap (residues 261 to 279), a mo- LRR 6 zebrafish NILEKMPDLRVLNLMGNEVIKKIPNYRKTLIVRLKQLTYLDDRPVFPKDRACAEAWA.// 558 tif occurring C-terminally to LRR in lrrc50-/- human SILESMPDLRVLNLMGNPVIRQIPNYRRTVTVRLKHLTYLDDRPVFPKDRACAEAWA.// 725 GGGTTA GGAAG mouse SVLESMPCLRVLNLMGNPVTKHIPNYRRTVTVRLKHLTYLDDRPVFPKDRACAEAWA.// 634 “SDS22-like” and “typical” LRR-con- rat SVLETMPCLRVLNLMGNPVTKHIPNYRRTVTVRLKHLTYLDDRPVFPKDRACAEAWA.// 633 chlamydomonas DILKQIPDLRCLYLKGNPVVSNIKNYRKVLVTSIPSLTYLDDRPVFDNERKIAQAWL.// 432 30,31 fruitfly KIFEQMLNLKVLVLQGNPVVSRLPQYRKTLILACKELTYLDSRPVFPRDRACAEAWK.// 1483 taining proteins. .::: : *: * * ** * .: :**:.: *****.**** .:* *:** To examine normal subcellular local- ization, we fused the human LRRC50 to D lrrc50-EGFP mRNA injection (320pg) body axis cilia motility cysts (3dpf) EGFP and examined the localization in straight curved + +/- - + - living and fixed 293T and MDCK kidney sibling 100% (n=125) - 100% (125) - - - 100% (125) cells. Distribution of the LRRC50-EGFP lrrc50-/- 94.7% (36) 5.3% (2) 89.5% (34) 5.3% (2) 5.3% (2) 7.9% (3) 92.1% (35) construct was diffusely cytoplasmic, with bright foci at the spindle poles of mitotic 0

LRR1 I lrrc5 cells, as seen by co-localization with

LRR1 II f d z n

E Hi F ␥ 514 spindle pole component -tubulin (Fig- zf lrrc50 ndIII Hi LRR1 EGFP ure 5D). In 28 of 44 ciliated MDCK cells 94 99 557 468 448 LxxLxL STOP scored, LRRC50-EGFP localized to the HindIII ciliary structure extending from the basal body immunostained with ␥-tubulin is G H xis (Figure 5E). LRR1-lrrc50-EGFP urved a LRR1 300 pg LRR1 300 pg c straight ax mRNA injection (300pg) uninjected

curved LRRC50 Is Required for Brush 25.2% anti-GFP 95 kDa (n = 40) Border and Cilia Regulation in anti-ILK 59 kDa straight Human Kidney Cells 74.8% (n = 119) To address the role of LRRC50 in a mam- malian cell system, we transfected nor-

sibling (straight body axis, no cysts) mal human proximal tubule (HK-2) cells

lrrc50-/- (curved body axis, cyst development) with either of two RNAi constructs (RNAi-1 and RNAi-2). Reverse tran- Figure 3. Genomic organization and expression pattern of zebrafish lrrc50. (A) Genomic scriptase–PCR confirmed both RNAi’s hu255H organization of zebrafish lrrc50. lrrc50 was mapped to chromosome 7, flanked by to completely knock down levels of the single-sequence length polymorphism markers z15270 (39 cM) and z8693 (41.5 cM) on gene product as compared with mock the MGH mapping panel (http://www.zfin.org). Exons are represented as black boxes. (empty vector)-transfected HK-2 cells Sequencing revealed a mutation in exon 4 of lrrc50, highlighted in red. (B) In lrrc50hu255H (Figure 6A). Interphase cells were immu- mutants, a T/A mutation is introduced, changing a conserved leucine into a premature stop codon. (C) ClustalW alignment shows Lrrc50 to be a highly conserved protein from nostained for ezrin, a cytoskeletal pro- Chlamydomonas to human (for accession numbers, see Supplemental Figure 5). LRRs are tein that anchors actin polymers to the outlined and lrrc50hu255H L88X is highlighted in red. (D) The mutant phenotype is rescued plasma membrane and major compo- by injection of 320 pg lrrc50-EGFP mRNA, as scored for body curvature, cilia motility, and nent of cellular brush borders.32 Similar pronephric cyst formation after 3 dpf. Absolute numbers of animals scored are in to zebrafish lrrc50 mutants, both parentheses. (E) Schematic representation of the ⌬LRR1-lrrc50-EGFP deletion construct LRRC50 RNAi constructs severely re- in which the consensus LxxLxL sequence (nucleotides 391 to 408) of the first LRR is duced the brush border at the apical cell removed in frame. (F) HindIII digestion on plasmid DNA shows a reduced band in the surface of polarized HK-2 cells (Figure ⌬ deletion construct as compared with full-length lrrc50-EGFP. (G) LRR1-lrrc50-EGFP 6B). Co-staining with zona occulans-1 mRNA is unable to rescue the mutant phenotype; the Mendelian sibling-to-mutant ratio (ZO-1) did not reveal gross alterations in is maintained. (H) Western blot analysis on ⌬LRR1-lrrc50-EGFP–injected embryos with apical tight junctions. ␣-GFP shows that despite robust expression, the ⌬LRR1-lrrc50-EGFP deletion construct cannot rescue the lrrc50hu255H phenotype. Reducing LRRC50 levels in HK-2 cells induced a reduction in ciliated inter-

1132 Journal of the American Society of Nephrology J Am Soc Nephrol 19: 1128–1138, 2008 www.jasn.org BASIC RESEARCH

DISCUSSION A B otic placode floor plate In this study, we present the first func- tional characterization of zebrafish and dorsal pronephric duct forerunner primordium cells human LRRC50, a recently reported or- lateral dorsal tholog of the axonemal dynein-associ- 90% epiboly 12 somites ated Oda7 protein in Chlamydomonas. C D LRRC50 was identified as a putative cili- neural tube lateral line organs ary protein in two independent bioinfor- matic studies.11,12 Reminiscent of the lo- pronephric duct nose calization of other PKD-associated proteins (for a comprehensive overview, see reference10), we showed localization nose 20 hpf 72 hpf of human LRRC50-EGFP at the mitotic Figure 4. Zebrafish lrrc50 is expressed in all ciliated tissues. Whole-mount in situ spindle poles and cilium in human and hybridization displays expression at 90% epiboly in the dorsal forerunner cells, which dog kidney cells, suggesting LRRC50 to aggregate to form the ciliated KV (A); 12 somites (14 hpf) in the otic placode, pronephric be a ciliary component in vertebrates. duct primordia, and floor plate (B); 20 hpf in the pronephric duct, neural tube, nose, and Oda7 mutants lack ODAs normally diffusely in the brain (C); and 72 hpf in nose and lateral line organs (D). present on the outer-doublet microtu- bules of ciliary axonemes and display a reduced flagellar beat frequency.17,33 The phase cells, from 70% (Ϯ 2.3) in mock-transfected to 38.7% zebrafish lrrc50hu255HϪ/Ϫ ciliary phenotype is more severe, in- (Ϯ 2) in RNAi-2–transfected cells, as visualized by ␣-acety- cluding ciliary dyskinesia, ODA defects, and additional dynein lated ␣-tubulin immunostaining (Figure 6C). Moreover, in arm irregularities and misalignments of outer-microtubule cells that did form a cilium, the ciliary length was reduced from doublets suggestive of architectural instability. Structures on 8.43 (Ϯ 0.43) to 2.74 ␮m(Ϯ 0.29) in mock- and RNAi-2– the inner surface of microtubules have been suggested to in- transfected cells, respectively. RNAi-1 produced similar results crease their stability.34 Consistently, Oda7p regulates protozo- (data not shown). an-specific dynein heavy-chain ␣ stability and resides in a dou- blet-associated complex that interacts with both outer row and Lrrc50 Ultrastructural Cilia Defects I1 inner row dyneins,17 possibly in the recently identified out- Because our data suggest a ciliary function for Lrrc50, we er-inner dynein complex.34 Furthermore, Oda7p is localized to performed electron microscopy of pronephric lrrc50hu255H the , where it regulates preassembly of the outer-row mutant cilia. We observed that these cilia showed ultra- dynein complex.33 Cytoplasmic distribution of LRRC50-EGFP structural irregularities. Most mutant cilia lack the ODA might suggest a conserved role for vertebrate LRRC50 in this normally present on the outer-doublet microtubules of cil- sense. iary axonemes (Figure 7A); however, some axonemes were Motile cilia in the zebrafish pronephros (9 ϩ 2), KV (9 ϩ 2), observed to completely lack all dynein arms or to have mis- neural tube (9 ϩ 0) and teleost lateral line sensory kinocilia (9 placed IDA. ϩ 2) contain dynein arms on their ciliary axonemes21,35; how- In addition, we observed increased misalignments of outer ever, it is debated whether immotile mammalian renal mono- microtubule doublets in mutant ciliary axonemes. In a wild- cilia (9 ϩ 0) do. A pioneer study on the ultrastructure of hu- type cilium, outer microtubule doublets were aligned in a circle man and rat renal monocilia clearly showed a dynein arm–like (or ellipse, depending on the cilium angle at sectioning) structure pointing outward from the microtubule doublets,36 around the central pair (for schematic representation, see Fig- whereas in mouse, no dynein arms were observed.37 The liter- ure 7B). By determining the tangent angle of the microtubule ature remains elusive on this issue, and, as Iban˜ez-Tallon et al.1 doublet with respect to the peripheral circle as a whole, we stated, electron microscopy has not yet verified whether dy- observed that most (88.7%) phenotypically normal siblings nein arms are actually present in human renal monocilia. displayed outer doublets aligned ina0to3°angle on the ax- Nonetheless, we report conserved LRRC50 function in the reg- oneme. On the contrary, in lrrc50hu255H mutants, this number ulation of microtubule-based cilia and actin-based brush bor- was decreased two-fold (43.5%; Figure 7C). We observed out- der microvilli between zebrafish and humans. It is surprising er-doublet misalignments in 56.5% of the measured doublets that an axonemal dynein-associated protein might underlie (n ϭ 52), with angles ranging from 5 up to 75°. Although no the regulation of both types of apical structures, and we there- obvious correlation between the severity of doublet misalign- fore speculate LRRC50 might have additional and/or con- ment and nature of the ultrastructural irregularity could be verged functions in vertebrates. Interestingly, made, these data suggest instability of the ciliary architecture in function was recently found to be regulated by both microtu- lrrc50hu255H mutants. bular organization involving IFT component KIF3A16 and the

J Am Soc Nephrol 19: 1128–1138, 2008 LRRC50 Implicated in PKD 1133 BASIC RESEARCH www.jasn.org

Ϫ Ϫ human LRRC50 3 dpf. Even though lrrc50hu255H / glo- A 16q24.1 centromere telomere merular cells are enlarged, glomerular

ATG 1 ATG 2 TAA 32.63 Kb filtration did not seem to be affected, since TAMRA dye accumulation could B M88 be seen in both the pronephric tubule hsLRRC50 and anterior pronephric duct. Glomer- 1 128 279 313 343 391 513 725 ular filtration pressure might contrib- LRR_SDS22 LRRcap proline rich LRR coiled-coil ute to the observed residual pronephric fluid flow. C LRR_SDS22-like LXXLXXLXLXXNXIXXIXXLXX LRR cons. segment LXXLXLXXNXL In mice and humans, it is hypothe- hsLRR 1 (128-149) YTGLRCLWLQSNGIQKIENLEA hsLRR 4 (195-218) LPVLNTLQMAHNHLETVEDIQHLQE hsLRR 2 (150-171) QTELRCLFLQMNLLRKIENLEP hsLRR 5 (219-243) CLRLCVLDLSHNKLSDPEILSILES sized that cyst-lining kidney epithelial hsLRR 3 (172-193) LQKLDALNLSNNYIKTIENLSC hsLRR 6 (244-270) MPDLRVLNLMGNPVIRQIPNYRRTVTV * .* *. * ::.****. * .*:: * : hsLRRcap (261-279) IPNYRRTVTVRLKHLTYLD cells lose their ability to sense mechanical fluid flow, which may result in dediffer- 293T 293T γ-tubulin LRRC50-EGFP overlay 293T ention and hyperproliferation of mutant D DAPI cells.42,43 We showed pronephric hyper- proliferation to be closely associated with, yet never preceding, cystic disease Ϫ Ϫ centrosome hu255H / spindle pole progression in lrrc50 zebrafish embryos, indicating this to be a second- E MDCK γ-tubulin LRRC50-EGFP overlay DAPI ary defect. cilium LRRC50 seems to be the newest addi- tion to an emerging pattern linking pro- basal body teins containing the subfamily SDS22- like LRRs to flagellar/ciliary function. Other members include the Chlamydo- Figure 5. Genomic organization and subcellular localization of human LRRC50. (A) monas dynein light chain 1, involved in Genomic organization of human LRRC50. A putative alternative translation start site at protein–protein interactions in motor methionine 88 is in exon 3 (ATG2). (B) Schematic representation of human LRRC50 complexes44,45; the Ciona flagellar radial showing the locations of six putative N-terminal LRRs, an LRRcap partially overlapping spoke protein LRR3746; and Chlamydo- with the sixth LRR, a coiled coil, and a nonconserved proline-rich domain. Residue monas VFL1 protein, which establishes numbers below the bar mark putative domain boundaries. (C) hsLRR 1 to 3 show close the correct rotational orientation of basal homology to the LRR_SDS22-like consensus motif, whereas hsLRR 4 to 6 contain only the 47 conserved consensus LRR sequence. (D) Subcellular localization of LRRC50 was deter- bodies. Trypanosomal TbLRTP was mined by transfection of 293T cells with a full-length LRRC50-EGFP construct. Distribu- shown to suppress basal body replication tion of LRRC50-EGFP (green) appears diffusely cytoplasmic with bright spots at the and flagellar biogenesis and to play a crit- spindle poles of mitotic cells as shown by co-staining with ␥-tubulin (red). (E) LRRC50- ical role in cell-cycle control.48 The ze- EGFP was observed to localize to the ciliary structure extending from the basal body in brafish TbLRTP homolog seahorse MDCK cells but never to the basal body itself as determined by co-staining with ␣-␥- (lrrc6l) develops pronephric cysts.5,49 We tubulin. Nuclei are counterstained with DAPI (blue). provided direct evidence that the SDS22- like LRRs present in Lrrc50 are necessary actin cytoskeleton via anchoring with actin bundling protein for proper protein function, because injection of a deletion ␣-actinin.38 Furthermore, myosin VIIa was shown to be a com- construct lacking the consensus LxxLxL sequence of the first Ϫ Ϫ mon component of cilia and renal microvilli in mice.39 It is LRR did not rescue lrrc50hu255h / ciliary dyskinesia and pro- intriguing to speculate that interaction with actin might be a nephric cyst development. Our data suggest LRRC50 to be a common denominator. Flagellar inner-arm dyneins contain novel candidate gene for human disease affecting the kidney. an actin subunit,40 and actin is dynamically turned over in Identifying LRRC50 interaction partners will help to place static Chlamydomonas flagella; however, its function remains LRRC50 in the complex context of ciliogenesis, cilia mainte- unknown.41 nance, planar cell polarity, and cell-cycle control. Pronephric fluid flow deficiency is representative of pro- nephric ciliary dysfunction in zebrafish.21,25 Although most lrrc50hu255H mutants have impaired fluid excretion, in CONCISE METHODS 39.1% of animals tested, this was markedly slower but not absent, suggestive of residual movement. Similarly, pro- Zebrafish Strains, Screening Methods, and Positional nephric ciliary movement was not detected in the recently Cloning described shy mutant25 either, although 1 of 10 of injected Zebrafish were maintained as described previously,50 and experi- mutants displayed fluorescence at the cloaca after 30 min at ments were conducted in accordance with the Dutch guidelines for

1134 Journal of the American Society of Nephrology J Am Soc Nephrol 19: 1128–1138, 2008 www.jasn.org BASIC RESEARCH

and 3G8 (1:40; gift from Dr. E. Jones56) were C HK-2 MOCK transfected acetylated α−tubulin DAPI performed as described previously.21 Em- bryos were fixed with Dent’s (20% DMSO/ A NAi-2 80% MeOH) or 4% paraformaldehyde mock RNAi-1 R cilium hs LRRC50 (3G8). Secondary antibodies were goat anti- mouse Alexa 488 (1:500; Molecular Probes, http://www.invitrogen.com/) and goat anti- 18S rabbit Cy3 (1:250; Jackson Laboratories, HK-2 http://www.jacksonimmuno.com) or Alexa 555 (1:500, Molecular Probes). Confocal im-

B HK-2 MOCK transfected DAPI ages were collected using a Leica DMI 4000B ezrin HK-2 LRRC50 RNAi-2 ZO-1 microscope.

brush border BrdU Labeling Studies

5 µm Two- and 3-d-old embryos were pulsed with

LRRC50 RNAi-1 10 mM BrdU (Sigma-Aldrich) and 15% DMSO in embryo medium on ice for 20 cilium min.57 After several short washes, embryos were incubated at 28°C for 20 or 60 min, re-

5 µm 10 µm spectively. Six-day-old embryos were pulsed LRRC50 RNAi-2 with 3 mM BrdU in embryo medium for 6 h D at 28°C. Embryos were fixed in 4% parafor- maldehyde, and BrdU incorporation was de- tected with primary anti-BrdU antibody (1: 5 µm 100; DAKO, http://www.dako.com/) and secondary anti-mouse IgG–horseradish per- Figure 6. LRRC50 is required for brush border and cilia regulation in human proximal oxidase (1:300; DAKO) as described previ- tubule cells. (A) Reverse transcriptase–PCR analysis of human LRRC50 in HK-2 cells, either ously.58 mock (empty vector) transfected or with the indicated RNAi construct, demonstrating complete knockdown of LRRC50 mRNA levels. 18S RNA was taken as loading control. (B) LRRC50 RNAi knockdown with two different constructs in human proximal tubule HK-2 Histology cells reduces the ezrin stained brush border (green). Immunostaining with zona occu- Transversal plastic sections (7 ␮m) were lans-1 (ZO-1; red) indicates intact tight junctions. (C and D) LRRC50 RNAi-2 knockdown stained with hematoxylin and eosin using in HK-2 cells reduces the number of ciliated cells and cilia length as shown by anti- standard protocols. acetylated ␣-tubulin staining (green). Nuclei are counterstained with DAPI (blue). Generation of Full-Length Zebrafish the care and use of laboratory animals. ENU mutagenesis was per- lrrc50-EGFP and ⌬LRR1-lrrc50- formed on TL males,51 and F3 embryos were morphologically EGFP Constructs and mRNA screened for ciliary motility defects in the neural tube and nose on a Full-length zebrafish lrrc50 cDNA was amplified from EST clone Zeiss (http://www.zeiss.com/) Axioplan microscope. Positional clon- IRAKp961I11103Q (BC045963; RZPD) and cloned into the pCS2ϩ ing was performed using single-sequence length polymorphisms vector, C-terminally fused with EGFP (derived from pEGFP-N2; from the MGH mapping panel,52 the Ensembl Zv4 zebrafish genome Clontech, http://www.clontech.com/). Primers were 5Ј-ACTCCTG- assembly, and single-nucleotide polymorphism markers (http://cas- GATCCTAATATCACTGCGAAATGAAGACAAA-3Ј and 5Ј-GT- cad.niob.knaw.nl). Candidate genes were sequenced. Marker infor- CAGTGGATCCTGTCCAGTTCCTCAATCAGAACAGT-3Ј (Bio- mation is available upon request. legio, http://www.biolegio.com). Sequence was verified. A deletion construct was generated by removing nucleotides 391 to In Situ Hybridization and Immunohistochemistry 408 encoding the consensus LxxLxL of the first LRR (⌬LRR1) us- Whole-mount in situ hybridizations were performed as described ing the QuickChange Site-Directed Mutagenesis Kit (Stratagene, previously53 with minor modifications. A partial 1.3-kb antisense http://www.stratagene.com) according to the manufacturer’s digoxygenin-labeled (Roche, http://www.roche-applied-science. instructions, on the lrrc50-EGFP fusion construct described com) mRNA probe for lrrc50 was synthesized from EST clone previously. Primers were 5Ј-GTATAGAGGGTTTGGAAGAAT- IMAGp998E0413112Q (BQ419779; RZPD, http://www.imagenes-bio. ATA CTGGCGAGTGTATGGTATTCGAAAGATTGAAAA-3Ј and de). lrrc50 and spaw54 transcripts were purified using NucleoSpin 5Ј-TTTTCAATCTTTCGAATACCATTACACTCGCCAGTATATTC- RNA clean-up columns (Macherey-Nagel, http://www.mn-net. TTCCAAACCCTCTATAC-3Ј (Biolegio). Clones were verified by Hin- com/). Immunohistochemistry for acetylated ␣-tubulin (1:400; Sig- dIII digestion and sequencing. After KpnI linearization, lrrc50-EGFP and ma-Aldrich, http://www.sigmaaldrich.com/), Ntl (notail; 1:100),55 ⌬LRR1-lrrc50-EGFP mRNA were transcribed in vitro using the SP6

J Am Soc Nephrol 19: 1128–1138, 2008 LRRC50 Implicated in PKD 1135 BASIC RESEARCH www.jasn.org

A sibling lrrc50-/- Probes) solution was injected into the heart of ODA I. IDA 4.5 dpf embryos that were anesthetized with III IV 100 mM MS222 and embedded in 0.5% aga- II. rose. Embryos were monitored under a Leica II III. fluorescence stereo microscope, and the time IV. from injection to the first visible excretion at I V. the cloaca was recorded. Only embryos that V immediately after injection showed TAMRA throughout the cardiovascular system were

o B 0 C included. 1 2 3 a microtubule doublet alignment on axoneme (angle in degrees) 27o b 0-3 5-10 10-20 > 20 Scanning and Transmission Electron sibling 88.7% (n=86) 4.1% (4) 1% (1) 6.2% (6) Microscopy lrrc50-/- 43.5% (40) 9.8% (9) 18.5% (17) 28.3% (26) Embryos were fixed in Karnovsky fixative (2% paraformaldehyde, 2.5% glutaralde- hyde, 0.08 M Na-cacodylate [pH 7.4], 0.25 Figure 7. Aberrant ciliary ultrastructure in lrrc50 mutant zebrafish. (A) Electron microscopy mM calcium chloride, and 0.5 mM magne- hu255H Ϫ/Ϫ reveals ultrastructural irregularities in lrrc50 . Most mutant cilia lack the ODA nor- sium chloride [pH 7.4]) for at least 24 h at mally present on the outer-doublet microtubules of ciliary axonemes (I); however, misplace- 4°C. Briefly, for scanning electron micros- ment of the IDA (II/III) or ODA (IV) or complete lack of all dynein arms (V) is also observed. (B) copy, embryos were dehydrated in ethanol Outer-doublet alignments were measured as depicted in the schematic overview. Briefly, a and critical point–dried from CO . Samples best-fitting ellipse was drawn through the center (red dot 2) of the doublets. Then, a straight 2 were sputter-coated with 5 nm of Au/Pd, and line (a) was drawn through the middle points of the individual doublet microtubules (dots 1 to 3). The angle between the drawn tangent line (b) through dot 2 with the best fitting ellipse analyses were conducted at 20-kv accelerating and line (a) was measured. (C) lrrc50hu255H mutants show a marked increase in outer-doublet voltage with a Hitachi (http://www.hitachi- misalignments. hitec.com/global/index.html) S-800 field emission scanning electron microscope. For mMESSAGE mMACHINE kit (Ambion, http://www.ambion.com/). transmission electron microscopy, samples Respectively, 320 and 300 pg of mRNA was injected into one-cell-stage were postfixed in 1% osmiumtetroxide and embedded in Epon 812. embryos within 30 min after fertilization. Only GFP-positive embryos Ultrathin sections (60 nm) were contrasted with 3% uranyl magne- were analyzed, and all embryos were genotyped after termination of the sium acetate and lead citrate and viewed with a Jeol (http://www.je- experiment. Furthermore, transcription of the mRNA was verified by ol.com/) JEM 1010 or a Philips (Eindhoven, The Netherlands) CM10 Western analysis using an anti-GFP antibody (1:500; Santa Cruz Biotech- transmission electron microscope. nology, Santa Cruz, CA). In short, anesthetized embryos were frozen in liquid nitrogen and sonicated on ice in 15 ␮l of RIPA buffer per embryo, Cell Culture and Transfection supplemented with protease inhibitors (1:1000 apoprotein, leupeptin, Cells were cultured in DMEM supplemented with antibiotics and 6 to and PMSF [Roche]). Laemmli sample buffer was added (1:1), and sam- 10% FCS. Transfection with LRRC50-EGFP in pCDNA3 was per- ples were boiled for 10 min. One embryo equivalent was loaded per lane, formed by electroporation (270 V, 1.0 mF). Constructs were co-trans- and proteins were separated on an 8% SDS polyacrylamide gel as de- fected with pBABE-puro (gift of the Nolan Laboratory, http://www. scribed previously.59 Anti–integrin-linked kinase (ILK) was used as a stanford.edu/group/nolan/index.html) in a 5:1 ratio, then selected loading control (1:4000; Sigma). with puromycin (1 ␮g/ml; Sigma-Aldrich) 24 h later.

Generation of Full-Length Human LRRC50-EGFP Full-length human LRRC50 cDNA was amplified from verified EST Immunofluorescence Staining clone IRATp970C0324D6 (BC024009; RZPD) using the Advantage-2 Confluent cells were cultured on coverslips for 4 d in serum-free me- PCR enzyme system (Clontech), introducing a 5Ј EcoRI and a 3Ј SacII dium, then fixed either in 4% paraformaldehyde for 15 min and per- restriction site. Primers were 5Ј-ACTCCTGAATTCACCACCATG- meabilized in 0.1% Triton/PBS for 10 min or in ice-cold methanol for CACCCTGAGCCCTC-3Ј and 5Ј-GTCAGTCCGCGGTATGAT- 2 min. Primary antibodies used were: ␣-ezrin (1:400; BD-Pharmin- GCTTTCGGTGCTGGG-3Ј (Biolegio). After EcoRI/SacII digestion, gen, http://www.bdbiosciences.com), ␣-acetylated ␣-tubulin the 2.2-kb product was cloned into the pcDNA3 vector (Stratagene), (1:10,000; Sigma-Aldrich), ␣-␥-tubulin (1:500; Sigma-Aldrich), establishing a C-terminal EGFP fusion (derived from pEGFP-N2; ␣-zona occulans-1 (1:1000; Zymed, htp://www.invitrogen.com/). Clontech). Sequence was verified. Secondary antibody was goat anti-mouse Alexa 488 (1:500; Molecular Probes) and/or goat anti-rabbit Alexa 568 (1:500; Molecular Probes). Fluorescent Dye Injection and Excretion Assay Confocal analysis was performed on a Zeiss LSM510. For cilia detec- For dye excretion assays, 1 nl of a 7-mg/ml tetramethylrhodamine- tion, 300 interphase nuclei (as determined by DAPI) were scored for conjungated 70-k molecular weight dextran (TAMRA; Molecular the presence of cilia without knowledge of the sample identity. Exper-

1136 Journal of the American Society of Nephrology J Am Soc Nephrol 19: 1128–1138, 2008 www.jasn.org BASIC RESEARCH iments were performed on at least two different days, and the data 9. Harris PC, Torres VE: Understanding pathogenic mechanisms in poly- were combined. cystic kidney disease provides clues for therapy. Curr Opin Nephrol Hypertens 15: 456–463, 2006 10. Bisgrove BW, Yost HJ: The roles of cilia in developmental disorders RNAi and disease. Development 133: 4131–4143, 2006 shRNAi clones v2HS_73906 (RNAi-1, target CCTTCACAGA- 11. Stolc V, Samanta MP, Tongprasit W, Marshall WF: Genome-wide CATCTTTAA), v2HS_73908 (RNAi-2, target CAGAATATGT- transcriptional analysis of flagellar regeneration in Chlamydomonas GCTTTCCGA), or empty vector (“mock,” pSM2c) were isolated as reinhardtii identifies orthologs of ciliary disease genes. Proc Natl Acad Sci U S A 102: 3703–3707, 2005 recommended by the manufacturer (Open Biosystems, http://www. 12. Avidor-Reiss T, Maer AM, Koundakjian E, Polyanovsky A, Keil T, Sub- openbiosystems.com/). Ten micrograms of plasmid were co-trans- ramaniam S, Zuker CS: Decoding cilia function: Defining specialized fected with 2 ␮g of pBABE-puro into approximately 2 million HK-2 genes required for compartmentalized cilia biogenesis. Cell 117: 527– cells by electroporation on two consecutive days before puromycin 539, 2004 selection 24 h later. LRRC50 expression in mock- and RNAi-trans- 13. Blacque OE, Reardon MJ, Li C, McCarthy J, Mahjoub MR, Ansley SJ, Badano JL, Mah AK, Beales PL, Davidson WS, Johnsen RC, Audeh M, fected HK-2 cells was detected by reverse transcriptase–PCR, using Plasterk RH, Baillie DL, Katsanis N, Quarmby LM, Wicks SR, Leroux standard touchdown PCR (25 cycles) with primers 5Ј-ATGCACCCT- MR: Loss of C. elegans BBS-7 and BBS-8 protein function results in GAGCCCTC-3Ј and 5Ј-GCAGAGTTTTTGCAGGGAAC-3Ј. 18S cilia defects and compromised . Genes Dev 18: RNA was taken as loading control: Primers 5Ј-AGTTGGTGGAGC- 1630–1642, 2004 GATTTGTC-3Ј and 5Ј-TATTGCTCAATCTCGGGTGG-3Ј. 14. Pazour GJ, Agrin N, Leszyk J, Witman GB: Proteomic analysis of a eukaryotic cilium. J Cell Biol 170: 103–113, 2005 15. Smith JC, Northey JG, Garg J, Pearlman RE, Siu KW: Robust method for proteome analysis by MS/MS using an entire translated genome: ACKNOWLEDGMENTS Demonstration on the ciliome of Tetrahymena thermophila. J Pro- teome Res 4: 909–919, 2005 This project was supported by the Dutch Cancer Association 16. Li Q, Montalbetti N, Wu Y, Ramos AJ, Raychowdhury MK, Chen XZ, Cantiello, HF: Polycystin-2 cation channel function is under the control (UU2006-3565). F.V.E. is supported by the MRC Centre for Devel- of microtubular structures in primary cilia of renal epithelial cells. J Biol opmental and Biomedical Genetics, and R.H.G. is supported by a Chem 281: 37566–37575, 2006 VIDI award (Netherlands Scientific Organization). 17. Freshour J, Yokoyama R, Mitchell DR: Chlamydomonas flagellar outer We gratefully acknowledge the Hubrecht Screening Team, Jeroen row dynein assembly protein ODA7 interacts with both outer row and Korving for tissue sectioning, Leon Tertoolen for data acquisition, I1 inner row dyneins. J Biol Chem 282: 5404–5412, 2007 18. Essner JJ, Amack JD, Nyholm MK, Harris EB, Yost HJ: Kupffer’s vesicle Chris Hill (Sheffield University) for transmission, and Ju¨rgen Berger is a ciliated organ of asymmetry in the zebrafish embryo that initiates (MPI Tu¨bingen) for scanning electron microscopy. We thank Dr. D. left-right development of the brain, heart and gut. Development 132: Mitchell for sharing reagents and unpublished data. 1247–1260, 2005 19. Essner JJ, Vogan KJ, Wagner MK, Tabin CJ, Yost HJ, Brueckner M: Conserved function for embryonic nodal cilia. Nature 418: 37–38, 2002 DISCLOSURES 20. Bisgrove BW, Snarr BS, Emrazian A, Yost HJ: Polaris and polycystin-2 in dorsal forerunner cells and Kupffer’s vesicle are required for spec- None. ification of the zebrafish left-right axis. Dev Biol 287: 274–288, 2005 21. Kramer-Zucker AG, Olale F, Haycraft CJ, Yoder BK, Schier AF, Drum- mond IA: Cilia-driven fluid flow in the zebrafish pronephros, brain and REFERENCES Kupffer’s vesicle is required for normal organogenesis. Development 132: 1907–1921, 2005 1. Ibanez-Tallon I, Heintz N, Omran H: To beat or not to beat: Roles of 22. Bisgrove BW, Morelli SH, Yost HJ: Genetics of human laterality disor- cilia in development and disease. Hum Mol Genet 12: R27–R35, 2003 ders: Insights from vertebrate model systems. Annu Rev Genomics 2. Brokaw CJ: Flagellar movement: A sliding filament model. Science Hum Genet 4: 1–32, 2003 178: 455–462, 1972 23. Hamada H, Meno C, Watanabe D, Saijoh Y: Establishment of verte- 3. Dutcher SK: Flagellar assembly in two hundred and fifty easy-to-follow brate left-right asymmetry. Nat Rev Genet 3: 103–113, 2002 steps. Trends Genet 11: 398–404, 1995 24. Peeters H, Devriendt K: Human laterality disorders. Eur J Med Genet 4. Li JB, Gerdes JM, Haycraft CJ, Fan Y, Teslovich TM, May-Simera H, Li 49: 349–362, 2006 H, Blacque OE, Li L, Leitch CC, Lewis RA, Green JS, Parfrey PS, Leroux 25. Zhao C, Malicki J: Genetic defects of pronephric cilia in zebrafish. MR, Davidson WS, Beales PL, Guay-Woodford LM, Yoder BK, Stormo Mech Dev 124: 605–616, 2007 GD, Katsanis N, Dutcher SK: Comparative genomics identifies a flagel- 26. Kobe B, Deisenhofer J: The leucine-rich repeat: A versatile binding lar and basal body proteome that includes the BBS5 human disease motif. Trends Biochem Sci 19: 415–421, 1994 gene. Cell 117: 541–552, 2004 27. Kobe B, Deisenhofer J: Proteins with leucine-rich repeats. Curr Opin 5. Sun Z, Amsterdam A, Pazour GJ, Cole DG, Miller MS, Hopkins N: A Struct Biol 5: 409–416, 1995 genetic screen in zebrafish identifies cilia genes as a principal cause of 28. Sigrist CJ, Cerutti L, Hulo N, Gattiker A, Falquet L, Pagni M, Bairoch A, cystic kidney. Development 131: 4085–4093, 2004 Bucher P: PROSITE: A documented database using patterns and 6. Rosenbaum JL, Witman GB: Intraflagellar transport. Nat Rev Mol Cell profiles as motif descriptors. Brief Bioinform 3: 265–274, 2002 Biol 3: 813–825, 2002 29. Letunic I, Copley RR, Pils B, Pinkert S, Schultz J, Bork P: SMART 5: 7. Simons M, Walz G: Polycystic kidney disease: Cell division without a Domains in the context of genomes and networks. Nucleic Acids Res c(l)ue? Kidney Int 70: 854–864, 2006 34: D257–D260, 2006 8. Singla V, Reiter JF: The primary cilium as the cell’s antenna: Signaling 30. Kobe B, Kajava AV: The leucine-rich repeat as a protein recognition at a sensory organelle. Science 313: 629–633, 2006 motif. Curr Opin Struct Biol 11: 725–732, 2001

J Am Soc Nephrol 19: 1128–1138, 2008 LRRC50 Implicated in PKD 1137 BASIC RESEARCH www.jasn.org

31. Ceulemans H, De Maeyer M, Stalmans W, Bollen M: A capping 47. Silflow CD, LaVoie M, Tam LW, Tousey S, Sanders M, Wu W, domain for LRR protein interaction modules. FEBS Lett 456: 349–351, Borodovsky M, Lefebvre PA: The Vfl1 protein in Chlamydomonas 1999 localizes in a rotationally asymmetric pattern at the distal ends of the 32. Bretscher A, Reczek D, Berryman M: Ezrin: A protein requiring conforma- basal bodies. J Cell Biol 153: 63–74, 2001 tional activation to link microfilaments to the plasma membrane in the 48. Morgan GW, Denny PW, Vaughan S, Goulding D, Jeffries TR, Smith assembly of cell surface structures. J Cell Sci 110: 3011–3018, 1997 DF, Gull K, Field MC: An evolutionarily conserved coiled-coil protein 33. Fowkes ME, Mitchell DR: The role of preassembled cytoplasmic com- implicated in polycystic kidney disease is involved in basal body plexes in assembly of flagellar dynein subunits. Mol Biol Cell 9: 2337– duplication and flagellar biogenesis in Trypanosoma brucei. Mol Cell 2347, 1998 Biol 25: 3774–3783, 2005 34. Nicastro D, Schwartz C, Pierson J, Gaudette R, Porter ME, McIntosh 49. Amsterdam A, Nissen RM, Sun Z, Swindell EC, Farrington S, Hopkins JR: The molecular architecture of axonemes revealed by cryoelectron N: Identification of 315 genes essential for early zebrafish develop- tomography. Science 313: 944–948, 2006 ment. Proc Natl Acad Sci U S A 101: 12792–12797, 2004 35. Flock A, Duvall AJ 3rd: The ultrastructure of the kinocilium of the sensory 50. Westerfield M: The Zebrafish Book: A Guide for the Laboratory Use of cells in the inner ear and lateral line organs. J Cell Biol 25: 1–8, 1965 Zebrafish (Danio rerio), Eugene, University of Oregon Press, 1995 36. Webber WA, Lee J: Fine structure of mammalian renal cilia. Anat Rec 51. van Eeden FJ, Granato M, Odenthal J, Haffter P: Developmental 182: 339–343, 1975 mutant screens in the zebrafish. Methods Cell Biol 60: 21–41, 37. Mokrzan EM, Lewis JS, Mykytyn K: Differences in renal tubule primary 1999 cilia length in a mouse model of Bardet-Biedl syndrome. Nephron Exp 52. Shimoda N, Knapik EW, Ziniti J, Sim C, Yamada E, Kaplan S, Jackson Nephrol 106: e88–e96, 2007 D, de Sauvage F, Jacob H, Fishman MC: Zebrafish genetic map with 38. Li Q, Montalbetti N, Shen PY, Dai XQ, Cheeseman CI, Karpinski E, Wu G, 2000 microsatellite markers. Genomics 58: 219–232, 1999 Cantiello HF, Chen XZ: Alpha-actinin associates with polycystin-2 and 53. Schulte-Merker S: Looking at embryos. In: Zebrafish: A Practical Ap- regulates its channel activity. Hum Mol Genet 14: 1587–1603, 2005 proach, edited by Nusslein-Volhard C, Dahm R, Oxford, Oxford Uni- 39. Wolfrum U, Liu X, Schmitt A, Udovichenko IP, Williams DS: Myosin VIIa versity Press, 2002 as a common component of cilia and microvilli. Cell Motil Cytoskele- 54. Long S, Ahmad N, Rebagliati M: The zebrafish nodal-related gene ton 40: 261–271, 1998 southpaw is required for visceral and diencephalic left-right asymme- 40. Piperno G, Luck DJ: An actin-like protein is a component of axonemes try. Development 130: 2303–2316, 2003 from Chlamydomonas flagella. J Biol Chem 254: 2187–2190, 1979 55. Schulte-Merker S, Ho RK, Herrmann BG, Nusslein-Volhard C: The 41. Watanabe Y, Hayashi M, Yagi T, Kamiya R: Turnover of actin in protein product of the zebrafish homologue of the mouse T gene is Chlamydomonas flagella detected by fluorescence recovery after expressed in nuclei of the germ ring and the notochord of the early photobleaching (FRAP). Cell Struct Funct 29: 67–72, 2004 embryo. Development 116: 1021–1032, 1992 42. Nauli SM, Alenghat FJ, Luo Y, Williams E, Vassilev P, Li X, Elia AE, Lu 56. Vize PD, Jones EA, Pfister R: Development of the Xenopus pronephric W, Brown EM, Quinn SJ, Ingber DE, Zhou J: Polycystins 1 and 2 system. Dev Biol 171: 531–540, 1995 mediate mechanosensation in the primary cilium of kidney cells. Nat 57. Appel B, Givan LA, Eisen JS: Delta-Notch signaling and lateral Genet 33: 129–137, 2003 inhibition in zebrafish spinal cord development. BMC Dev Biol 1: 43. Yamaguchi T, Hempson SJ, Reif GA, Hedge AM, Wallace DP: Calcium 13, 2001 restores a normal proliferation phenotype in human polycystic kidney 58. Kimmel CB, Miller CT, Kruze G, Ullmann B, BreMiller RA, Larison KD, disease epithelial cells. J Am Soc Nephrol 17: 178–187, 2006 Snyder HC: The shaping of pharyngeal cartilages during early devel- 44. Benashski SE, Patel-King RS, King SM: Light chain 1 from the Chlamy- opment of the zebrafish. Dev Biol 203: 245–263, 1998 domonas outer dynein arm is a leucine-rich repeat protein associated 59. Giles RH, Lolkema MP, Snijckers CM, Belderbos M, van der Groep P, with the motor domain of the gamma heavy chain. Biochemistry 38: Mans DA, van Beest M, van Noort M, Goldschmeding R, van Diest PJ, 7253–7264, 1999 Clevers H, Voest EE: Interplay between VHL/HIF1alpha and Wnt/beta- 45. Wu H, Maciejewski MW, Marintchev A, Benashski SE, Mullen GP, King catenin pathways during colorectal tumorigenesis. Oncogene 25: SM: Solution structure of a dynein motor domain associated light 3065–3070, 2006. chain. Nat Struct Biol 7: 575–579, 2000 46. Padma P, Satouh Y, Wakabayashi K, Hozumi A, Ushimaru Y, Kamiya R, Inaba K: Identification of a novel leucine-rich repeat protein as a component of flagellar in the Ascidian Ciona intestinalis. Supplemental information for this article is available online at http://www. Mol Biol Cell 14: 774–785, 2003 jasn.org/.

1138 Journal of the American Society of Nephrology J Am Soc Nephrol 19: 1128–1138, 2008