View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Developmental Biology 287 (2005) 274 – 288 www.elsevier.com/locate/ydbio

Polaris and Polycystin-2 in dorsal forerunner cells and Kupffer’s vesicle are required for specification of the zebrafish left–right axis

Brent W. Bisgrove, Brian S. Snarr 1, Anoush Emrazian, H. Joseph Yost *

Department of Oncological Sciences, Huntsman Cancer Institute Center for Children, University of Utah, Salt Lake City, UT 84112, USA

Received for publication 26 May 2005, revised 23 August 2005, accepted 26 August 2005 Available online 7 October 2005

Abstract

Recently, it has become clear that motile cilia play a central role in initiating a left-sided signaling cascade important in establishing the LR axis during mouse and zebrafish embryogenesis. Two proposed to be important in this cilia-mediated signaling cascade are polaris and polycystin-2 (pkd2). Polaris is involved in ciliary assembly, while Pkd2 is proposed to function as a Ca2+-permeable cation channel. We have cloned zebrafish homologues of polaris and pkd2. Both genes are expressed in dorsal forerunner cells (DFCs) from gastrulation to early somite stages when these cells form a ciliated Kupffer’s vesicle (KV). Morpholino-mediated knockdown of Polaris or Pkd2 in zebrafish results in misexpression of left-side-specific genes, including southpaw, lefty1 and lefty2, and randomization of heart and gut looping. By targeting morpholinos to DFCs/KV, we show that polaris and pkd2 are required in DFCs/KV for normal LR development. Polaris morphants have defects in KV cilia, suggesting that the laterality phenotype is due to problems in cilia function per se. We further show that expression of polaris and pkd2 is dependent on the T-box transcription factors no tail and spadetail, respectively, suggesting that these genes have a previously unrecognized role in regulating ciliary structure and function. Our data suggest that the functions of polaris and pkd2 in LR patterning are conserved between zebrafish and mice and that Kupffer’s vesicle functions as a ciliated organ of asymmetry. D 2005 Elsevier Inc. All rights reserved.

Keywords: polaris; polycystin-2; Cilia; Left–right asymmetry; Kupffer’s vesicle

Introduction defects in early embryonic processes such as gastrulation and midline formation, or otherwise alter asymmetric nodal, lefty or Although vertebrates are bilaterally symmetric externally, all pitx2 expression, result in perturbations in left–right organo- share highly conserved asymmetries in the orientation of internal genesis later in development (Beddington and Robertson, 1999; organs including the heart, gut and brain, with respect to the Bisgrove and Yost, 2001; Burdine and Schier, 2000; Capdevila left–right (LR) axis. Prior to organogenesis, all vertebrates also et al., 2000). share conserved molecular asymmetries evident by the asym- While it is clear that asymmetric patterns of expression metric (left-sided) expression of TGF-h signaling molecules of in the lateral plate mesoderm are highly conserved, the initial the nodal and lefty families and of the bicoid-like transcription mechanisms that break symmetry within the developing factor pitx2 in the lateral plate mesoderm (Burdine and Schier, embryo and the degree of conservation of these mechanisms 2000; Capdevila et al., 2000). In zebrafish, these genes are also among distinct classes of vertebrates are less clear. In expressed specifically in the left diencephalon of the developing mammals, a link between ciliary motility and the regulation brain (Bisgrove et al., 1999; Essner et al., 2000; Rebagliati et al., of LR asymmetry has been implied for some time. Humans 1998a,b; Sampath et al., 1998; Thisse and Thisse, 1999). with primary ciliary diskinesia (PCD) often exhibit mirror Experimental manipulations or gene mutations that cause image reversal of their internal organs along with respiratory problems caused by immotile tracheal cilia and male infertility due to defects in sperm motility (Afzelius, 1976). This * Corresponding author. Fax: +1 801 585 5470. E-mail address: [email protected] (H.J. Yost). syndrome has been shown to result from mutations in 1 Present address: Department of Cell Biology and Anatomy, Medical components of the ciliary motor, including intermediate University of South Carolina, Charleston, SC 29425, USA. chains (Guichard et al., 2001; Pennarun et al., 1999) and

0012-1606/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ydbio.2005.08.047 B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 275 dynein heavy chains (Bartoloni et al., 2002; Olbrich et al., loss of function of either gene in mice also causes LR 2002). A link between cilia and the regulation of LR patterning defects (Moyer et al., 1994; Murcia et al., 2000; asymmetry was strengthened by the discovery that the inversus Pennekamp et al., 2002; Wu et al., 1998). viscerum (iv) mouse mutant results from a mutation in a ciliary Pkd2 is estimated to be mutated in about 15% of families dynein heavy chain gene, Left–Right Dynein (lrd), expressed affected with autosomal dominant polycystic kidney disease in the embryonic node (Supp et al., 1997). with the majority of the remaining cases attributed to mutations In the mouse embryo, a single large projects from in polycystin-1 (pkd1)(Peters et al., 1993). Pkd1 and pkd2 each cell in a pit-like cavity in the node (Sulik et al., 1994). encode multipass transmembrane that physically These node monocilia rotate in a clockwise direction and interact (Qian et al., 1997; Xu et al., 2003) and can associate generate flow of extracellular fluid which moves in a net at the plasma membrane to form calcium-permeable non- leftward direction (Nonaka et al., 1998). In several mouse selective cation channels (Cahalan, 2002; Hanaoka et al., mutants, there is a correlation between defects in monocilia 2000). In human and mouse kidney cells, Pkd2 is localized to structure or movement, alterations in nodal flow and aberrant primary cilia and the ER (Cai et al., 1999; Pazour et al., 2002), LR development. Knock-out mutants of the KIF3B suggesting a role for this in ciliary function. In cultured and KIF3A, which are -dependent motor mole- Madin–Darby canine kidney cells, manipulation of a single cules, lack node monocilia and nodal flow and have LR defects primary cilium caused an increase in intracellular calcium in (Marszalek et al., 1999; Nonaka et al., 1998; Takeda et al., neighboring cells as well as in the stimulated cell (Praetorius 1999). Mutants in lrd exhibit normal node morphology and and Spring, 2001, 2003). These calcium transients were node cilia are normally formed, but these cilia are immotile and dependent on both extracellular and intracellular calcium pools there is no discernable nodal flow (Okada et al., 1999; Supp et and did not occur in cells treated with antibodies blocking Pkd1 al., 1999). Nodal flow was initially proposed to move an or Pkd2 (Nauli et al., 2003). It has been suggested that ciliary unknown LR determinant to the left side of the node, but more Pkd2 acts as a mechanoreceptive calcium-entry channel, recent experiments using in vitro embryo culture suggest that triggering calcium release from intracellular stores perhaps directional fluid flow per se may be necessary and sufficient for via Pkd2 localized to the ER (Koulen et al., 2002). specifying the LR axis in mouse (Nonaka et al., 2002). In mice, the Oak Ridge Polycystic Kidney disease gene Recently, McGrath et al. (2003) have provided evidence of two causes a phenotype resembling human autosomal recessive types of monocilia in the mouse node. GFP-tagged LRD fusion polycystic kidney disease (Moyer et al., 1994). The gene protein is expressed in a centrally located subset of motile node responsible for this phenotype, Tg737 or polaris (Murcia et al., monocilia, while the periphery of the node has non-LRD 2000; Yoder et al., 1995), is homologous to Clamydomonas containing monocilia. These authors proposed that asymmetric ift88 (Cole et al., 1998) and C. elegans osm-5 (Haycraft et al., nodal fluid flow functions to activate non-motile mechan- 2001; Qin et al., 2001), genes required for intraflagellar osensory cilia on the periphery of the node which initiates an transport (IFT). IFT is the - and dynein-mediated asymmetric calcium flux. movement of large protein rafts that occurs along microtubule In non-mammalian vertebrates, expression of lrd homo- filaments within the of cilia and flagella (Kozminski logues and monocilia also occur in a small group of cells et al., 1993; Pazour et al., 1998). In mammals, Polaris protein within the organizer in Xenopus, the node in chick and the localizes to the basal bodies and the axoneme of motile or dorsal shield in zebrafish at stages prior to the onset of immotile cilia (Taulman et al., 2001). asymmetric in the lateral plate mesoderm, Protein localization data suggest that Polaris and Pkd2 suggesting a conserved role for monocilia in LR patterning functions in cilia may be linked. Pkd1 and Pkd2 colocalize function across vertebrate species (Essner et al., 2002). In with Polaris and a novel cytoplasmic protein, Cystin, in the zebrafish, dorsal forerunner cells (DFCs) express an lrd primary cilia of kidney cells (Hou et al., 2002; Pazour et al., homologue, lrdr1 and form a ciliated epithelium inside a 2002; Yoder et al., 2002). Similarly, the C. elegans homo- fluid-filled organ called Kupffer’s vesicle (KV). These cilia are logues of Pkd1, Pkd2 and Polaris colocalize in the cilia of motile and create directional fluid flow (Essner et al., 2005; sensory neurons (Barr et al., 2001; Barr and Sternberg, 1999; Kramer-Zucker et al., 2005). Morpholino knockdown of lrdr1 Haycraft et al., 2001; Qin et al., 2001). In addition, high levels function disrupts KV fluid flow and perturbs LR development, of Pkd2 accumulate in the shortened cilia of cells derived from suggesting that KV cilia in zebrafish play a similar role in the kidneys of polaris mutant mice (Pazour et al., 2002), and establishing the LR axis as do node monocilia in mice (Essner high levels of LOV-10GFP (Pkd1) and PKD20GFP accumu- et al., 2005). late in osm-5 stunted cilia of C. elegans (Qin et al., 2001), To further understand the mechanism by which the zebrafish suggesting that OSM-5/Polaris and IFT are not required to LR axis is established, we have investigated the role of two move Pkd2 into cilia but may be required to recycle it to the proteins proposed to be important in cilia-mediated signaling cell body. cascades. Polaris is involved in ciliary assembly, while In this report, we show that zebrafish polaris and pkd2 are Polycystin-2 (Pkd2) is a Ca2+-permeable cation channel expressed in several ciliated tissues during early development associated with cilia and thought to play a role as a ciliary including KV. Knockdown of Polaris or Pkd2 function results mechanosensor. Defects in polaris and pkd2 function were in misexpression of normally left-side-specific genes, including originally linked to polycystic kidney disease in mammals, and southpaw (spaw), lefty1 (lft1) and lefty2 (lft2), and causes 276 B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 randomization of heart and gut looping. Given the broad Pol AUG MO were tagged with fluorescein to facilitate the tracking of their expression of these genes, analysis of mouse mutants or distribution in injected embryos. Morpholinos were injected into the yolk just under the blastoderm of 1- to zebrafish morphants does not allow us to identify the tissue in 2-cell stage embryos or into the center of the yolk at midblastula stages in a which these genes function during LR specification. To address volume of 1.0 nl. Morpholino induced phenotypes were rescued by coinjection this important issue, we used a technique that targets of in vitro synthesized RNAs with polaris or pkd2 morpholinos. Capped RNAs morpholinos specifically to KV (Amack and Yost, 2004) and were synthesized from pCS2+ expression plasmid (Rupp et al., 1994; Turner found that polaris and pkd2 morpholinos are required in KV and Weintraub, 1994) constructs containing the entire open-reading frames of Polaris and Pkd2 using the mMessage machine SP6 transcription kit (Ambion). for normal LR development. This is the first demonstration that polaris and pkd2 function in node/KV cells to control LR RNA in situ localization development. In addition, polaris morphants have defects in KV cilia suggesting that the laterality phenotype is due to For in situ hybridizations, embryos were fixed in 4% parafomaldehyde in problems in cilia function per se. Analysis of polaris and pkd2 sucrose buffer (Westerfield, 1995), rinsed in PBST, dehydrated into absolute - expression levels in zebrafish laterality mutants suggests that methanol and stored at À20 C. Riboprobes were synthesized from linearized DNA templates using T3 or T7 polymerases and digoxigenin or fluorescein the T-box transcription factors no-tail (ntl) and spadetail (spt) labeling mixes (BMB). In situ hybridizations were carried out as previously play a role in the regulation of these genes in DFCs. Our data described (Bisgrove et al., 1999). Anti-sense probes used include: polaris, pkd2 suggest that the functions of polaris and pkd2 in LR patterning (this report), lefty1 (lft1), lefty2 (lft2)(Bisgrove et al., 1999); southpaw (spaw) are conserved between zebrafish and mice and support the (Long et al., 2003); cardiac light chain 2 (cmlc2)(Yelon et al., 1999); hypothesis that KV functions as a ciliated organ of asymmetry. forkhead 2 (fkd2)(Odenthal and Nusslein-Volhard, 1998); Wilms’ tumor suppressor 1 (wt1)(Drummond et al., 1998); myoD (Weinberg et al., 1996); sonic hedgehog (shh)(Krauss et al., 1993); ntl (Schulte-Merker et al., 1994); Materials and methods patched 1 (ptc)(Karlstrom et al., 1999); axial (Strahle et al., 1993); sox17 (Alexander and Stainier, 1999). Embryos were cleared in 70% glycerol/PBST Embryo culture and zebrafish stocks and photographed with a Leica MZ12 dissecting microscope. Images were captured with a Nikon Coolpix 995 digital camera and processed using Adobe Zebrafish, Danio rerio, were maintained at 28.5-C on a 14 h/10 h light/dark Photoshop. cycle. Embryos were collected from natural matings, cultured and staged by developmental time and morphological criteria (Westerfield, 1995). Wild-type Immunohistochemistry embryos were obtained from Oregon AB strain zebrafish (Zebrafish Interna- tional Resource Center, Oregon). For immunohistochemistry, embryos were fixed in 4% parafomaldehyde in sucrose buffer (Westerfield, 1995), rinsed in PBS, dehydrated into Cloning zebrafish polaris and pkd2 cDNAs absolute methanol and stored at À20-C. After rehydration, embryos were boiled in 1 mM EDTA for 5 min then and blocked for 1 h in 5% goat Candidates for homologues of mouse Polaris and Pkd2 were identified from serum (Sigma), 2% bovine serum albumin (BSA) (Sigma) and 1% DMSO in zebrafish genome sequences (Sanger Institute) using tblastn searches. Genomic PBST. Embryos were incubated overnight at 4-C in anti-acetylated sequences surrounding the zebrafish polaris candidate were syntenic with a antibody (1:400, Sigma) in the blocking solution. After washing in 2% BSA region of mouse chromosome14 that contains murine polaris; and the genomic and 1% DMSO in PBST, embryos were incubated with Alexa 568-labeled region containing the zebrafish pkd2 candidate was syntenic with a region of goat anti-mouse antibodies (1:500, Molecular Probes). After washing, the tail mouse 19 that contains murine pkd2. Oligonucleotide primers to region of the embryos was removed and mounted in Slow Fade (Molecular these sequences were designed, and RT-PCR was used to generate cDNA Probes). Embryos were imaged on an Olympus Fluoview scanning laser products from late gastrula, 24 h post-fertilization (hpf) or adult kidney mRNA confocal microscope, and images were processed using Image J software. using the Superscript First Strand Synthesis System (Invitrogen). PCR products were ligated into pCRII TOPO (Invitrogen) and sequenced. Results Morpholino knock-down of Polaris and Pkd2 function Identification of polaris and pkd2 homologues in zebrafish Morpholino antisense oligonucleotides, synthesized by Gene Tools (Oregon) were used to knock-down the function of Polaris and Pkd2: Polaris To address whether the functions of polaris and pkd2 in translation-blocking morpholino (Pol AUG MO) is complementary to a region establishing LR asymmetry are conserved among vertebrates, spanning the translation start site, 5V-ATGCACATTCTCCATCGAGGAAAGT- we identified and characterized homologues of these genes 3V; Polaris splice-blocking morpholino 1 (Pol SP MO1) overlaps the splice donor site of the first exon–intron junction, 5V-AAGTATTTACGCACCTCT- in zebrafish (GenBank DQ175628, DQ175629). The length of GAGTCAA-3Vand Polaris splice-blocking morpholino 2 (Pol SP MO2) the predicted Polaris protein encoded by the zebrafish overlaps the splice donor site of the exon3/intron3 junction, 5V-CAACTC- homologue is identical to that of mouse (824 amino acids). CACTCACCCCATAAGCTGT-3V (previously reported as ift88SP (Tsujikawa Amino acid identity between the zebrafish and mouse or and Malicki, 2004). Pkd2 translation-blocking morpholino (Pkd2 AUG MO) is zebrafish and human proteins is 74 and 77%, respectively complementary to a region spanning the translation start site, 5V-ACTGGAGtT- CATCGTGTATTTCTAC-3V (lower case ‘‘t’’ is a substitution to reduce (Supplementary Fig. 1). The sizes of predicted zebrafish mouse morpholino secondary structure), and Pkd2 splice-blocking morpholino and human homologues of Pkd2 are 904, 966 and 968 amino (Pkd2 SP MO) 5V-TAGTTTTCAACCTACTTATGCAGAG-3V overlaps the acids, respectively, with approximately 60% amino acid exon2/intron2 splice donor site. Other morpholinos used included no tail conservation between the zebrafish and both mammalian (ntl) translation-blocking morpholino 5V-GACTTGAGGCAGGCATATTTCC- proteins (Supplementary Fig. 2). The high degree of sequence GAT-3V(Nasevicius and Ekker, 2000) and spadetail (spt) translation-blocking morpholino, 5V-GCTTGAGGTCTCTGATAGCCTGCAT-3V (provided by D.J. similarity across much of each protein suggests that Polaris and Grunwald) and a standard negative control morpholino, 5V-CCTCTTACCT- Pkd2 functions are likely to be conserved between zebrafish CAGTTACAATTTATA-3V, provided by Gene Tools. All morpholinos except and mammals. B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 277

Polaris and Pkd2 expression domains overlap in Kupffer’s higher levels of expression in KV, the derivative of the DFCs vesicle, a proposed organ of asymmetry (Fig. 1D). Polaris expression in KV is downregulated at the 5 somite stage (Fig. 1E). Polaris expression remains ubiquitous Both polaris and pkd2 transcripts are expressed broadly at later somite stages with higher levels detected in the otic during early development with domains of highest expression vesicle and in the pronephric duct primordia (Figs. 1F, G). At localized to tissues that will become ciliated, including the 24 hpf, polaris is highly expressed in the brain and in the pronephric duct primordia and KV (Fig. 1). Polaris transcript caudal tip of the floorplate, and at 3 days post-fertilization is present maternally, with high levels of mRNA detectable by (dpf), polaris is weakly expressed in the brain, the nasal in situ hybridization from the 1 cell to the 2000 cell stage (Fig. placodes and the eye (Figs. 1H, I). 1A). Polaris transcript is subsequently undetectable until 80% Maternal pkd2 message is not detected by in situ hybrid- epiboly when it is localized to DFCs (Figs. 1B, C). During ization (Fig. 1J). Zygotic expression is initiated at the onset of early somitogenesis, polaris is ubiquitously expressed, with gastrulation in a band of 3–4 cell diameters at the blastoderm

Fig. 1. Polaris and pkd2 are expressed in multiple ciliated tissues during early development. In situ localization of polaris (A–I) and pkd2 (J–R) transcripts at early cleavage through 72 hpf (all panels are lateral views except panels C, G, I, and P, dorsal views, and panel R, ventral view). Maternal polaris RNA is present in cleavage stage embryos (A) and becomes undetectable by early gastrula stages (B). At late gastrula stages (C), polaris expression is restricted to dorsal forerunner cells (dfc) which later aggregate to form the primordium of Kupffer’s vesicle (Kv) at the end of gastrulation (D). Polaris expression is ubiquitous at later stages with high levels detected in the otic vesicle (ov) (F), the pronephric duct primordia (pnd) and floorplate (fp) (G). At 24 hpf, polaris is expressed in the brain and in the caudal floorplate (H). At 3 dpf, polaris is expressed in the brain (I). Maternal pkd2 message is not detected by in situ hybridization (J). At 40% epiboly, zygotic expression is detected in a band of 3–4 cell diameters at the blastoderm margin (K). Pkd2 RNA is localized to the hypoblast of the dorsal midline and to DFCs during gastrulation (L). From tailbud stage through 24 hpf, pkd2 is ubiquitous with higher levels of expression in KV (M) that persist to the 6 somite stage (N). pkd2 message is also detected in the pronephric duct primordia and floorplate (P) and at 24 hpf is highly expressed in the brain (Q). At 3 dpf, pkd2 is expressed in the pharyngeal arches (pa) and pectoral fins (pf) (R). 278 B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 margin (Fig. 1K). Pkd2 RNA is subsequently localized to the embryos injected with 12 ng of polaris splice-blocking hypoblast of the dorsal midline and to DFCs (Fig. 1L). During morpholino 1 (Pol SP MO1) or 2 ng of pkd2 splice-blocking early somite stages, pkd2 is ubiquitously expressed, with morpholino (Pkd2 SP MO). The amplicon produced from higher levels of expression in KV that persist to the 6 somite gene-specific primers that span polaris intron 1 was increased stage (Figs. 1M, N). During subsequent stages, pkd2 is in size in embryos injected with Pol SP MO1 (Fig. 2A). detected in the pronephric duct primordia and neural floorplate Sequence analysis showed that this PCR product included an (Figs. 1O, P) and at 24 hpf is highly expressed throughout the 88 bp intron that causes a frame shift in the second exon brain (Fig. 1Q). At 3 dpf, pkd2 is expressed in the pharyngeal leading to a termination codon at the fifth position in exon 2. arches and the pectoral fin bud (Fig. 1R). Similarly, RT-PCR of Pkd2 SP MO-injected embryos indicated the presence of a 752 bp unspliced intron that encodes in-frame Polaris and pkd2 morpholinos perturb development of tissues stops at intron 2 codons 7 and 8. Splice inhibition by Pol SP expressing those genes MO2 has previously been shown (Tsujikawa and Malicki, 2004). Thus, at the concentrations injected, polaris and pkd2 To assess the functions of Polaris and Pkd2 during zebrafish splice-blocking morpholinos should lead to a substantial development, we designed translation- and splice-blocking decrease in functional gene product. morpholino antisense oligonucleotides to knock down gene The most obvious phenotypes of polaris and pkd2 function. To test the efficacy of the splice-blocking morpho- morphants were alterations in body shape. At 40 hpf, linos, RT-PCR was carried out on total RNA prepared from polaris morphants often exhibit a curled down tail, while

Fig. 2. Knock-down of polaris and pkd2 gene function leads to several morphological phenotypes. Pol SP MO1 (A) and Pkd2 SP MO1 (B) morpholinos interfere with splicing of polaris and pkd2 message, respectively. RT-PCR products produced from gene-specific primers from RNA extracted from uninjected wild-type (wt) embryos versus morpholino-injected embryos are increased in size reflecting incorporation of intron sequences into the mRNAs. As an RT-PCR control, B- primers give the same size products from cDNAs prepared from wt and morpholino-injected embryos (B). Body shape is altered in 40 hpf morphant embryos (C–D). Polaris morphants exhibit a ventrally curved trunk and tail (D) while pkd2 morphants exhibit a dorsally curved trunk and tail (E) compared to uninjected wt embryos (C). Kidney cysts are evident in polaris and pkd2 morphants at 3 dpf (F, G). In uninjected wt embryos, wt1 expression identifies the glomerulus (g) of the pronephric kidney, while in morphants (Pol SP MO1 shown), the glomerulus appears enlarged and the pronephric tubules (pt) are distended (G). Polaris and pkd2 morphants exhibit hydrocephalus and cardiac edema at 2 dpf (H, I). Expansion of the hindbrain (single arrow) and pericardium (asterisk) is evident in morphants (Pkd2 SP MO shown, I) compared to an uninjected wt embryo (H). The direction of heart and gut looping is often reversed in 40 hpf polaris and pkd2 morphants (J–M). In a wt embryo (J), the atrium (a) of the heart loops toward the left side of the embryo while in morphants (Pol AUG MO shown) the atrium often loops toward the right side (K). In wt embryos, the liver (l) and intestinal bulb are located on the left side of the embryo (L) while in morphants (Pol AUG MO shown) these organs are often on the right side (M). B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 279

Table 1 Effect of 1–2 cell morpholino injections on the direction of heart and gut looping Treatment n Heart looping (%) Gut looping (%) Normal Reversed Normal Reversed Uninjected control 1143 98 2 98 2 Control MO (12 ng) 75 93 7 92 7 Pol AUG MO (4.5 ng) 95 62 38 60 37 Pol SP MO1 (12 ng) 197 69 31 65 34 Pol SP MO2 (1.5 ng) 197 60 40 59 41 Pol SP MO2 rescue (1.5 ng/200 pg RNA) 256 79 21 78 22 Pkd2 AUG MO (12 ng) 203 64 32 62 33 Pkd2 SP MO (2 ng) 163 49 51 49 50 Pkd2 SP MO rescue (2 ng/360 pg RNA) 206 70 30 61 39 Heart and gut looping data not adding up to 100% indicate a small percentage of embryos with straight or dysmorphic hearts and bilateral or midline guts. in pkd2 morphants, the tail often curls up (Figs. 2C–E). rates were partially rescued in morphants coinjected with the Embryos injected with polaris or pkd2 morpholinos also cognate in vitro synthesized RNA. Heart reversal rates exhibit kidney cysts (Figs. 2F, G), cardiac edema and recovered from 40% in Pol SP MO2 morphants to 21% in hydrocephalus (Figs. 2H, I). Pol Sp MO2 morphants coinjected with polaris RNA, and heart reversals recovered from 51% in Pkd2 SP MO morphants Polaris and pkd2 morphants have defects in left–right to 30% in Pkd2 SP MO morphants coinjected with pkd2 RNA. organogenesis Over-expression of polaris or pkd2 RNA in wild-type embryos gave no discernible phenotype. In normal embryos at 40 hpf, the ventricle of the heart loops to the right while the atrium loops to the left, and the liver and Polaris and pkd2 morphants have altered LR asymmetric gene intestinal bulb lie on the left side of the midline (Figs. 2J, L). In expression patterns polaris and pkd2 morphants, LR asymmetries in the heart and gut are often reversed (Figs. 2K, M; Table 1). The percentage To address when defects in LR organogenesis occurred of morphants with reversed orientation of the heart ranged from during development, we assayed the expression of genes that 31% (Pol SP MO1) to 51% (Pkd2 SP MO), with similar rates are normally expressed only on the left side of the embryo of reversals in the direction of gut looping. Organ reversal rates (Bisgrove et al., 1999; Bisgrove et al., 2000; Long et al., 2003). in uninjected embryos or embryos injected with control At the 19–21 somite stage, alterations in expression of the morpholino ranged from 2–7%, respectively. Organ reversal nodal-related gene southpaw (spaw) were observed in polaris

Fig. 3. Polaris and pkd2 morpholinos cause randomized expression of spaw (A–E). Polaris and pkd2 morphants show left-sided (A), right-sided (B), bilateral (C) or an absence (D) of spaw expression in the LPM (Pol AUG MO morphants shown). MyoD expression in somites was used to stage embryos. (E) Histogram showing the percentage of control and morphant embryos with left-sided (L), right-sided (R), bilateral (B) or absent (A) spaw expression. 280 B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 and pkd2 morphants (Figs. 3A–D). In uninjected embryos and with left-sided expression of lft1 in 19–60% of embryos and embryos injected with control morpholino at the 1–2 cell stage, left-sided expression of lft2 in 24–61% of embryos. The spaw was expressed in the left lateral plate mesoderm (LPM) in integrity of the midline was not obviously perturbed in these about 95% of embryos (Fig. 3E). Polaris and pkd2 morphants morphants as evident by the normal expression of lft1 in the showed a randomization of spaw expression. 28–54% of notochord (Figs. 4A–D). Other markers of notochord (ntl) and morphant embryos exhibited normal left-sided spaw expres- floorplate (shh, ptc1, axial) were also expressed normally in sion, while the remainder showed right-sided, bilateral or an polaris and pkd2 morphants, confirming that the midlines in absence of spaw expression in the LPM. The bilateral domain these morphants were intact (data not shown). Overall, altera- of spaw in the tailbud (Long et al., 2003) was unaffected in tions in lft1 and lft2 expression patterns reflect those seen for these morphants. spaw expression, including the large numbers of Pol SP MO2 The expression of two other members of the asymmetric and Pkd2 SP MO morphant embryos with no spaw, lft1 or lft2 nodal signaling cascade, the zebrafish lefty-related genes, lefty1 expression, suggesting that spaw signaling lies upstream of the (lft1) and lefty2 (lft2), was also altered in polaris and pkd2 asymmetric expression of lft1 and lft2. morphants (Figs. 4A–D). The majority (94–98%) of uninjected embryos and embryos injected with control morpholino showed Polaris and Pkd2 have distinct roles in cilia function normal left-sided expression of lft1 in the left diencephalon and lft2 in the left LPM (Figs. 4E, F). In Polaris and pkd2 morphants, In polaris and pkd2 mutant mice, alterations in LR these asymmetric gene expression patterns were randomized development have been attributed to defects associated with

Fig. 4. Polaris and pkd2 morpholinos cause randomized expression of lft1 and lft2 in 22–24 somite zebrafish embryos (A–E). Polaris and pkd2 morphants show left-sided (A), right-sided (B), bilateral (C) or absent (D) lft1 expression in the diencephalon and lft2 expression in the LPM (Pol AUG MO morphants shown). MyoD expression in somites was used to stage embryos. (E, F) Histograms show the percentage of control and morphant embryos with left-sided (L), right-sided (R), bilateral (B) or absent (A) lft1 and lft2 expression. B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 281

control embryo, there were 59 cilia with an average length of 3.5 Am(Fig. 5A). In contrast, a similarly sized KV in a Pol AUG MO morphant embryo (Fig. 5B), had a similar number of cilia (50), but the average length of the cilia was shorter (2.2 Am). All cilia within KV were shorter suggesting that all KV cilia require Polaris. Analysis of 5 uninjected control embryos and 5 Pol AUG MO morphant embryos with similar size KVs yielded a mean cilia length of 3.4 T 0.7 Am for control embryos (n = 172 cilia) and 2.1 T 0.6 Am for Fig. 5. KV cilia are shorter in polaris morphant embryos. Anti-acetylated morphant embryos (n = 190 cilia); the mean cilia lengths are tubulin staining of KV cilia in 8 somite embryos. (A) Uninjected wt control significantly different ( P < 0.001, two-tailed Student’s t test). embryo; cilia average 3.5 Am in length. (B) Embryo injected with 4.5 ng Pol Immunohistochemical labeling of KV cilia with anti-IFT188 AUG MO; cilia average 2.2 Am in length. (Pazour et al., 2002) which recognizes zebrafish Polaris (Tsujikawa and Malicki, 2004) was used to confirm the loss node monocilia (Murcia et al., 2000; Pennekamp et al., 2002). of Polaris from KV cilia of morphant embryos (data not Recent evidence suggests that KV in zebrafish functions in an shown). Examination of pkd2 morphant embryos revealed no analogous manner to that proposed for the mouse node in the significant differences in the number or length of KV cilia establishment of the LR axis (Essner et al., 2005). To determine between control embryos and morphants (data not shown). if defects were present in KV cilia, we carried out immuno- histochemistry with acetylated tubulin antibodies on uninjected Polaris and pkd2 morpholinos targeted to DFCs and KV 8-somite stage embryos and similarly staged embryos injected perturb left–right organogenesis at the 1–2 cell stage with polaris or pkd2 morpholinos. At the 8-somite stage, KV appears as a hollow sphere lined with long In mouse polaris and pkd2 mutants, defects in LR patterning cilia (Fig. 5). The size of the vesicle, and correspondingly the have been attributed to defects in node cilia. However, the number of cilia, varies between embryos at the same mutant phenotypes are pleiotropic, affecting many different cell developmental (somite) stage, so we limited our analysis to and tissue types (Murcia et al., 2000; Pennekamp et al., 2002), embryos with a similar size KV. For example, in an uninjected making it impossible to conclude that these genes function in

Fig. 6. Polaris and pkd2 morpholinos targeted to DFCs by midblastula stage yolk injections cause randomized expression of lft1 and lft2 in 22–24 somite zebrafish embryos. (A, B) Histograms show the percentage of uninjected, DFC-control and DFC-morphant embryos with left-sided (L), right-sided (R), bilateral (B) or absent expression (A) of lft1 in the diencephalon and lft2 in the lateral plate mesoderm. 282 B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288

Table 2a not accumulate in DFCs and did not cause significant reversals Effect of DFC-targeted morpholino injections on the direction of heart and gut in heart and gut looping (Table 2b). These data indicate that looping polaris and pkd2 gene functions are required specifically in Treatment n Heart looping (%) Gut looping (%) DFCs/KV for correct establishment of the embryonic LR axis. Normal Reversed Normal Reversed Uninjected control 828 98 2 97 2 Genetic regulation of polaris and pkd2 expression DFCControl MO (24 ng) 129 97 3 97 3 Pol SP MO1 DFC (24 ng) 90 86 13 86 14 A number of zebrafish mutants exhibit LR patterning DFCPol SP MO2 (3 ng) 102 81 15 82 17 DFCPkd2 AUG MO (24 ng) 350 84 15 83 15 defects (Chen et al., 1997), and several of these have been DFCPkd2 SP MO (4 ng) 112 77 22 77 22 classified based on altered patterns of disruption of left-sided Heart and gut looping data not adding up to 100% indicate a small percentage marker genes including lft1, lft2 and pitx2 (Bisgrove and Yost, of embryos with straight or dysmorphic hearts and bilateral or midline guts. 2001; Bisgrove et al., 2000). To address whether the laterality phenotype of any of these mutants might be manifested node cells to control LR development. To address whether the through alterations in the expression of polaris or pkd2,we laterality defects in polaris and pkd2 morphants are due to examined expression of these genes in KV at bud stage in a perturbations in gene function in KV, we targeted morpholinos collection of mutants (Table 3). Clutches of embryos obtained specifically to dorsal forerunner cells, a cell lineage that from crosses of heterozygous carriers should be comprised of contributes to KV (Cooper and D’Amico, 1996). Morpholinos 25% homozygous mutant embryos and 75% phenotypically injected into the yolk during midblastula stages (256–1024 wild-type embryos. No alteration in polaris or pkd2 expression cells) diffuse throughout the yolk and in a subset of embryos are was detected in the predicted 25% of mutant embryos derived to237b n1 tc321 taken up by DFCs but are excluded from the rest of the embryo from heterozygous crosses of lok , flh , cup , m134 tf219 b229 tm84 b641 tl211 (Amack and Yost, 2004). By using fluorescence microscopy to oep , cyc , cyc , din , smu , mom or ty68b ta56 identify embryos in which fluorescein-tagged morpholinos had sur . As expected, in clutches derived from cas loaded specifically into DFCs, we were able to select chimeric heterozygous parents, about 25% lacked expression of polaris embryos in which gene function is knocked down in DFCs but and pkd2 in KV. Embryos lacking cas function fail to develop not other embryonic cells. normal DFCs and their derivatives (Alexander et al., 1999; b195 Midblastula stage injections of polaris (DFCPol MO) and Essner et al., 2005). Interestingly, about 25% of ntl and b104 pkd2 (DFCPkd2 MO) morpholinos randomized asymmetric spt embryos showed a loss in KV expression of polaris and gene expression patterns (Fig. 6). Left-sided expression of pkd2, respectively, while expression of these genes in other lft1 in the diencephalon of DFC-morphants ranged from 61% tissues within the embryos was unaffected. To confirm that KV (DFCPkd SP MO) to 85% (DFCPol SP MO1) versus 91% in cells were present and of a similar number in bud-stage ntl and embryos injected with control MO (Fig. 6A) while left-sided spt embryos, we carried out in situ hybridization with sox17,a expression of lft2 in the LPM of DFC-morphants ranged from transcription factor that is highly expressed in DFCs/KV 64% (DFCPkd2 SP MO) to 83% (DFCPol SP MO1) versus 92% in embryos injected with control MO (Fig. 6B). DFC-morphants Table 3 also showed an increase in heart and gut reversals (13–22%) Expression of polaris and pkd2 in laterality mutants and in ntl and spt compared to embryos injected with control morpholino (3%) or morphants to uninjected controls (2%) (Table 2a). None of the other Gene expression in KV at bud stage morphological phenotypes produced from 1–2 cell injections, Mutant (het. Â het.) polaris expression pkd2 expression including curled tails, cardiac edema and cystic kidneys, were lokto237b 44/46 = 96% 38/39 = 97% seen in DFC-morphants produced from midblastula stage flhn1 35/35 = 100% 31/31 = 100% tc321 injections (Figs. 2C–M). cup 24/25 = 96% 52/61 = 85% oepm134 32/34 = 94% 27/27 = 100% As a control to test whether polaris and pkd2 morpholinos ty68b sur 37/42 = 88% 35/39 = 90% must enter DFCs to perturb LR development, we injected Pol SP cyctf219 35/35 = 100% 50/54 = 93% MO and Pkd2 SP MO2 into the yolk at late blastula stages (dome cycb229 38/42 = 90% 39/39 = 100% stage to 30% epiboly), when connections between the yolk and dintm84 62/65 = 95% 51/51 = 100% b641 DFCs are thought to be closed (Cooper and D’Amico, 1996). In smu 54/59 = 92% 48/48 = 100% momtl211 30/31 = 97% 64/72 = 89% these embryos, morpholino diffused throughout the yolk but did ta56 cas 31/46 = 67% 18/24 = 75% ntlb195 46/61 = 75% 26/26 = 100% Table 2b sptb104 24/25 = 96% 93/126 = 73% Effect of late blastula morpholino injections on the direction of heart and gut looping Gene expression in DFCs at 85–95% epiboly Treatment n Heart looping (%) Gut looping (%) Morphant polaris expression pkd2 expression Normal Reversed Normal Reversed ntl MO 78/145 = 54% 47/47 = 100% Uninjected control 196 99 1 99 1 Uninjected 112/122 = 92% 38/38 = 100% YolkPol SP MO2 (3 ng) 102 99 1 98 2 spt MO 71/87 = 82% 86/134 = 64% YolkPkd2 SP MO (4 ng) 79 96 4 95 5 Uninjected 112/121 = 93% 134/136 = 99% B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 283

(Alexander and Stainier, 1999). In all embryos derived from pronephric ducts and not necessarily an indication of a specific ntlb195 (n = 42) and sptb104 (n = 58) heterozygous crosses, role for Pkd2 or Polaris in the heart. Hydrocephalus in mice is sox17 was expressed in a similar sized group of cells in KV, associated with mutations in polaris, the dynein heavy chain indicating that although this organ is often dismorphic in these gene mdnah5 and a transcriptional regulator of ciliary genes, mutants (Amack and Yost, 2004; Essner et al., 2005), the cells hfh4 (Chen et al., 1998; Ibanez-Tallon et al., 2002; Taulman et comprising KV are present and retain at least some of their al., 2001). This condition is also present in some humans with normal characteristics. To confirm the role of ntl and spt in PCD (Ibanez-Tallon et al., 2003). Cystic kidneys and hydro- regulating polaris and pkd2 expression, we attempted to cephalus have both been attributed to defects in ciliary function phenocopy these mutants by injecting ntl and spt morpholinos (Ibanez-Tallon et al., 2004; Igarashi and Somlo, 2002; Kramer- at the 1–2 cell stage. Polaris expression in DFCs at 85–95% Zucker et al., 2005; Lakkis and Zhou, 2003; Pazour, 2004; epiboly was absent in 46% of ntl morphants while pkd2 Taulman et al., 2001). expression was unaffected. In contrast, pkd2 expression was Polaris and pkd2 morphants also exhibit alterations to body absent in 36% of spt morphants, while polaris expression was shape. Polaris morphants have ventrally curved tails, while in affected to a lesser extent. In contrast to what was observed in pkd2 morphants, the tail curls dorsally. These phenotypes are the mutants, ntl and spt morpholinos had only a partial effect consistent with previous reports of zebrafish polaris (oval) on abolishing polaris and pkd2 expression, respectively, mutants (Kramer-Zucker et al., 2005; Tsujikawa and Malicki, perhaps due to incomplete protein knockdown. However, the 2004), and pkd2 morphants (Sun et al., 2004). A number of morphants corroborate the data from analysis of mutants that zebrafish mutants with curly tail phenotypes have been reported polaris expression is regulated by ntl function and that pkd2 and many of these have additional phenotypes including midline expression is dependent on spt gene function. defects, kidney cysts and laterality defects. Although the developmental mechanisms involved are poorly understood, a Discussion curved body shape is often coincident with ventral spinal cord or other midline defects (Brand et al., 1996). Neither polaris nor Polaris and pkd2 functions are conserved in zebrafish pkd2 morphants have obvious midline defects but both genes are expressed early in tissues that will give rise to the embryonic Zebrafish homologues of mammalian Polaris and Pkd2 are midline: Polaris is maternally expressed and ubiquitous and 60–76% similar to their mammalian counterparts, suggesting pkd2 is expressed at high levels in the blastoderm margin and that many aspects of protein function are conserved. Like their subsequently in the hypoblast, the tissue that will give rise to the murine counterparts, these genes are ubiquitously expressed notochord and spinal floorplate. As such, subtle alterations in the during much of early embryogenesis (Fig. 1) and in particular function of midline cells may be responsible for the alterations in are enriched in ciliated tissues where these genes are critical for body shape in polaris and pkd2 morphants. Many curly tail development and function (Murcia et al., 2000; Pennekamp et mutants also have kidney cysts (Brand et al., 1996; Drummond al., 2002; Taulman et al., 2001). These ciliated tissues include et al., 1998; Sun et al., 2004). Kidney cyst formation does not, the pronephric duct primordia of the developing kidney and however, appear to be caused by body curvature, since many Kupffer’s vesicle, a structure proposed to be analogous to the curly tail mutants do not develop cysts in the kidney and cyst mouse node in LR development (Essner et al., 2002). formation often occurs prior to the development of axis Many aspects of the morphological phenotypes of zebrafish curvature (Drummond et al., 1998). As both polaris and pkd2 embryos injected at the 1–2 cell stage with polaris or pkd2 are highly expressed in the developing pronephric ducts, it is morpholinos, including kidney cysts and hydrocephalus, are likely that the kidney cysts in these morphants are a consequence analogous to phenotypes present in the respective murine of loss of ciliary function in the pronephric ducts. A curved body mutants. Both polaris and pkd2 mouse mutants were first shape is also often coincident with laterality defects (Brand et al., studied as models of polycystic kidney disease (Moyer et al., 1996; Chen et al., 1997), but is not absolutely predictive of 1994; Taulman et al., 2001; Wu et al., 1998, 2000). In laterality defects. In 17 mutant lines with curly tails and no mammals, kidney cysts are the result of grossly expanded obvious midline defects, only 8 lines had altered cardiac LR kidney tubule lumens. Similarly, in zebrafish, the anterior asymmetry (Chen et al., 1997). Polaris and pkd2 DFC- pronephric tubules and the glomerulus (Figs. 2F, G) are the morphants do not exhibit curved bodies, but have laterality initial site of cyst formation in all kidney cyst mutants reported defects (Fig. 6, Tables 2a and 2b), indicating that laterality (Drummond et al., 1998; Sun et al., 2004). Mutations in other defects in these morphants are independent of body curvature ciliary protein genes also cause cystic kidney phenotypes. and a consequence of alterations in the function of DFCs/KV. These genes include pkd1 (Yoder et al., 2002), inv (Otto et al., 2003), kif3A (Lin et al., 2003), cystin (Hou et al., 2002), genes Polaris and pkd2 are important in establishing the zebrafish implicated in Bardet–Biedl syndrome (Ansley et al., 2003; LR axis Blacque et al., 2004; Kim et al., 2004; Mykytyn et al., 2004) and other IFT protein genes (Kramer-Zucker et al., 2005; Sun A role for ciliary function in LR patterning has been et al., 2004). Cardiac edema observed in polaris and pkd2 postulated in mammals and more recently in zebrafish. morphants may be a consequence of perturbed osmoregulation Consistent with this, perturbation of polaris or pkd2 function resulting from the knockdown of ciliary function in the in mice (Murcia et al., 2000; Pennekamp et al., 2002) and 284 B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 zebrafish randomizes the directions of heart and gut looping evident by the normal expression of lft1 in the notochord (Figs. (Figs. 2J–M,Table 1). Alterations in gut and cardiac situs reflect 4A–D). Other markers of notochord (ntl) and floorplate (shh, altered expression of LR patterning genes in the LPM (Figs. 3, ptc1, axial) were also expressed normally in polaris (and pkd2) 4). Polaris and pkd2 morpholinos cause an increase in the morphants, indicating that in contrast to the mouse polaris percentage of embryos exhibiting right-sided, bilateral or absent mutant the midlines in these morphants were unaffected. expression of spaw, lft1 and lft2, genes normally expressed only The asymmetric gene expression phenotypes of zebrafish in the left LPM or diencephalon. The extent and nature of altered pkd2 morphants and mouse pkd2 mutants are similar; both gene expression patterns vary somewhat among the different exhibit a high percentage of embryos with no expression of morphants examined. In the 1–2 cell injections, the polaris and nodal (spaw)orlefty-2 in the LPM (Figs. 3, 4; Pennekamp pkd2 morpholinos that were effective at the lowest concentration et al., 2002). Interestingly, lefty-1 expression was absent (Pol SP MO2, Pkd2 SP MO) gave the highest percentage of from the floorplate of pkd2 mouse mutants, although marker embryos in which spaw, lft1 and lft2 expression was absent. This gene expression and histological examination suggested that may in part reflect the efficiency of protein knockdown by the the floorplate was intact (Pennekamp et al., 2002). In different morpholinos, since injecting higher amounts of each contrast, expression of lft1 and midline integrity in zebrafish these morpholinos increased the penetrance of laterality defects pkd2 morphants was unaffected. The loss of nodal (spaw) (data not shown). In this study, we chose morpholino concentra- and lefty expression in the LPM associated with loss of tions for which RT-PCR showed loss of correctly spliced RNAs pkd2 function in zebrafish and mice is consistent with a role where applicable (Figs. 2A, B), and gave a significant for this protein in transducing a signal which acts upstream percentage of embryos with curled tails and laterality pheno- of nodal in the LR axis specification pathway. The types, but did not cause phenotypes such as brain necrosis or developmental timing of this event in zebrafish and mice lethality. Although brain necrosis might be consistent with a is consistent with Pkd2 functioning as a mechanotransducer gene-specific knockdown of polaris or pkd2, both of which are of nodal/KV fluid flow. The fact that in both mice and expressed in the brain, this phenotype is difficult to distinguish zebrafish there is not a complete loss of nodal expression in from a phenotype that can be associated with injection of too 100% of mutants or morphants raises the possibility that a much morpholino (Nasevicius and Ekker, 2000). We wished to second protein has a partially redundant role in transmitting avoid any possibility that the laterality defects we observed were a signal form the node/KV to the LPM. a secondary consequence of non-specific developmental dis- tress, and as such the amounts of morpholino injected likely gave Role of polaris and pkd2 in KV cilia incomplete protein knockdown. A high degree of concordance was observed between the The functions of polaris, pkd2 and other ciliary genes pattern of lft1 in the diencephalon and lft2 expression in the including kif3A, kif3B, inv and lrd in establishing the LR axis LPM (e.g. both genes expressed on the right side) in individual of murine embryos are postulated to be mediated through their morphants (91% concordance across all polaris morphants, n = functions in monocilia within the mouse node (Marszalek et al., 339; 89% concordance across all pkd2 morphants, n = 205; 1999; Nonaka et al., 1998; Okada et al., 1999; Supp et al., Figs. 4E, F). This observation is consistent with the idea that 1997; Takeda et al., 1999). Acetylated tubulin labeling of the alterations of LR patterning due to knock-down of Polaris or cilia associated with zebrafish KV, a structure proposed to be Pkd2 function occurred early during LR axis establishment, analogous to the mouse node in LR development, revealed that upstream of the asymmetric expression of nodal signaling cilia were significantly shorter in polaris AUG-morphants (Fig. pathway genes in the LPM and diencephalon (Bisgrove and 5). In mouse polaris knock-out mutants, node cells fail to form Yost, 2001; Bisgrove et al., 2003). or maintain monocilia (Murcia et al., 2000). A role for polaris Alterations in the patterns of asymmetric gene expression in and its homologues in cilia assembly in other tissues and in zebrafish polaris morphants were different from those ob- other organisms is highly conserved. In the polaris Tg737orpk served in polaris KO mouse mutants. Zebrafish polaris insertion allele, primary cilia associated with the kidney are morphants exhibit randomized expression of spaw and lft2 in shorter than normal and ependymal cilia are absent, while in C. the LPM. In contrast, mouse polaris mutants exhibit bilateral elegans, loss of polaris (osm-5) function results in truncated expression of lefty-2 and nodal in the LPM (Murcia et al., sensory cell cilia, and in Chlamydomonas, loss of ift-88 2000). At E8.0, polaris knock-out mutants also have reduced function results in loss of flagella (Haycraft et al., 2001; Pazour expression of shh and a loss of hnf3B expression in the et al., 2000; Qin et al., 2001; Taulman et al., 2001). In addition floorplate (Murcia et al., 2000). The bilateral expression of to a role in ciliary assembly, Polaris also functions to transport nodal and lefty-2 in the LPM likely reflects these floorplate protein cargo along cilia. In C. elegans, OSM-50GFP (Polaris) defects as an intact floorplate expressing the nodal inhibitor particles show anterograde and retrograde migration along lefty-1 is essential in preventing left-sided signals from sensory cell cilia, and in osm-5 mutants, LOV10GFP (Pkd1) crossing the midline and activating nodal and lefty-2 expres- and PKD20GFP accumulate at the distal tips of cilia indicating sion on the right side (Meno et al., 1998). Although polaris a disruption in retrograde transport (Haycraft et al., 2001; Qin (and pkd2) morphants exhibit curled tails, a phenotype often et al., 2001). associated with midline defects (Brand et al., 1996), the We suggest that the shortened cilia seen in Pol AUG MO midline was not obviously perturbed in these morphants as morphants reflect a knockdown in the amount of Polaris B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 285 translated from maternal polaris RNA. The alterations in LR (Fig. 6B). Polaris and pkd2 DFC-morphants showed patterning in these morphants likely reflect a combination of corresponding reversals in heart and gut looping (Tables 2a shortened cilia which would be expected to alter KV fluid and 2b). The weak penetrance of the laterality phenotype flow, and a block in IFT of factors necessary for ciliary relative to that observed from morpholino injections at the 1–2 signaling, potentially including PKD2. In Pol SP MO cell stage may be due to inefficiencies of DFC-targeted morphants, cilia length was relatively unaffected, presumably injections, such as variability in the amount of morpholino due to the presence of maternal Polaris and therefore the that enters DFCs from the yolk and morpholino not entering all laterality phenotype is likely due to a block in IFT. No DFCs in a given embryo. It is not possible to determine obvious difference in KV cilia length or morphology was whether Polaris and Pkd2 have any function specific to DFCs, seen in pkd2 morphants. Similarly, the structure of node but it is more likely that these proteins are required once DFCs monocilia is unaffected in mouse pkd2 mutants and sensory and their progenitors in KV become ciliated. DFCs have no cilia are normal in C. elegans pkd2 mutants (Barr et al., 2001; known developmental function other than their role in giving McGrath et al., 2003). The lack of a morphological phenotype rise to KV (Warga and Stainier, 2002). Ablation of these cells in cilia of pkd2 morphants and mutants is consistent with its during late gastrula stages results in the loss or altered proposed role as a calcium ion channel and not a structural morphogenesis of KV and concomitant LR patterning defects, component of cilia. but has no other effect on anteriorposterior, dorsoventral or midline development (Essner et al., 2005). In summary, Polaris and pkd2 function in Kupffer’s vesicle to control LR although a role for polaris and pkd2 in LR patterning has patterning been known for some time, by using a technique to target morpholinos to DFC/KV cells, we provide the first direct The phenotypes of mouse mutants and zebrafish morphants evidence in any vertebrate that polaris and pkd2 function in suggest that Polaris and Pkd2 function in node and KV cilia to ciliated cells in the node/KV to specify the LR axis. control laterality. However, several observations leave open the possibility that node/KV cilia are not central to LR specifica- Genetic regulation of polaris and pkd2 expression tion. Both polaris and pkd2 are broadly expressed in mice and zebrafish, and murine mutants have pleiotropic phenotypes, Mutations in a large number of genes in mice and zebrafish including midline defects, that could be responsible for cause laterality defects and these defects can be broadly alterations in LR patterning. Both proteins may also have grouped into different classes based on alterations in asym- functions unrelated to cilia. For example, Polaris has recently metric gene expression patterns (Bisgrove and Yost, 2001). For been implicated as a modulator of Hedgehog (Hh) signaling some classes, the mutant phenotype can be explained due to (Huangfu et al., 2003) and Hh signaling has been implicated in loss of midline structures or defects in nodal signaling, but in LR patterning in mice and chick (Levin et al., 1995; Meyers others, the mechanism is unclear (Bisgrove et al., 2000). To ask and Martin, 1999; Tsukui et al., 1999; Zhang et al., 2001). It is if alterations in polaris or pkd2 expression might play a role in unlikely that the role for polaris in zebrafish LR patterning is the phenotypes of zebrafish laterality mutants, we assayed for through an effect on Hh signaling since ptc1 expression expression of these genes in 13 mutant lines at bud stage. appears unaltered in polaris morphants (data not shown), and Expression of polaris was greatly reduced or absent in KV of several zebrafish mutants with defects in the Hh pathway have about 25% of the embryos from a ntl heterozygous intercross normal LR patterning (Chen et al., 1997; Chen et al., 2001). and pkd2 expression was similarly affected in embryos from a Neither polaris nor pkd2 has been knocked out exclusively in spt intercross (Table 3). Expression of polaris and pkd2 in the mouse node, so even if the LR patterning defects are other regions of the embryos was unaffected in these mutants. specific to a ciliary function for these genes, it is formally Analysis of sox17 expression indicates that, in both ntl and spt possible that cilia found in other embryonic tissues such as mutant embryos, DFC derivatives in KV are present and midline (Sulik et al., 1994) are responsible for the LR express some of their normal genetic program. In wild-type phenotype in mice. embryos, ntl and spt are expressed in DFCs and throughout One of the unique advantages of studying polaris and pkd2 KV (Melby et al., 1996; our unpublished observations). In both in zebrafish is that gene function can be knocked down ntl and spt mutant embryos, KV primordia are present and specifically in DFCs, the progenitors of ciliated cells in KV, by although KV is often dismorphic the cells express cilia at later injecting morpholino into the yolk at midblastula stages stages of development (Amack and Yost, 2004; Essner et al., (Amack and Yost, 2004). While targeting of morpholinos to 2005) further indicating that much of the genetic program of DFCs is not 100% efficient, this technique has provided a these cells is intact. Thus, our analysis of polaris and pkd2 valuable approach in examining DFC/KV-specific gene func- expression in ntl and spt mutants and morphants suggests that tion (Amack and Yost, 2004; Essner et al., 2005). Knockdown Ntl directly modulates expression of polaris while Spt directly of Polaris and Pkd2 in DFCs/KV resulted in alterations of the modulates expression of pkd2. Ntl has recently been shown to normal left-sided expression of lft1 and lft2.Left-sided also regulate transcription of lrdr1 in DFCs (Essner et al., expression of lft1 in the diencephalon of DFC-morphants 2005), suggesting that T-box transcription factors have a ranged from 61–85% (Fig. 6A) while left-sided expression of previously unrecognized role in regulating genes involved in lft2 in the LPM of DFC-morphants ranged from 64%–83% ciliary structure and function. 286 B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288

Transcriptional regulation of ciliary genes may be complex. development by antagonism of lefty and nodal signaling. Development In C. elegans, polaris (osm-5) expression is under direct 126, 3253–3262. Bisgrove, B.W., Essner, J.J., Yost, H.J., 2000. Multiple pathways in the midline control of the RFX-type transcription factor daf-19, while pkd2 regulate concordant brain, heart and gut left–right asymmetry. Develop- is directly regulated by egl-46, a zinc-finger transcription factor ment 127, 3567–3579. (Haycraft et al., 2001; Yu et al., 2003). In mice, RFX3 has been Bisgrove, B.W., Morelli, S.H., Yost, H.J., 2003. Genetics of human laterality shown to regulate the transcription of D2liC, a gene encoding disorders: insights from vertebrate model systems. Annu. Rev. Genomics an IFT protein required for maintenance of node cilia (Bonnafe Hum. Genet. 4, 1–32. Blacque, O.E., Reardon, M.J., Li, C., McCarthy, J., Mahjoub, M.R., Ansley, et al., 2004). It will be interesting to assess the complexity of S.J., Badano, J.L., Mah, A.K., Beales, P.L., Davidson, W.S., Johnsen, R.C., these regulatory cascades to determine the role of individual Audeh, M., Plasterk, R.H., Baillie, D.L., Katsanis, N., Quarmby, L.M., transcription factors in the development of KV cilia and to Wicks, S.R., Leroux, M.R., 2004. Loss of C. elegans BBS-7 and BBS- assess the evolutionary conservation of the transcriptional 8 protein function results in cilia defects and compromised intraflagellar regulatory network that controls ciliogenesis. transport. Genes Dev. 18, 1630–1642. Bonnafe, E., Touka, M., AitLounis, A., Baas, D., Barras, E., Ucla, C., Moreau, A., Flamant, F., Dubruille, R., Couble, P., Collignon, J., Durand, B., Reith, Acknowledgments W., 2004. The transcription factor RFX3 directs nodal cilium development and left–right asymmetry specification. Mol. Cell. Biol. 24, 4417–4427. We thank J. D. Amack and W. W. Branford, for their Brand, M., Heisenberg, C.P., Warga, R.M., Pelegri, F., Karlstrom, R.O., thoughtful comments on the manuscript, and J. N. Campbell for Beuchle, D., Picker, A., Jiang, Y.J., Furutani-Seiki, M., van Eeden, F.J., Granato, M., Haffter, P., Hammerschmidt, M., Kane, D.A., Kelsh, R.N., expert technical assistance. This research was supported by Mullins, M.C., Odenthal, J., Nusslein-Volhard, C., 1996. Mutations grants from NHLBI and the Huntsman Cancer Foundation to affecting development of the midline and general body shape during HJY. zebrafish embryogenesis. Development 123, 129–142. Burdine, R.D., Schier, A.F., 2000. Conserved and divergent mechanisms in left –right axis formation. Genes Dev. 14, 763–776. Appendix A. Supplementary data Cahalan, M.D., 2002. The ins and outs of polycystin-2 as a calcium release channel. Nat. Cell Biol. 4, E56–E57. Supplementary data associated with this article can be Cai, Y., Maeda, Y., Cedzich, A., Torres, V.E., Wu, G., Hayashi, T., Mochizuki, found, in the online version, at doi:10.1016/j.ydbio.2005. T., Park, J.H., Witzgall, R., Somlo, S., 1999. Identification and character- 08.047. ization of polycystin-2, the PKD2 gene product. J. Biol. Chem. 274, 28557–28565. Capdevila, J., Vogan, K.J., Tabin, C.J., Izpisua Belmonte, J.C., 2000. References Mechanisms of left–right determination in vertebrates. Cell 101, 9–21. Chen, J.N., van Eeden, F.J., Warren, K.S., Chin, A., Nusslein-Volhard, C., Afzelius, B.A., 1976. A human syndrome caused by immotile cilia. Science Haffter, P., Fishman, M.C., 1997. Left–right pattern of cardiac BMP4 may 193, 317–319. drive asymmetry of the heart in zebrafish. Development 124, 4373–4382. Alexander, J., Stainier, D.Y., 1999. A molecular pathway leading to endoderm Chen, J., Knowles, H.J., Hebert, J.L., Hackett, B.P., 1998. Mutation of the formation in zebrafish. Curr. Biol. 9, 1147–1157. mouse hepatocyte nuclear factor/forkhead homologue 4 gene results in an Alexander, J., Rothenberg, M., Henry, G.L., Stainier, D.Y., 1999. casanova absence of cilia and random left–right asymmetry. J. Clin. Invest. 102, plays an early and essential role in endoderm formation in zebrafish. Dev. 1077–1082. Biol. 215, 343–357. Chen, W., Burgess, S., Hopkins, N., 2001. Analysis of the zebrafish Amack, J.D., Yost, H.J., 2004. The T box transcription factor no tail in smoothened mutant reveals conserved and divergent functions of hedgehog ciliated cells controls zebrafish left–right asymmetry. Curr. Biol. 14, activity. Development 128, 2385–2396. 685–690. Cole, D.G., Diener, D.R., Himelblau, A.L., Beech, P.L., Fuster, J.C., Ansley, S.J., Badano, J.L., Blacque, O.E., Hill, J., Hoskins, B.E., Leitch, C.C., Rosenbaum, J.L., 1998. Chlamydomonas kinesin-II-dependent intraflagel- Kim, J.C., Ross, A.J., Eichers, E.R., Teslovich, T.M., Mah, A.K., Johnsen, lar transport (IFT): IFT particles contain proteins required for ciliary R.C., Cavender, J.C., Lewis, R.A., Leroux, M.R., Beales, P.L., Katsanis, N., assembly in Caenorhabditis elegans sensory neurons. J. Cell Biol. 141, 2003. dysfunction is a likely cause of pleiotropic Bardet–Biedl 993–1008. syndrome. Nature 425, 628–633. Cooper, M.S., D’Amico, L.A., 1996. A cluster of noninvoluting endocytic cells Barr, M.M., Sternberg, P.W., 1999. A polycystic kidney-disease gene at the margin of the zebrafish blastoderm marks the site of embryonic shield homologue required for male mating behaviour in C. elegans. Nature formation. Dev. Biol. 180, 184–198. 401, 386–389. Drummond, I.A., Majumdar, A., Hentschel, H., Elger, M., Solnica-Krezel, L., Barr, M.M., DeModena, J., Braun, D., Nguyen, C.Q., Hall, D.H., Sternberg, Schier, A.F., Neuhauss, S.C., Stemple, D.L., Zwartkruis, F., Rangini, Z., P.W., 2001. The Caenorhabditis elegans autosomal dominant polycystic Driever, W., Fishman, M.C., 1998. Early development of the zebrafish kidney disease gene homologs lov-1 and pkd-2 act in the same pathway. pronephros and analysis of mutations affecting pronephric function. Curr. Biol. 11, 1341–1346. Development 125, 4655–4667. Bartoloni, L., Blouin, J.L., Pan, Y., Gehrig, C., Maiti, A.K., Scamuffa, N., Essner, J.J., Branford, W.W., Zhang, J., Yost, H.J., 2000. Mesendoderm and Rossier, C., Jorissen, M., Armengot, M., Meeks, M., Mitchison, H.M., left –right brain, heart and gut development are differentially regulated by Chung, E.M., Delozier-Blanchet, C.D., Craigen, W.J., Antonarakis, S.E., pitx2 isoforms. Development 127, 1081–1093. 2002. Mutations in the DNAH11 (axonemal heavy chain dynein type 11) Essner, J.J., Vogan, K.J., Wagner, M.K., Tabin, C.J., Yost, H.J., Brueckner, M., gene cause one form of situs inversus totalis and most likely primary ciliary 2002. Conserved function for embryonic nodal cilia. Nature 418, 37–38. dyskinesia. Proc. Natl. Acad. Sci. U. S. A. 99, 10282–10286. Essner, J.J., Amack, J.D., Nyholm, M.K., Harris, E.B., Yost, H.J., 2005. Beddington, R.S., Robertson, E.J., 1999. Axis development and early Kupffer’s vesicle is a ciliated organ of asymmetry in the zebrafish embryo asymmetry in mammals. Cell 96, 195–209. that initiates left–right development of the brain, heart and gut. Develop- Bisgrove, B.W., Yost, H.J., 2001. Classification of left –right patterning defects ment 132, 1247–1260. in zebrafish, mice, and humans. Am. J. Med. Genet. 101, 315–323. Guichard, C., Harricane, M.C., Lafitte, J.J., Godard, P., Zaegel, M., Tack, V., Bisgrove, B.W., Essner, J.J., Yost, H.J., 1999. Regulation of midline Lalau, G., Bouvagnet, P., 2001. Axonemal dynein intermediate-chain gene B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288 287

(DNAI1) mutations result in situs inversus and primary ciliary dyskinesia populations of node monocilia initiate left–right asymmetry in the mouse. (Kartagener syndrome). Am. J. Hum. Genet. 68, 1030–1035. Cell 114, 61–73. Hanaoka, K., Qian, F., Boletta, A., Bhunia, A.K., Piontek, K., Tsiokas, L., Melby, A.E., Warga, R.M., Kimmel, C.B., 1996. Specification of cell fates at Sukhatme, V.P., Guggino, W.B., Germino, G.G., 2000. Co-assembly of the dorsal margin of the zebrafish gastrula. Development 122, 2225–2237. polycystin-1 and -2 produces unique cation-permeable currents. Nature 408, Meno, C., Shimono, A., Saijoh, Y., Yashiro, K., Mochida, K., Ohishi, S., Noji, 990–994. S., Kondoh, H., Hamada, H., 1998. lefty-1 is required for left–right Haycraft, C.J., Swoboda, P., Taulman, P.D., Thomas, J.H., Yoder, B.K., 2001. determination as a regulator of lefty-2 and nodal. Cell 94, 287–297. The C. elegans homolog of the murine cystic kidney disease gene Tg737 Meyers, E.N., Martin, G.R., 1999. Differences in left–right axis pathways in functions in a ciliogenic pathway and is disrupted in osm-5 mutant worms. mouse and chick: functions of FGF8 and SHH. Science 285, 403–406. Development 128, 1493–1505. Moyer, J.H., Lee-Tischler, M.J., Kwon, H.Y., Schrick, J.J., Avner, E.D., Hou, X., Mrug, M., Yoder, B.K., Lefkowitz, E.J., Kremmidiotis, G., Sweeney, W.E., Godfrey, V.L., Cacheiro, N.L., Wilkinson, J.E., Woychik, D’Eustachio, P., Beier, D.R., Guay-Woodford, L.M., 2002. Cystin, a novel R.P., 1994. Candidate gene associated with a mutation causing recessive cilia-associated protein, is disrupted in the cpk mouse model of polycystic polycystic kidney disease in mice. Science 264, 1329–1333. kidney disease. J. Clin. Invest. 109, 533–540. Murcia, N.S., Richards, W.G., Yoder, B.K., Mucenski, M.L., Dunlap, J.R., Huangfu, D., Liu, A., Rakeman, A.S., Murcia, N.S., Niswander, L., Anderson, Woychik, R.P., 2000. The Oak Ridge Polycystic Kidney (orpk) disease K.V., 2003. Hedgehog signalling in the mouse requires intraflagellar gene is required for left–right axis determination. Development 127, transport proteins. Nature 426, 83–87. 2347–2355. Ibanez-Tallon, I., Gorokhova, S., Heintz, N., 2002. Loss of function of Mykytyn, K., Mullins, R.F., Andrews, M., Chiang, A.P., Swiderski, R.E., Yang, axonemal dynein Mdnah5 causes primary ciliary dyskinesia and hydro- B., Braun, T., Casavant, T., Stone, E.M., Sheffield, V.C., 2004. Bardet– cephalus. Hum. Mol. Genet. 11, 715–721. Biedl syndrome type 4 (BBS4)-null mice implicate Bbs4 in flagella Ibanez-Tallon, I., Heintz, N., Omran, H., 2003. To beat or not to beat: formation but not global cilia assembly. Proc. Natl. Acad. Sci. U. S. A. roles of cilia in development and disease. Hum. Mol. Genet. 12 (1), 101, 8664–8669. R27–R35. Nasevicius, A., Ekker, S.C., 2000. Effective targeted gene Fknockdown_ in Ibanez-Tallon, I., Pagenstecher, A., Fliegauf, M., Olbrich, H., Kispert, A., zebrafish. Nat. Genet. 26, 216–220. Ketelsen, U.P., North, A., Heintz, N., Omran, H., 2004. Dysfunction of Nauli, S.M., Alenghat, F.J., Luo, Y., Williams, E., Vassilev, P., Li, X., Elia, axonemal dynein heavy chain Mdnah5 inhibits ependymal flow and reveals A.E., Lu, W., Brown, E.M., Quinn, S.J., Ingber, D.E., Zhou, J., 2003. a novel mechanism for hydrocephalus formation. Hum. Mol. Genet. 13, Polycystins 1 and 2 mediate mechanosensation in the primary cilium of 2133–2141. kidney cells. Nat. Genet. 33, 129–137. Igarashi, P., Somlo, S., 2002. Genetics and pathogenesis of polycystic kidney Nonaka, S., Tanaka, Y., Okada, Y., Takeda, S., Harada, A., Kanai, Y., Kido, M., disease. J. Am. Soc. Nephrol. 13, 2384–2398. Hirokawa, N., 1998. Randomization of left–right asymmetry due to loss of Karlstrom, R.O., Talbot, W.S., Schier, A.F., 1999. Comparative synteny cloning nodal cilia generating leftward flow of extraembryonic fluid in mice lacking of zebrafish you-too: mutations in the Hedgehog target gli2 affect ventral KIF3B motor protein. Cell 95, 829–837. forebrain patterning. Genes Dev. 13, 388–393. Nonaka, S., Shiratori, H., Saijoh, Y., Hamada, H., 2002. Determination of left– Kim, J.C., Badano, J.L., Sibold, S., Esmail, M.A., Hill, J., Hoskins, B.E., right patterning of the mouse embryo by artificial nodal flow. Nature 418, Leitch, C.C., Venner, K., Ansley, S.J., Ross, A.J., Leroux, M.R., Katsanis, 96–99. N., Beales, P.L., 2004. The Bardet–Biedl protein BBS4 targets cargo to the Odenthal, J., Nusslein-Volhard, C., 1998. Fork head domain genes in zebrafish. pericentriolar region and is required for microtubule anchoring and cell Dev. Genes Evol. 208, 245–258. cycle progression. Nat. Genet. 36, 462–470. Okada, Y., Nonaka, S., Tanaka, Y., Saijoh, Y., Hamada, H., Hirokawa, N., 1999. Koulen, P., Cai, Y., Geng, L., Maeda, Y., Nishimura, S., Witzgall, R., Ehrlich, Abnormal nodal flow precedes situs inversus in iv and inv mice. Mol. Cell B.E., Somlo, S., 2002. Polycystin-2 is an intracellular calcium release 4, 459–468. channel. Nat. Cell Biol. 4, 191–197. Olbrich, H., Haffner, K., Kispert, A., Volkel, A., Volz, A., Sasmaz, G., Kozminski, K.G., Johnson, K.A., Forscher, P., Rosenbaum, J.L., 1993. A Reinhardt, R., Hennig, S., Lehrach, H., Konietzko, N., Zariwala, M., motility in the eukaryotic flagellum unrelated to flagellar beating. Proc. Noone, P.G., Knowles, M., Mitchison, H.M., Meeks, M., Chung, E.M., Natl. Acad. Sci. U. S. A. 90, 5519–5523. Hildebrandt, F., Sudbrak, R., Omran, H., 2002. Mutations in DNAH5 cause Kramer-Zucker, A.G., Olale, F., Haycraft, C.J., Yoder, B.K., Schier, A.F., primary ciliary dyskinesia and randomization of left–right asymmetry. Nat. Drummond, I.A., 2005. Cilia-driven fluid flow in the zebrafish pronephros, Genet. 30, 143–144. brain and Kupffer’s vesicle is required for normal organogenesis. Otto, E.A., Schermer, B., Obara, T., O’Toole, J.F., Hiller, K.S., Mueller, A.M., Development 132, 1907–1921. Ruf, R.G., Hoefele, J., Beekmann, F., Landau, D., Foreman, J.W., Krauss, S., Concordet, J.P., Ingham, P.W., 1993. A functionally conserved Goodship, J.A., Strachan, T., Kispert, A., Wolf, M.T., Gagnadoux, M.F., homolog of the Drosophila segment polarity gene hh is expressed in tissues Nivet, H., Antignac, C., Walz, G., Drummond, I.A., Benzing, T., with polarizing activity in zebrafish embryos. Cell 75, 1431–1444. Hildebrandt, F., 2003. Mutations in INVS encoding inversin cause Lakkis, M., Zhou, J., 2003. Molecular complexes formed with polycystins. nephronophthisis type 2, linking renal cystic disease to the function of Nephron, Exp. Nephrol. 93, e3–e8. primary cilia and left–right axis determination. Nat. Genet. 34, 413–420. Levin, M., Johnson, R.L., Stern, C.D., Kuehn, M., Tabin, C., 1995. A Pazour, G.J., 2004. and cilia-dependent renal disease: molecular pathway determining left–right asymmetry in chick embryo- the ciliary hypothesis of polycystic kidney disease. J. Am. Soc. Nephrol. 15, genesis. Cell 82, 803–814. 2528–2536. Lin, F., Hiesberger, T., Cordes, K., Sinclair, A.M., Goldstein, L.S., Somlo, S., Pazour, G.J., Wilkerson, C.G., Witman, G.B., 1998. A dynein light chain is Igarashi, P., 2003. Kidney-specific inactivation of the KIF3A subunit of essential for the retrograde particle movement of intraflagellar transport kinesin-II inhibits renal ciliogenesis and produces polycystic kidney (IFT). J. Cell Biol. 141, 979–992. disease. Proc. Natl. Acad. Sci. U. S. A. 100, 5286–5291. Pazour, G.J., Dickert, B.L., Vucica, Y., Seeley, E.S., Rosenbaum, J.L., Witman, Long, S., Ahmad, N., Rebagliati, M., 2003. The zebrafish nodal-related gene G.B., Cole, D.G., 2000. Chlamydomonas IFT88 and its mouse homologue, southpaw is required for visceral and diencephalic left–right asymmetry. polycystic kidney disease gene tg737, are required for assembly of cilia and Development 130, 2303–2316. flagella. J. Cell Biol. 151, 709–718. Marszalek, J.R., Ruiz-Lozano, P., Roberts, E., Chien, K.R., Goldstein, L.S., Pazour, G.J., Baker, S.A., Deane, J.A., Cole, D.G., Dickert, B.L., Rosenbaum, 1999. Situs inversus and embryonic ciliary morphogenesis defects in J.L., Witman, G.B., Besharse, J.C., 2002. The intraflagellar transport mouse mutants lacking the KIF3A subunit of kinesin-II. Proc. Natl. protein, IFT88, is essential for vertebrate photoreceptor assembly and Acad. Sci. U. S. A. 96, 5043–5048. maintenance. J. Cell Biol. 157, 103–113. McGrath, J., Somlo, S., Makova, S., Tian, X., Brueckner, M., 2003. Two Pennarun, G., Escudier, E., Chapelin, C., Bridoux, A.M., Cacheux, V., Roger, 288 B.W. Bisgrove et al. / Developmental Biology 287 (2005) 274–288

G., Clement, A., Goossens, M., Amselem, S., Duriez, B., 1999. Loss-of- insights in determination of laterality and mesoderm induction by function mutations in a human gene related to Chlamydomonas reinhardtii kif3AÀ/À mice analysis. J. Cell Biol. 145, 825–836. dynein IC78 result in primary ciliary dyskinesia. Am. J. Hum. Genet. 65, Taulman, P.D., Haycraft, C.J., Balkovetz, D.F., Yoder, B.K., 2001. Polaris, a 1508–1519. protein involved in left –right axis patterning, localizes to basal bodies and Pennekamp, P., Karcher, C., Fischer, A., Schweickert, A., Skryabin, B., cilia. Mol. Biol. Cell 12, 589–599. Horst, J., Blum, M., Dworniczak, B., 2002. The ion channel polycystin- Thisse, C., Thisse, B., 1999. Antivin, a novel and divergent member of the 2 is required for left –right axis determination in mice. Curr. Biol. 12, TGFh superfamily, negatively regulates mesoderm induction. Development 938–943. 126, 229–240. Peters, D.J., Spruit, L., Saris, J.J., Ravine, D., Sandkuijl, L.A., Fossdal, R., Tsujikawa, M., Malicki, J., 2004. Intraflagellar transport genes are essential Boersma, J., van Eijk, R., Norby, S., Constantinou-Deltas, C.D., et al., for differentiation and survival of vertebrate sensory neurons. Neuron 42, 1993. Chromosome 4 localization of a second gene for autosomal dominant 703–716. polycystic kidney disease. Nat. Genet. 5, 359–362. Tsukui, T., Capdevila, J., Tamura, K., Ruiz-Lozano, P., Rodriguez-Esteban, Praetorius, H.A., Spring, K.R., 2001. Bending the MDCK cell primary cilium C., Yonei-Tamura, S., Magallon, J., Chandraratna, R.A., Chien, K., increases intracellular calcium. J. Membr. Biol. 184, 71–79. Blumberg, B., Evans, R.M., Belmonte, J.C., 1999. Multiple left–right Praetorius, H.A., Spring, K.R., 2003. Removal of the MDCK cell primary asymmetry defects in Shh(À/À) mutant mice unveil a convergence of cilium abolishes flow sensing. J. Membr. Biol. 191, 69–76. the shh and retinoic acid pathways in the control of Lefty-1. Proc. Natl. Qian, F., Germino, F.J., Cai, Y., Zhang, X., Somlo, S., Germino, G.G., 1997. Acad. Sci. U. S. A. 96, 11376–11381. PKD1 interacts with PKD2 through a probable coiled-coil domain. Nat. Turner, D.L., Weintraub, H., 1994. Expression of achaete-scute homolog 3 in Genet. 16, 179–183. Xenopus embryos converts ectodermal cells to a neural fate. Genes Dev. 8, Qin, H., Rosenbaum, J.L., Barr, M.M., 2001. An autosomal recessive 1434–1447. polycystic kidney disease gene homolog is involved in intraflagellar Warga, R.M., Stainier, D.Y., 2002. The guts of endoderm formation. Results transport in C. elegans ciliated sensory neurons. Curr. Biol. 11, 457–461. Probl. Cell Differ. 40, 28–47. Rebagliati, M.R., Toyama, R., Fricke, C., Haffter, P., Dawid, I.B., 1998a. Weinberg, E.S., Allende, M.L., Kelly, C.S., Abdelhamid, A., Murakami, T., Zebrafish nodal-related genes are implicated in axial patterning and Andermann, P., Doerre, O.G., Grunwald, D.J., Riggleman, B., 1996. establishing left–right asymmetry. Dev. Biol. 199, 261–272. Developmental regulation of zebrafish MyoD in wild-type, no tail and Rebagliati, M.R., Toyama, R., Haffter, P., Dawid, I.B., 1998b. Cyclops spadetail embryos. Development 122, 271–280. encodes a nodal-related factor involved in midline signaling. Proc. Natl. Westerfield, M., 1995. The Zebrafish Book: A Guide for the Laboratory Use of Acad. Sci. U. S. A. 95, 9932–9937. Zebrafish. University of Oregon Press, Eugene, OR. Rupp, R.A., Snider, L., Weintraub, H., 1994. Xenopus embryos regulate the Wu, G., D’Agati, V., Cai, Y., Markowitz, G., Park, J.H., Reynolds, D.M., nuclear localization of XMyoD. Genes Dev. 8, 1311–1323. Maeda, Y., Le, T.C., Hou, H. Jr., Kucherlapati, R., Edelmann, W., Somlo, Sampath, K., Rubinstein, A.L., Cheng, A.M., Liang, J.O., Fekany, K., Solnica- S., 1998. Somatic inactivation of Pkd2 results in polycystic kidney disease. Krezel, L., Korzh, V., Halpern, M.E., Wright, C.V., 1998. Induction of the Cell 93, 177–188. zebrafish ventral brain and floorplate requires cyclops/nodal signalling. Wu, G., Markowitz, G.S., Li, L., D’Agati, V.D., Factor, S.M., Geng, L., Tibara, Nature 395, 185–189. S., Tuchman, J., Cai, Y., Park, J.H., van Adelsberg, J., Hou Jr., H., Schulte-Merker, S., van Eeden, F.J., Halpern, M.E., Kimmel, C.B., Nusslein- Kucherlapati, R., Edelmann, W., Somlo, S., 2000. Cardiac defects and renal Volhard, C., 1994. No tail (ntl) is the zebrafish homologue of the mouse T failure in mice with targeted mutations in Pkd2. Nat. Genet. 24, 75–78. (Brachyury) gene. Development 120, 1009–1015. Xu, G.M., Gonzalez-Perrett, S., Essafi, M., Timpanaro, G.A., Montalbetti, N., Strahle, U., Blader, P., Henrique, D., Ingham, P.W., 1993. Axial, a zebrafish Arnaout, M.A., Cantiello, H.F., 2003. Polycystin-1 activates and stabilizes gene expressed along the developing body axis, shows altered expression in the polycystin-2 channel. J. Biol. Chem. 278, 1457–1462. cyclops mutant embryos. Genes Dev. 7, 1436–1446. Yelon, D., Horne, S.A., Stainier, D.Y., 1999. Restricted expression of cardiac Sulik, K., Dehart, D.B., Iangaki, T., Carson, J.L., Vrablic, T., Gesteland, K., myosin genes reveals regulated aspects of heart tube assembly in zebrafish. Schoenwolf, G.C., 1994. Morphogenesis of the murine node and Dev. Biol. 214, 23–37. notochordal plate. Dev. Dyn. 201, 260–278. Yoder, B.K., Richards, W.G., Sweeney, W.E., Wilkinson, J.E., Avener, E.D., Sun, Z., Amsterdam, A., Pazour, G.J., Cole, D.G., Miller, M.S., Hopkins, N., Woychik, R.P., 1995. Insertional mutagenesis and molecular analysis of a 2004. A genetic screen in zebrafish identifies cilia genes as a principal cause new gene associated with polycystic kidney disease. Proc. Assoc. Am. of cystic kidney. Development 131, 4085–4093. Physicians 107, 314–323. Supp, D.M., Witte, D.P., Potter, S.S., Brueckner, M., 1997. Mutation of an Yoder, B.K., Hou, X., Guay-Woodford, L.M., 2002. The polycystic kidney axonemal dynein affects left–right asymmetry in inversus viscerum mice. disease proteins, polycystin-1, polycystin-2, polaris, and cystin, are co- Nature 389, 963–966. localized in renal cilia. J. Am. Soc. Nephrol. 13, 2508–2516. Supp, D.M., Brueckner, M., Kuehn, M.R., Witte, D.P., Lowe, L.A., Yu, H., Pretot, R.F., Burglin, T.R., Sternberg, P.W., 2003. Distinct roles of McGrath, J., Corrales, J., Potter, S.S., 1999. Targeted deletion of the transcription factors EGL-46 and DAF-19 in specifying the functionality ATP binding domain of left–right dynein confirms its role in of a polycystin-expressing sensory neuron necessary for C. elegans specifying development of left–right asymmetries. Development 126, male vulva location behavior. Development 130, 5217–5227. 5495–5504. Zhang, X.M., Ramalho-Santos, M., McMahon, A.P., 2001. Smoothened Takeda, S., Yonekawa, Y., Tanaka, Y., Okada, Y., Nonaka, S., Hirokawa, N., mutants reveal redundant roles for Shh and Ihh signaling including 1999. Left–right asymmetry and kinesin superfamily protein KIF3A: new regulation of L/R symmetry by the mouse node. Cell 106, 781–792.