Developmental Biology 367 (2012) 15–24

Contents lists available at SciVerse ScienceDirect

Developmental Biology

0012-16

http://d

n Corr

E-m

journal homepage: www.elsevier.com/locate/developmentalbiology

Defective Nodal and Cerl2 expression in the Arl13bhnn mutant node underlieits heterotaxia

Christine E. Larkins a,b, Alyssa Bushey Long a, Tamara Caspary a,n

a Department of Human Genetics, Emory University School of Medicine, 615 Michael Street, Suite 301, Atlanta, GA 30322, USAb Graduate Program in Biochemistry, Cell and Developmental Biology, Emory University, Atlanta, GA, USA

a r t i c l e i n f o

Article history:

Received 23 August 2011

Received in revised form

5 April 2012

Accepted 6 April 2012Available online 24 April 2012

Keywords:

Nodal

Cerl2

Arl13b

Left–right axis

06/$ - see front matter & 2012 Elsevier Inc. A

x.doi.org/10.1016/j.ydbio.2012.04.011

esponding author. Fax: þ 404 727 3949.

ail address: tcaspar@emory.edu (T. Caspary).

a b s t r a c t

Specification of the left–right axis during embryonic development is critical for the morphogenesis of

asymmetric organs such as the heart, lungs, and stomach. The first known left–right asymmetry to

occur in the mouse embryo is a leftward fluid flow in the node that is created by rotating cilia on the

node surface. This flow is followed by asymmetric expression of Nodal and its inhibitor Cerl2 in the

node. Defects in cilia and/or fluid flow in the node lead to defective Nodal and Cerl2 expression and

therefore incorrect visceral organ situs. Here we show the cilia protein Arl13b is required for left right

axis specification as its absence results in heterotaxia. We find the defect originates in the node where

Cerl2 is not downregulated and asymmetric expression of Nodal is not maintained resulting in

symmetric expression of both genes. Subsequently, Nodal expression is delayed in the lateral plate

mesoderm (LPM). Symmetric Nodal and Cerl2 in the node could result from defects in either the

generation and/ or the detection of Nodal flow, which would account for the subsequent defects in the

LPM and organ positioning.

& 2012 Elsevier Inc. All rights reserved.

Introduction

The vertebrate body plan is established during embryonicdevelopment through the specification of three axes: the ante-rior-posterior, dorsal-ventral, and left–right (LR). In vertebrates,the LR axis is the final axis to be specified just before visceraldevelopment, and its specification is necessary for asymmetricdevelopment of organs, such as the heart, lungs, and stomach(Peeters and Devriendt, 2006; Shiratori and Hamada, 2006). Workover the last 15 years has provided tremendous advances in ourunderstanding of how the LR axis is established, but many stepsalong the way, from the initial breaking of bilateral symmetry tothe development of asymmetric organs, remain elusive (Bisgroveet al., 2003; Raya and Izpisua Belmonte, 2008; Speder et al., 2007;Tabin, 2005; Vandenberg and Levin, 2009, 2010).

In mouse, the first known event required for establishment ofthe LR axis is the leftward flow of extraembryonic fluid in thenode, which is created by the rotation of motile cilia on the nodesurface (Nonaka et al., 2002; Nonaka et al., 1998; Okada et al.,1999). At the molecular level, two proteins are initially expressedsymmetrically on the left and right side of the node: Nodal and itsinhibitor cerberus-like 2 (Cerl2) (Collignon et al., 1996; Lowe

ll rights reserved.

et al., 1996). Once fluid flow is initiated, Nodal shows increasedexpression at the left node periphery while Cerl2 is downregulated on the left side of the node (Marques et al., 2004). AsNodal feeds back to induce its own expression, Cerl2 repression onthe left side of the node would allow for the observed left-sidedincrease in Nodal expression (Kawasumi et al., 2011; Norris et al.,2002; Schweickert et al., 2010). While both Nodal and Cerl2 arelikely regulated in response to flow, the exact mechanisms of flowdetection and the cellular response to flow have been vigorouslydebated. One hypothesis posits that the flow physically enrichesthe left side of the node with a morphogen such as SonicHedgehog (Shh) (Tanaka et al., 2005). Alternatively, anotherhypothesis (the two cilia model) proposes that the flow generatedby motile cilia at the center of the node is detected via mechan-osensory cilia at the node periphery (McGrath et al., 2003).Although the debate is not settled, two recent observations areconsistent with the two cilia model: first, very little flow isneeded to break asymmetry within the node (Shinohara et al.,2012); and second, the putative sensors in the mechanosensorycilia at the node periphery were identified as Pkd1l1 and Pkd2

(Field et al., 2011; McGrath et al., 2003).From the node, asymmetry is established throughout the

embryo. Nodal becomes expressed in the left lateral plate meso-derm (LPM) where it is required for correct asymmetric develop-ment of visceral organs. Evidence suggests that the Nodal liganditself diffuses from the node to the LPM to induce its own

C.E. Larkins et al. / Developmental Biology 367 (2012) 15–2416

expression there (Oki et al., 2007). Therefore, it is important thatthe levels of active Nodal ligand be tightly regulated in the nodeand LPM to achieve expression in only the left LPM.

In addition to Cerl2, Nodal is also inhibited by the TGF-betaproteins Lefty1 and Lefty2, which are expressed at their highestlevels in the midline and LPM of the embryo, respectively, and aretargets of the Nodal ligand (Meno et al., 2001; Saijoh et al., 2000;Yamamoto et al., 2003). Because Nodal can diffuse over longdistances to induce its own expression, Lefty1 acts as a barrier atthe midline preventing Nodal from inducing expression in theright LPM (Meno et al., 1998; Meno et al., 2001). The expression ofLefty2 in the left LPM limits the amount of time Nodal is expressedthere, and this expression pattern of both Lefty1 and Lefty2 createsa situation in which the level of active Nodal ligand must reach aconcentration threshold in the LPM to overcome inhibition by theLefty proteins (Nakamura et al., 2006), thus adding another layerto the tight regulation of active Nodal protein in the embryo.

Several other signaling events help mediate the left-sidedenrichment of Nodal in the node and its subsequent expressionin the left LPM. Wnt signaling, specifically Wnt3a, induces Notchat the node, and Notch signaling is required for Nodal to beenriched in the left node periphery (Krebs et al., 2003; Nakayaet al., 2005; Raya et al., 2003). Shh signaling is needed to induceexpression of Gdf1 at the node periphery, which is required forNodal ligand diffusion to the left LPM (Tanaka et al., 2007; Zhanget al., 2001). The genetic network of LR axis establishmentinvolves many players, some of which are still unknown, buttogether they turn a localized leftward flow of extraembryonicfluid into a robust LR asymmetry in the LPM that can guidevisceral organ development.

Here we describe the left–right axis defects found in theArl13b hennin (hnn) mouse mutant. Arl13bhnn mutants lack the ciliaprotein Arl13b (ADP ribosylation factor-like 13b), a small regula-tory GTPase, which leads to shortened cilia in the node (Casparyet al., 2007). We show here that the Arl13bhnn mutants areheterotaxic following an unusual pattern of Nodal expression: inthe LPM, Nodal initially shows delayed and randomized expres-sion, with subsequent bilateral expression, while in the nodethere is an inability to maintain asymmetric Nodal expression.Interestingly, we found that although Nodal initially showsexpression in the node in a pattern that is similar to wild-type,Cerl2 is not correctly down regulated on the left side of the node,indicating that, in our mutants, the defect in Nodal expressionmay be secondary to a defect in Cerl2 expression. Mouse mutantsthat are unable to generate flow as well as mutants that cannotdetect flow also express Nodal and Cerl2 symmetrically within thenode raising the possibility that one or both of these processes isdisrupted in Arl13bhnn mutants.

Materials and methods

Mouse strains

The strains (Arl13bhnn, IFT172wim, BATgal, Patched-lacZ, andNodallacZ) were genotyped as previously described and bred atleast 10 generations on the C3H background (Collignon et al.,1996; Garcia-Garcia et al., 2005; Goodrich et al., 1997; Huangfuet al., 2003; Maretto et al., 2003).

Phenotypic analysis

Timed matings of mice carrying the Arl13bhnn mutation wereperformed, and embryos were harvested using somite number forstaging. In situ hybridization was as previously described, andembryos were genotyped after analyzing expression patterns.

Antisense probes were generated from the following cDNA plasmids(with sources): Nodal (Elizabeth Robertson), Gdf1 (Nancy Wall), Shh

and Foxa2 (Andy McMahon), NotchI (Janet Rossant), Lefty1 and Lefty2

(Hiroshi Hamada), Pitx2 (Axel Schweickart), Cerl2 (IMAGE clone ID790229), and Dll1 (Radhika Atit).

Embryos homozygous for Arl13bhnn and carrying either theBATgal, Patched-lacZ, or Nodal-lacZ reporter alleles were harvestedat E8.5 and fixed in 4% PFA with 0.2% glutaraldehyde for 15 min(Goodrich et al., 1997; Maretto et al., 2003). Embryos werewashed (0.1 M phosphate buffer, 2 mM MgCl2, 0.01% sodiumdeoxycholate, 0.02% NP-40) and then treated with 1 mg/mLX-gal in 5 mM potassium ferricyanide and 5 mM potassiumferrocyanide overnight at room temperature.

For organ analysis, embryos were harvested at E12.5. Organswere removed from the body cavity to allow imaging of the lungs.Images were taken using a Leica DM6000B upright fluorescencemicroscope and processed using QCapture software.

Results

Arl13bhnn mutants have heterotaxia

Midgestation Arl13bhnn embryos display a variety of pheno-types including randomized heart looping, an indication of LR axisdefects (Caspary et al., 2007; Garcia-Garcia et al., 2005). Tocharacterize the extent of the LR axis defects in Arl13bhnn mutants,we examined the orientation of the heart, lungs, and stomach inE12.5 embryos. Normally, the left lung has a single lobe, whereasthe right lung has four, so we determined the left versus right lungby the presence of lobes. We found that in Arl13bhnn mutants, thelungs most often showed left isomerism (44.4% of embryos) andnormal orientation at a high rate (38.9%), although the right lunghad smaller and fewer lobes (Fig. 1(A)–(F), Table 1, and data notshown). The heart and stomach orientation were almost comple-tely randomized and were not coordinated. For example, theheart could be on the left while the stomach was on the right(Fig. 1(A)–(F), Table 1) indicating that, in addition to the globaldefect in specifying the LR axis, individual organs interpret the LRaxis distinctly, a defect called heterotaxia. This anomaly is seen inseveral other mouse mutants with LR defects, as well as in humanlaterality disorders (Chen et al., 1998; Lin et al., 1999; Lowe et al.,2001; Meno et al., 1998; Peeters and Devriendt, 2006).

Nodal and Pitx2 are misexpressed in the LPM of Arl13bhnn mutants

Asymmetric organ development is guided by the asymmetricexpression of Nodal and its downstream target Pitx2 in the LPM,both of which we assessed through in situ hybridization (Campioneet al., 1999; Lin et al., 1999; Liu et al., 2001; Ryan et al., 1998;Shiratori et al., 2001). In all cases, we scored molecular expressionpatterns prior to genotyping alleviating any ascertainment bias.Wild-type embryos display Nodal expression in the left LPM startingat the 2–3 somite stage, and this expression is extinguished by the6–7 somite stage (Fig. 2(A)–(D)). In Arl13bhnn mutants, we saw aninitial delay in Nodal expression in the LPM until the 4–5 somitestage. Nodal began to be expressed in the LPM at the 4–5 somitestage, but still half of the embryos at this stage did not show Nodal

expression (Fig. 2(H)). Those embryos with expression at the 4–5somite stage most often showed expression in the left LPM orbilaterally (Fig. 2(H)). At the 6–7 somite stage, when most wild-typeembryos have no expression in the LPM (Fig. 2(A)–(D)), almost allArl13bhnn embryos had Nodal expression that was bilateral(Fig. 2(E)–(H)). Therefore, in Arl13bhnn mutants, not only is there aspatial defect in Nodal expression, but there is also a temporaldefect. This expression pattern was in marked contrast to mutants

C.E. Larkins et al. / Developmental Biology 367 (2012) 15–24 17

lacking cilia. Consistent with previous reports, we observed bilateralNodal expression in the LPM in IFT172wim mutants starting at earlysomite stages (Huangfu et al., 2003). Because there is not a delay inNodal expression in mutants lacking cilia, our data indicate thatNodal expression in the LPM is more severely affected in Arl13bhnn

embryos with abnormal cilia than in IFT172wim embryos lacking ciliaaltogether.

In the LPM and splanchnopleure of E8.5 and E9.5 Arl13bhnn

embryos, we found that Pitx2 was most often expressed bilater-ally or exclusively on the left side of the embryo (Fig. 3(A), (B)–(D)). This pattern is consistent with the laterality in the lungs ofArl13bhnn mutants, which most often showed left isomerism or

Fig. 1. Organ laterality in Arl13bhnn mutants shows heterotaxia, while Arl13bhnn;

NodallacZ/þ shows right isomerism. (A)–(C) Wild-type E12.5 embryos. (D)–(F)

E12.5 Arl13bhnn mutant embryos. (G)–(I) E12.5 Arl13bhnn; NodallacZ/þ embryos.

(B) A ventral view of the internal organs of the wild-type embryo shows normal

placement of the heart with its apex towards the left. (C) A dorsal view of the

organs shows the left–right asymmetry of the lungs, where there is a single left

lobe and 4 right lobes. The stomach can be seen on the left side of the embryo. (E)

A ventral view shows the heart apex toward the left side of the Arl13bhnn mutant

embryo. (F) A dorsal view of the same embryo in (D) and (E) shows the stomach on

the right, as well as a single lobe on the left and right lung. (H) A ventral view of

the Arl13bhnn; NodallacZ/þ mutant shows the heart apex toward the left. (I) A dorsal

view shows there are small lobes on both the left and right lung, with the stomach

in the center of the embryo. Inset is a ventral view of the lungs, with arrowheads

pointing to the lobes. rv, right ventricle; lv, left ventricle; st, stomach; l, left;

r, right.

Table 1Organ laterality in wild-type, Arl13bhnn, and Arl13bhnn; NodallacZ mutants. The orientat

embryos. n is the number of embryos examined and the percentage of total embryos e

Heart Lungs

Left (%) Right (%) n¼ Normal (%) Reversed (%)

Wild-type 100 0 6 100 0

Arl13bhnn/hnn 57.9 42.1 19 38.9 6

Arl13bhnn/hnn; NodallacZ/þ 62.5 37.5 8 0 0

normal orientation (Table 1). Arl13bhnn embryos lacking Pitx2

expression were most often found at early somite stages, suggest-ing Pitx2 expression is also delayed in the absence of Arl13b.

The midline barrier in Arl13bhnn mutants is lost

Because we saw a large number of Arl13bhnn embryos withbilateral Nodal expression, we wanted to examine a possible defectin maintaining the midline barrier. In wild-type embryos, Nodal fromthe LPM turns on expression of its inhibitor Lefty1 in the prospectivefloor plate at the 2–3 somite stage, and Lefty1 expression isextinguished by the 6–7 somite stage due to loss of Nodal expressionin the LPM (Meno et al., 1998; Yamamoto et al., 2003). Our previousanalysis of Arl13bhnn shows that by E9.5, mutants lack a floorplate inthe caudal embryo (Caspary et al., 2007); however, we did seeexpression of the floor plate marker Foxa2, as well as Lefty1, inmutant embryos at E8.5 (Fig. 4). Interestingly, Lefty1 was notexpressed in the midline at the 6–7 somite stage of our mutantembryos, although Nodal expression persisted in the LPM (Fig. 4(A)–(F), Fig. 2(G) and (H)). As there is no floorplate at E9.5 in our mutants,these results could indicate a loss of floorplate integrity in Arl13bhnn

mutants. While examining Lefty1 expression, we also observed Lefty2

expression and found it follows the same pattern as Nodal in the LPM(Fig. 4(A) and (B), (D) and (E), n¼13).

Nodal asymmetry is initiated but not maintained in the node

of Arl13bhnn mutants

Nodal and Cerl2 are initially expressed symmetrically in the crowncells of the node and are then asymmetrically expressed as flowcommences at early somite stages (Nonaka et al., 1998; Okada et al.,1999). The expression of Cerl2, an inhibitor of Nodal, is reduced onthe left while being maintained on the right, while Nodal expressionis increased on the left (Marques et al., 2004; Schweickert et al.,2010). This pattern of Cerl2 expression results in an increase in activeNodal protein on the left side of the node, which leads to increasedNodal expression in the left LPM (Kawasumi et al., 2011; Oki et al.,2009). We examined Nodal and Cerl2 expression in the node, andsurprisingly, we saw that almost half of Arl13bhnn embryos hadincreased Nodal expression on the left side of the node at presomitestages, which was nearly identical to wild-type (Fig. 5(A) and (B)).However, that asymmetry was gradually lost in Arl13bhnn mutantembryos at later somite stages (Fig. 5(A) and (B)). Consistent withCerl2 inhibiting Nodal, we saw that Cerl2 expression was not downregulated on the left side of the Arl13bhnn node to the same extent aswild-type from the presomite to 5 somite stage (Fig. 5(D) and (E)),indicating that the gradual loss in Nodal asymmetry could be due toan inability to down regulate Cerl2.

Left biased Nodal asymmetry is not initiated in the node of IFT172wim

mutants

We examined mutants completely lacking cilia, the IFT172wim

mutants, to see if they had similar defects in Nodal enrichment in

ion of the heart, lungs, and stomach were determined for wild-type and mutant

xamined is shown.

Stomach

Left isom (%) Right isom (%) n¼ Left (%) Right (%) Center (%) n¼

0 0 6 100 0 0 6

44.4 11.1 18 52.4 42.9 4.8 21

0 100 5 0 12.5 87.5 8

Fig. 2. Nodal expression is disrupted in Arl13bhnn mutants. (A)–(C), (E)–(G), (I)–(K), (M)–(O) in situ hybridization for Nodal. (A)–(C) Wild-type embryos showing Nodal

expression in the left LPM starting at the 2–3 somite stage (A) and being extinguished at the 6–7 somite stage (C). (E)–(G) Arl13bhnn mutant embryos showing no LPM

expression at the 2–3 somite stage (E), right-sided expression at the 4–5 somite stage (F), and bilateral expression at the 6–7 somite stage (G). (I)–(K) IFT172wim mutants

showing bilateral Nodal expression starting at the 2–3 somite stage. (M)–(O) Arl13bhnn; IFT172wim mutants have bilateral expression of Nodal starting at the 2–3 somite

stage (M). (D), (H), (L), (P) Graphs quantifying the number of embryos, with each Nodal expression pattern broken down by somite stage.

C.E. Larkins et al. / Developmental Biology 367 (2012) 15–2418

the node (Huangfu et al., 2003). If Arl13bhnn mutants completelylacked fluid flow in the node, we might expect them to show asimilar pattern of Nodal expression to mutants lacking cilia. InIFT172wim mutants, almost half of the embryos showed asym-metric Nodal expression in the node, but increased expressionwas seen more often on the right side of the node than in eitherwild-type or Arl13bhnn mutant embryos (Fig. 5(C)). Our findingthat Arl13bhnn mutants showed Nodal expression in the node in apattern that is distinct from mutants with no cilia suggests thatthe Nodal expression defect in the node of Arl13bhnn mutants isnot due to a complete loss of fluid flow.

Other signaling pathways are grossly intact in Arl13bhnn mutants

at e8.5

Several signaling pathways are active in the node and affectNodal expression in the node and LPM. First, a Wnt3a null mutantshows a similar delay in Nodal expression followed by bilateralexpression of Nodal in the LPM. Wnt3a is required for Notchsignaling at the node, which in turn is needed for Nodal

expression (Nakaya et al., 2005). There has also been a linkbetween cilia and Wnt signaling making it a candidate fordisruption in our mutants (Corbit et al., 2007; Germino, 2005;Simons et al., 2005; Lancaster et al., 2011). To examine canonicalWnt signaling activity in our mutant embryos, we incorporatedthe BATgal reporter allele, which expresses the lacZ gene inresponse to Wnt signaling activity (Maretto et al., 2003). Wesaw no difference in b-galactosidase activity between wild-typeand mutant embryos (Fig. 6(A) and (B)). Likewise, we saw nochange in the expression of Dll1, a Wnt3a target, or of Notch1 andits target, Lfng (Fig. 6(C)–(H)).

Another signaling pathway, Shh, induces expression of Gdf1 atthe node periphery, which is required for Nodal diffusion to theLPM (Tanaka et al., 2007; Zhang et al., 2001). Shh signaling is alsoknown to be required for Nodal expression in the LPM (Tsiairis andMcMahon, 2009), and we know there are Shh signaling defects inArl13bhnn mutants at later stages (Caspary et al., 2007). At E8.5, wesaw no change in Shh expression, and using Patched-lacZ tomonitor Shh signaling activity, there was no difference betweenwild-type and mutant embryos (Fig. 6(I)–(L)). Gdf1 expression was

Fig. 3. Pitx2 expression is disrupted in Arl13bhnn mutants. (A)–(C) Pitx2 in situ

hybridization in E8.5 embryos. (A) Wild-type with expression in the left LPM. (B)

Arl13bhnn embryo with bilateral Pitx2 expression. (C) Arl13bhnn; NodallacZ/þ embryo

with no expression in the LPM. (D) Graph quantifies the number of embryos with

each expression pattern of Pitx2 in the LPM.

C.E. Larkins et al. / Developmental Biology 367 (2012) 15–24 19

also normal (Fig. 6(M) and (N)). This is surprising since Gdf1 andPtch1-lacZ are Shh responsive genes and we previously showed Shhsignaling is abnormal in the absence of Arl13b, however, ourprevious work shows that Shh signaling is not completely abol-ished in the Arl13bhnn mutants and we predict that at e8.5 there isenough Shh signaling to allow for Ptch1 and Gdf1 expression. Thus,while we cannot rule out subtle quantitative differences within thepathways, in terms of the molecular events known to be critical inestablishing the LR axis, Wnt, Shh and Notch activities appeargrossly intact in the absence of Arl13b.

Arl13b and Nodal genetically interact

Our analysis shows that Arl13bhnn mutants have an inability tomaintain higher levels of Nodal expression in the left side of the nodeand also have delayed Nodal expression in the LPM. Because Nodal isin a positive feedback loop to enhance its expression in the node andtravels from the node to induce its expression in the LPM (Nakamuraet al., 2006; Norris et al., 2002; Oki et al., 2007), the expressionpattern we see in our mutants could be caused by a reduction inactive Nodal protein in Arl13bhnn mutants. To test this, we geneticallyreduced Nodal in our Arl13bhnn mutants using the null NodallacZ allele(Collignon et al., 1996). If Nodal is reduced in the Arl13bhnn mutant,then a further reduction in Nodal by introducing the null alleleshould cause a more severe Nodal expression phenotype. Indeed, theArl13bhnn; NodallacZ/þ embryos were more severe, as they lackedNodal expression in the LPM at all somite stages according to both

the b-galactosidase assay as well as in situ hybridizations for Nodal

(Fig. 7(D)–(F)). In the node, in situ hybridization showed that Nodal

expression was weakened such that any asymmetry could not berecorded (Fig. 7(F)). Because Nodal acts to enhance its own expres-sion in the node and induces its expression in the LPM, the simplestinterpretation of this genetic interaction is that the levels of activeNodal are reduced in the Arl13bhnn mutant, although we cannot saywhether this effect is direct or indirect. A reduction in active Nodal inthe LPM could be the direct consequence of reduced active Nodal inthe node.

We confirmed the loss of Nodal expression in Arl13bhnn; NodallacZ/þ

by examining Pitx2, and saw that it likewise was not expressed in theLPM (Fig. 3(C) and (D)). These molecular patterns were reflected bythe organ laterality in the Arl13bhnn; NodallacZ/þ mutants, whichshowed right isomerism in the lungs, with the lobes being smallerthan wild-type, as well as a midline stomach (Fig. 1(G)–(I)).

To determine whether there was a defect in Nodal expression atearlier stages in Arl13bhnn mutants, we also examined the Arl13bhnn;NodallacZ/þ at E7.5 and saw that its expression using b-galactosidasedetection was indistinguishable from wild-type (Fig. 7(A) and (D)).Similarly, at E12.5 in Arl13bhnn; NodallacZ/þ mutants, the only Nodalsignaling-dependent defect we saw was in the visceral LR axis, andnot in mesoderm or endoderm derivatives, which are specified byNodal in the pre-gastrulation embryo.

LPM Nodal expression requires cilia in Arl13bhnn mutants

In mutants that lack cilia and therefore have no flow, Nodal

asymmetry in the node is randomized and Nodal in the LPM isbilateral. This predicts that removing cilia in combination withArl13b should rescue Nodal expression in the LPM. To this end, wecrossed Arl13bhnn mutants to IFT172wim and made double homo-zygous mutants that lack Arl13b as well as cilia. The doublemutant embryos showed bilateral expression of Nodal starting atthe 2–3 somite stage, reminiscent of IFT172wim mutants (Fig. 2(I)–(P)) indicating that without cilia, Arl13b mutants can expressNodal at the correct time.

Discussion

Here we characterized the left–right patterning defects in micelacking the small regulatory GTPase, Arl13b. We found thatArl13bhnn embryos are heterotaxic due to misexpression of severalgenes known to be critical in L–R patterning. In the node weshowed that Nodal and Cerl2 are expressed symmetrically in theabsence of Arl13b. As Arl13b is highly enriched in cilia and its lossleads to short cilia in the node, it is easy to imagine that thesymmetric gene expression arises from abnormal nodal flow.However, we cannot rule out the possibility that Arl13bhnn maydisrupt the ability for flow to be detected and this inability leadsto the symmetric expression of Nodal and Cerl2 in the node.

Organ laterality and the dose of Pitx2 in Arl13bhnn mutants

The heterotaxia we observed in Arl13bhnn mutants corre-sponded to the abnormal expression of the molecules that guideorgan laterality. This phenotype could yield important clues tohelp us understand congenital defects in humans. In severaldifferent LR axis-defective mutants and human syndromes, thevisceral organs may have different lateralities, even though asingle global LR axis decision has been specified. The cause of thishas been perplexing, but the suggestion is that individual organsmay require different doses of Pitx2, in terms of either the amountof Pitx2 the developing organ is presented with or the lengthof time the developing organ sees Pitx2, or both (Liu et al., 2001).

Fig. 4. Midline patterning is normal is Arl13bhnn mutants. (A)–(F) Lefty1 and Lefty2 in situ hybridization in wild-type (A) and (B) and Arl13bhnn (D) and (E). Graphs in (C) and

(F) quantify the number of embryos with Lefty1 expression in the midline. (G)–(J) Foxa2 in situ hybridization in wild-type (G) and (H) and Arl13bhnn (I) and (J).

C.E. Larkins et al. / Developmental Biology 367 (2012) 15–2420

The complex pattern of Nodal and Pitx2 expression in Arl13bhnn

mutants supports this idea. Because the lungs follow almostexactly the Pitx2 expression pattern we saw in Arl13bhnn mutants,they may only require a small dose of Pitx2 for LR specification tooccur. However, the heart and stomach obviously require moreprecise amounts of Pitx2, as they developed with a randomorientation in Arl13bhnn mutants and chose a left or right sideindependently of one another.

Patients with mutations in Arl13b display Joubert syndrome,which is characterized by brain malformations (Cantagrel et al.,2008). In the context of our analysis it is interesting to note that

laterality defects in Joubert patients are rare. Since Arl13b is aGTPase and likely has multiple effectors, it will be interesting tosee whether any effectors are specific to LR specification. Thiswould also predict that distinct alleles of Arl13b may be identifiedin non-Joubert patients who only exhibit laterality defects.

Defective Nodal expression in the LPM

In the Arl13bhnn LPM, we have shown that Nodal expression isdelayed and randomized followed by late bilateral expression.This is most likely an indirect consequence of the other molecular

Fig. 5. Arl13bhnn embryos show Nodal and Cerl2 expression defects in the node. (A)–(E) in situ hybridization for Nodal and Cerl2. The graphs show the expression patterns

broken down by somite stage. (A) wild-type embryo showing left-sided asymmetry of Nodal. (B) Arl13bhnn mutant embryo with symmetric Nodal expression in the node.

(C) IFT172wim embryo showing right-sided enrichment of Nodal. (D) wild-type embryo showing Cerl2 has higher levels of expression on the right side of the node. (E)

Arl13bhnn mutant showing slightly higher levels of Cerl2 expression in the left side of the node.

Fig. 6. Wnt and Shh signaling is intact in Arl13bhnn mutants. (A) and (B) whole mount b-galactosidase assay shows no difference in Wnt signaling between wild-type

(A) and Arl13bhnn mutant (B) embryos. (C)–(H) Whole mount in situ hybridization for Notch1 (C) and (D), Dll1 (E) and (F), and Lfng (G) and (H) shows that Notch signaling is

intact in Arl13bhnn mutants compared to wild-type. (I) and (J) Whole mount in situ hybridization for Shh is identical between wild-type and Arl13bhnn mutants. (K) and (L)

Patched-lacZ shows no change in Hedgehog response between wild-type and Arl13bhnn mutant embryos. (M) and (N) the Shh signaling target Gdf1 is expressed normally in

Arl13bhnn mutants.

C.E. Larkins et al. / Developmental Biology 367 (2012) 15–24 21

defects we observed. The late bilateral expression of Nodal in theLPM may result from the late loss of midline integrity in ourmutants. The Nodal inhibitor Lefty1 is normally expressed in the

midline in response to Nodal from the left LPM and preventsNodal from diffusing to the right LPM to induce expression there(Meno et al., 1998; Sakuma et al., 2002; Yamamoto et al., 2003).

Fig. 7. Arl13b interacts genetically with Nodal. (A) and (B) wild-type whole mount

b-galactosidase assay in embryos carrying the nodal-lacZ allele at e7.5 (A) and the

6–7 somite stage (B). (D) and (E) whole mount b-galactosidase assay in Arl13bhnn

mutant embryos carrying the nodal-lacZ allele at e7.5 (D) and the 6–7 somite stage

(E). (C) and (F) Whole mount in situ hybridization confirms the lack of Nodal

expression in Arl13bhnn mutant embryos with the Nodal-lacZ allele (F), while

expression is intact in wild-type embryos (C).

C.E. Larkins et al. / Developmental Biology 367 (2012) 15–2422

We found that Lefty1 was initially expressed in the midline, butwas lost at late somite stages. This loss of expression occurredeven though Nodal was still expressed in the LPM, which wouldlikely allow for Nodal to diffuse to the opposing LPM to induce itsexpression there, thus causing bilateral LPM Nodal expressiononly at late somite stages.

Another interpretation for the late bilateral expression ofNodal in our mutants is that the inability to down regulate Cerl2

and the loss of asymmetric Nodal expression in the node leads toequal levels of active Nodal protein on both sides of the node atlater somite stages causing activation of Nodal expression in boththe left and right LPM. This inability to down regulate Cerl2 couldalso be the cause for delayed expression of Nodal in the LPM, asthere would be less active Nodal protein in the node to induce itsexpression in the LPM. The lack of Nodal in the LPM of Arl13bhnn;NodallacZ/þ embryos supports the model that there is less activeNodal protein in the LPM to activate its expression there. Onepossibility is that Arl13b functions in Nodal signaling from thenode to the LPM, however we cannot determine this definitivelyas the decreased Nodal expression in the node could also causeabnormal expression of Nodal in the LPM.

The fact that mutants lacking cilia can effectively signal fromthe node to the LPM while Arl13bhnn mutants do not is reminis-cent of other L–R patterning mouse mutants possessing cilia. Inthe node, Pkd2 and Pkd1l1 mutants display normal cilia but in theLPM they lack a response to the Nodal signaling cascade (Fieldet al., 2011). Inv mutants have cilia in the node that generate slow

abnormal flow towards the left, however they display delayedNodal expression in the right LPM (Lowe et al., 1996; Okada et al.,1999). For inv mutants the delay may be due to altered levels ofNodal signaling from the node as Cerl2 and Lefty1 expressionthere is abnormal. Such a mechanism may underlie the Arl13bhnn

phenotype although it is unclear whether the direct comparisoncan be made. Analysis of the Pkd1l1, Pkd2 and inv mutantsrevealed that asymmetric gene expression is highly sensitive togenetic background and the inv work was performed on a FVBstrain (Field et al., 2011; Lowe et al., 1996; Meno et al., 1996). Wework in the C3H strain background where the Pkd alleles werealso analyzed so can make direct phenotypic comparisons tothose. Pkd2 and Pkd1l1 mutants have normal cilia within the nodeyet the Nodal signaling cascade is not activated in the LPM ofeither mutant whereas the cascade is activated late in the absenceof Arl13b. The simplest model to reconcile these phenotypeswould posit that in the Pkd mutants, there is not enough activeNodal being produced in the node to allow for Nodal induction inthe LPM, whereas in Arl13bhnn mutants, there is just enough Nodalin the node to allow for activation of Nodal expression, albeitdelayed, in the LPM.

The fact that removing cilia in the context of Arl13b mutantscan rescue the activation of the Nodal signaling cascade suggeststhat cilia remove positive and negative regulators important foractivation of Nodal. For example, we know that removal of ciliacauses loss of Lefty1 expression in the midline (Nakamura et al.,2006), so removal of cilia in the Arl13bhnn mutants would likelyremove the Lefty1 inhibitor allowing for higher levels of activeNodal in the LPM (Nakamura et al., 2006). It would be interestingto delete Lefty1 in the Arl13bhnn mutants to determine if we seethe same phenotype that we found from removing cilia.

It is also possible that Arl13b has roles in the LPM independentof its function in the node. A precise understanding would requireus to spatially delete Arl13b in the LPM which has proventechnically challenging as Arl13b protein requires 42 h to turnover (our unpublished data). Regardless, the idea that Arl13b

affects the balance of positive and negative inputs to Nodalsignaling parallels its function in Shh signaling where we showedit controls activation but not repression of Shh signaling (Casparyet al., 2007).

Defective Nodal expression in the node

The primary LR axis specification defect in Arl13bhnn mutants isthe symmetric expression of Nodal and Cerl2 in the node as this canexplain all subsequent observations. Both Cerl2 and Nodal arepredicted to be downstream targets of fluid flow (Hirokawa et al.,2006; Schweickert et al., 2010), and our mutants have stunted ciliathat, if motile, would likely be unable to generate the same quality ofleftward flow that would be found in wild-type embryos. It wasrecently shown that very little fluid flow is required in the node togenerate asymmetric expression of Cerl2, and that in the absence offlow, Cerl2 is symmetric (Shinohara et al., 2012). We saw that almost50% of Arl13bhnn mutant embryos had higher levels of Nodal expres-sion in the left side of the node at presomite stages, a pattern thatwas nearly identical to wild-type. However, this pattern graduallyshifted to symmetric expression between the 2 to 5 somite stage.This may indicate that our mutants generate a fluid flow that is tooweak to allow for the signaling events required for down regulationof Cerl2, but this weak flow is sufficient for initial Nodal enrichment.Consistent with this, we found that the inhibitor of Nodal, Cerl2, wasnot down regulated normally in the left side of the node, and mostoften showed equal levels of expression in both sides of the node. Aninability to down regulate Cerl2 could lead to less active Nodal

protein in the node, preventing the Nodal feedback loop necessary forincreasing Nodal expression there (Norris et al., 2002). To confirm

C.E. Larkins et al. / Developmental Biology 367 (2012) 15–24 23

that reduced levels of Nodal expression are due to the defect in Cerl2

expression, it will be necessary to generate Cerl2� /�; Arl13bhnn

double mutants and observe Nodal expression.The symmetric Nodal and Cerl2 we observed in the Arl13bhnn

node could also be explained if Arl13b is required for detection offlow in the node. The two cilia hypothesis invokes a Pkd2-dependent mechanism for flow detection. Mutations in Pkd2 orits binding partner, Pkd1l1 have normal cilia but LR defectsstemming from symmetric expression of both Nodal and Cerl2 inthe 3 somite node. As this mimics the Arl13bhnn phenotype, itraises the possibility that Arl13bhnn mutants may also be unable todetect flow. Arl13b is a member of the Arf family whose membershave well-established roles in protein transport including thetransport of Pkd2 to the cilium by Arf4 (Ward et al., 2011).

Furthermore, both Arl13b and Pkd mutants possess cilia yetNodal is delayed or absent in the LPM. This has always beendifficult to reconcile with the phenotype of mutants lacking ciliawhich express Nodal in the LPM, albeit bilaterally. One model toexplain this would posit that cilia are required for both positiveand negative regulators that induce Nodal in the LPM and thatremoving cilia ablates both, resulting in bilateral Nodal expressionby the 2–3 somite stage. In contrast, in Pkd or Arl13b mutantswhere cilia are present (albeit distinctly) perhaps the balance ofthe positive and negative regulation is distorted resulting indistinct spatial and temporal patterns of Nodal expression in theLPM. The bilateral Nodal we saw in the Arl13bhnn IFT172wim doublemutants is consistent with this interpretation.

Taken together our analysis shows that Arl13b controls left–right patterning initially by disrupting the normal asymmetricexpression of Nodal and Cerl2 in the node. It remains possible thatthis is due to distinct roles of Arl13b in both the generation anddetection of flow. In support of this, mutants lacking cilia do notdisplay presomitic node asymmetry of Nodal as Arl13b mutantsdo. Mutants that cannot detect flow cannot activate the Nodalsignaling cascade in the LPM as Arl13b mutants do (Field et al.,2011). We have recently been able to bypass the Arl13b depen-dent ciliogenesis defects in a cell culture model indicating thatArl13b functions via distinct effectors (Mariani and Caspary,personal communication). As we identify specific Arl13b effectors,it will be interesting to determine if they distinguish roles ofArl13b in generating and detecting flow in the node.

Acknowledgments

We thank Miao Sun, Nicole Umberger, and Karolina Pio-trowska-Nitsche for helpful comments on the manuscript. Thiswork was funded by a Predoctoral Fellowship from the GreaterSoutheast Region American Heart Association (C.E.L.), a BasilO’Conner Starter Scholar Award from the March of Dimes (T.C.),as well as a Hitchings Elion Career Development Award from theBurroughs Wellcome Fund (T.C.).

References

Bisgrove, B.W., Morelli, S.H., Yost, H.J., 2003. Genetics of human lateralitydisorders: insights from vertebrate model systems. Annu. Rev. GenomicsHum. Genet. 4, 1–32.

Campione, M., Steinbeisser, H., Schweickert, A., Deissler, K., van Bebber, F., Lowe,L.A., Nowotschin, S., Viebahn, C., Haffter, P., Kuehn, M.R., Blum, M., 1999. Thehomeobox gene Pitx2: mediator of asymmetric left–right signaling in verte-brate heart and gut looping. Development 126, 1225–1234.

Cantagrel, V., Silhavy, J.L., Bielas, S.L., Swistun, D., Marsh, S.E., Bertrand, J.Y.,Audollent, S., Attie-Bitach, T., Holden, K.R., Dobyns, W.B., Traver, D., Al-Gazali,L., Ali, B.R., Lindner, T.H., Caspary, T., Otto, E.A., Hildebrandt, F., Glass, I.A.,Logan, C.V., Johnson, C.A., Bennett, C., Brancati, F., Valente, E.M., Woods, C.G.,Gleeson, J.G., 2008. Mutations in the cilia gene ARL13b lead to the classicalform of Joubert syndrome. Am. J. Hum. Genet. 83, 170–179.

Caspary, T., Larkins, C.E., Anderson, K.V., 2007. The graded response to SonicHedgehog depends on cilia architecture. Dev. Cell 12, 767–778.

Chen, J., Knowles, H.J., Hebert, J.L., Hackett, B.P., 1998. Mutation of the mousehepatocyte nuclear factor/forkhead homologue 4 gene results in an absence ofcilia and random left–right asymmetry. J. Clin. Invest. 102, 1077–1082.

Collignon, J., Varlet, I., Robertson, E.J., 1996. Relationship between asymmetricNodal expression and the direction of embryonic turning. Nature 381,155–158.

Corbit, K.C., Shyer, A.E., Dowdle, W.E., Gaulden, J., Singla, V., Reiter, J.F., 2007. Kif3aconstrains beta-catenin-dependent Wnt signalling through dual ciliary andnon-ciliary mechanisms. Nat. Cell Biol. 0, 0.

Field, S., Riley, K.L., Grimes, D.T., Hilton, H., Simon, M., Powles-Glover, N., Siggers,P., Bogani, D., Greenfield, A., Norris, D.P., 2011. Pkd1l1 establishes left–rightasymmetry and physically interacts with Pkd2. Development 138, 1131–1142.

Garcia-Garcia, M.J., Eggenschwiler, J.T., Caspary, T., Alcorn, H.L., Wyler, M.R.,Huangfu, D., Rakeman, A.S., Lee, J.D., Feinberg, E.H., Timmer, J.R., Anderson,K.V., 2005. Analysis of mouse embryonic patterning and morphogenesis byforward genetics. Proc. Nat. Acad. Sci. U. S. A. 102, 5913–5919.

Germino, G.G., 2005. Linking cilia to Wnts. Nat. Genet 37, 455–457.Goodrich, L.V., Milenkovic, L., Higgins, K.M., Scott, M.P., 1997. Altered neural cell

fates and medulloblastoma in mouse patched mutants. Science 277,1109–1113.

Hirokawa, N., Tanaka, Y., Okada, Y., Takeda, S., 2006. Nodal flow and the generationof left–right asymmetry. Cell 125, 33–45.

Huangfu, D., Liu, A., Rakeman, A.S., Murcia, N.S., Niswander, L., Anderson, K.V.,2003. Hedgehog signalling in the mouse requires intraflagellar transportproteins. Nature 426, 83–87.

Kawasumi, A., Nakamura, T., Iwai, N., Yashiro, K., Saijoh, Y., Belo, J.A., Shiratori, H.,Hamada, H., 2011. Left–right asymmetry in the level of active Nodal proteinproduced in the node is translated into left–right asymmetry in the lateralplate of mouse embryos. Dev. Biol. 353, 321–330.

Krebs, L.T., Iwai, N., Nonaka, S., Welsh, I.C., Lan, Y., Jiang, R., Saijoh, Y., O’Brien, T.P.,Hamada, H., Gridley, T., 2003. Notch signaling regulates left–right asymmetrydetermination by inducing Nodal expression. Genes Dev. 17, 1207–1212.

Lancaster, M.A., Schroth, J., Gleeson, J.G., 2011. Subcellular spatial regulation ofcanonical Wnt signalling at the primary cilium. Nat. Cell Biol. 13, 700–707.

Lin, C.R., Kioussi, C., O’Connell, S., Briata, P., Szeto, D., Liu, F., Izpisua-Belmonte, J.C.,Rosenfeld, M.G., 1999. Pitx2 regulates lung asymmetry, cardiac positioning andpituitary and tooth morphogenesis. Nature 401, 279–282.

Liu, C., Liu, W., Lu, M.F., Brown, N.A., Martin, J.F., 2001. Regulation of left–rightasymmetry by thresholds of Pitx2c activity. Development 128, 2039–2048.

Lowe, L.A., Supp, D.M., Sampath, K., Yokoyama, T., Wright, C.V., Potter, S.S.,Overbeek, P., Kuehn, M.R., 1996. Conserved left–right asymmetry of Nodalexpression and alterations in murine situs inversus. Nature 381, 158–161.

Lowe, L.A., Yamada, S., Kuehn, M.R., 2001. Genetic dissection of Nodal function inpatterning the mouse embryo. Development 128, 1831–1843.

Maretto, S., Cordenonsi, M., Dupont, S., Braghetta, P., Broccoli, V., Hassan, A.B.,Volpin, D., Bressan, G.M., Piccolo, S., 2003. Mapping Wnt/beta-catenin signal-ing during mouse development and in colorectal tumors. Proc. Nat. Acad. Sci.U.S.A. 100, 3299–3304.

Marques, S., Borges, A.C., Silva, A.C., Freitas, S., Cordenonsi, M., Belo, J.A., 2004. Theactivity of the Nodal antagonist Cerl-2 in the mouse node is required for correctL/R body axis. Genes Dev. 18, 2342–2347.

McGrath, J., Somlo, S., Makova, S., Tian, X., Brueckner, M., 2003. Two populations ofnode monocilia initiate left–right asymmetry in the mouse. Cell 114, 61–73.

Meno, C., Saijoh, Y., Fujii, H., Ikeda, M., Yokoyama, T., Yokoyama, M., Toyoda, Y.,Hamada, H., 1996. Left–right asymmetric expression of the TGF beta-familymember lefty in mouse embryos. Nature 381, 151–155.

Meno, C., Shimono, A., Saijoh, Y., Yashiro, K., Mochida, K., Ohishi, S., Noji, S.,Kondoh, H., Hamada, H., 1998. Lefty-1 is required for left–right determinationas a regulator of lefty-2 and Nodal. Cell 94, 287–297.

Meno, C., Takeuchi, J., Sakuma, R., Koshiba-Takeuchi, K., Ohishi, S., Saijoh, Y., Miyazaki,J., ten Dijke, P., Ogura, T., Hamada, H., 2001. Diffusion of Nodal signaling activity inthe absence of the feedback inhibitor Lefty2. Dev. Cell 1, 127–138.

Nakamura, T., Mine, N., Nakaguchi, E., Mochizuki, A., Yamamoto, M., Yashiro, K.,Meno, C., Hamada, H., 2006. Generation of robust left–right asymmetry in themouse embryo requires a self-enhancement and lateral-inhibition system.Dev. Cell 11, 495–504.

Nakaya, M.A., Biris, K., Tsukiyama, T., Jaime, S., Rawls, J.A., Yamaguchi, T.P., 2005.Wnt3a links left–right determination with segmentation and anteroposterioraxis elongation. Development 132, 5425–5436.

Nonaka, S., Shiratori, H., Saijoh, Y., Hamada, H., 2002. Determination of left–rightpatterning of the mouse embryo by artificial Nodal flow. Nature 418, 96–99.

Nonaka, S., Tanaka, Y., Okada, Y., Takeda, S., Harada, A., Kanai, Y., Kido, M.,Hirokawa, N., 1998. Randomization of left–right asymmetry due to loss ofNodal cilia generating leftward flow of extraembryonic fluid in mice lackingKIF3B motor protein. Cell 95, 829–837.

Norris, D.P., Brennan, J., Bikoff, E.K., Robertson, E.J., 2002. The Foxh1-dependentautoregulatory enhancer controls the level of Nodal signals in the mouseembryo. Development 129, 3455–3468.

Okada, Y., Nonaka, S., Tanaka, Y., Saijoh, Y., Hamada, H., Hirokawa, N., 1999. AbnormalNodal flow precedes situs inversus in iv and inv mice. Mol. Cell 4, 459–468.

Oki, S., Hashimoto, R., Okui, Y., Shen, M.M., Mekada, E., Otani, H., Saijoh, Y.,Hamada, H., 2007. Sulfated glycosaminoglycans are necessary for Nodal signaltransmission from the node to the left lateral plate in the mouse embryo.Development 134, 3893–3904.

C.E. Larkins et al. / Developmental Biology 367 (2012) 15–2424

Oki, S., Kitajima, K., Marques, S., Belo, J.A., Yokoyama, T., Hamada, H., Meno, C.,2009. Reversal of left–right asymmetry induced by aberrant Nodal signaling inthe node of mouse embryos. Development 136, 3917–3925.

Peeters, H., Devriendt, K., 2006. Human laterality disorders. Eur. J. Med. Genet 49,349–362.

Raya, A., Izpisua Belmonte, J.C., 2008. Insights into the establishment of left–rightasymmetries in vertebrates. Birth Defects Res. C Embryo Today 84, 81–94.

Raya, A., Kawakami, Y., Rodriguez-Esteban, C., Buscher, D., Koth, C.M., Itoh, T.,Morita, M., Raya, R.M., Dubova, I., Bessa, J.G., de la Pompa, J.L., Belmonte, J.C.,2003. Notch activity induces Nodal expression and mediates the establishmentof left–right asymmetry in vertebrate embryos. Genes Dev. 17, 1213–1218.

Ryan, A.K., Blumberg, B., Rodriguez-Esteban, C., Yonei-Tamura, S., Tamura, K.,Tsukui, T., de la Pena, J., Sabbagh, W., Greenwald, J., Choe, S., Norris, D.P.,Robertson, E.J., Evans, R.M., Rosenfeld, M.G., Izpisua Belmonte, J.C., 1998. Pitx2determines left–right asymmetry of internal organs in vertebrates. Nature394, 545–551.

Saijoh, Y., Adachi, H., Sakuma, R., Yeo, C.Y., Yashiro, K., Watanabe, M., Hashiguchi,H., Mochida, K., Ohishi, S., Kawabata, M., Miyazono, K., Whitman, M., Hamada,H., 2000. Left–right asymmetric expression of lefty2 and Nodal is induced by asignaling pathway that includes the transcription factor FAST2. Mol. Cell 5,35–47.

Sakuma, R., Ohnishi, Yi, Y., Meno, C., Fujii, H., Juan, H., Takeuchi, J., Ogura, T., Li, E.,Miyazono, K., Hamada, H., 2002. Inhibition of Nodal signalling by Leftymediated through interaction with common receptors and efficient diffusion.Genes Cells 7, 401–412.

Schweickert, A., Vick, P., Getwan, M., Weber, T., Schneider, I., Eberhardt, M., Beyer,T., Pachur, A., Blum, M., 2010. The Nodal inhibitor coco is a critical target ofleftward flow in Xenopus. Curr. Biol. 20, 738–743.

Shinohara, K., Kawasumi, A., Takamatsu, A., Yoshiba, S., Botilde, Y., Motoyama, N.,Reith, W., Durand, B., Shiratori, H., Hamada, H., 2012. Two rotating cilia in thenode cavity are sufficient to break left–right symmetry in the mouse embryo.Nat. Commun. 3, 622.

Shiratori, H., Hamada, H., 2006. The left–right axis in the mouse: from origin tomorphology. Development 133, 2095–2104.

Shiratori, H., Sakuma, R., Watanabe, M., Hashiguchi, H., Mochida, K., Sakai, Y.,Nishino, J., Saijoh, Y., Whitman, M., Hamada, H., 2001. Two-step regulation of

left–right asymmetric expression of Pitx2: initiation by Nodal signaling andmaintenance by Nkx2. Mol. Cell 7, 137–149.

Simons, M., Gloy, J., Ganner, A., Bullerkotte, A., Bashkurov, M., Kronig, C., Schermer,

B., Benzing, T., Cabello, O.A., Jenny, A., Mlodzik, M., Polok, B., Driever, W.,Obara, T., Walz, G., 2005. Inversin, the gene product mutated in nephro-

nophthisis type II, functions as a molecular switch between Wnt signalingpathways. Nat. Genet 37, 537–543.

Speder, P., Petzoldt, A., Suzanne, M., Noselli, S., 2007. Strategies to establish left/right asymmetry in vertebrates and invertebrates. Curr. Opin. Genet Dev. 17,351–358.

Tabin, C., 2005. Do we know anything about how left–right asymmetry is firstestablished in the vertebrate embryo? J. Mol. Histol. 36, 317–323.

Tanaka, C., Sakuma, R., Nakamura, T., Hamada, H., Saijoh, Y., 2007. Long-rangeaction of Nodal requires interaction with GDF1. Genes Dev. 21, 3272–3282.

Tanaka, Y., Okada, Y., Hirokawa, N., 2005. FGF-induced vesicular release of Sonichedgehog and retinoic acid in leftward Nodal flow is critical for left–right

determination. Nature 435, 172–177.Tsiairis, C.D., McMahon, A.P., 2009. An Hh-dependent pathway in lateral plate

mesoderm enables the generation of left/right asymmetry. Curr. Biol. 19,

1912–1917.Vandenberg, L.N., Levin, M., 2009. Perspectives and open problems in the early

phases of left–right patterning. Semin. Cell Dev. Biol. 20, 456–463.Vandenberg, L.N., Levin, M., 2010. Far from solved: a perspective on what we know

about early mechanisms of left–right asymmetry. Dev. Dyn. 239, 3131–3146.Ward, H.H., Brown-Glaberman, U., Wang, J., Morita, Y., Alper, S.L., Bedrick, E.J.,

Gattone 2nd, V.H., Deretic, D., Wandinger-Ness, A., 2011. A conserved signaland GTPase complex are required for the ciliary transport of polycystin-1. Mol.Biol. Cell 22, 3289–3305.

Yamamoto, M., Mine, N., Mochida, K., Sakai, Y., Saijoh, Y., Meno, C., Hamada, H.,2003. Nodal signaling induces the midline barrier by activating Nodal expres-

sion in the lateral plate. Development 130, 1795–1804.Zhang, X.M., Ramalho-Santos, M., McMahon, A.P., 2001. Smoothened mutants

reveal redundant roles for Shh and Ihh signaling including regulation of L/Rasymmetry by the mouse node. Cell 105, 781–792.