β/PKA and Gli1 regulate the maintenance of neural ... with which cells leave the IZ to populate the differentiating ... Furthermore, we identify the transcription factor Gli1, expressed

3137

INTRODUCTIONNeurogenesis in the embryonic vertebrate neural tube is spatio-temporally controlled to coordinate neuronal production andprogenitor cell maintenance. Neural progenitors are crucial for lateevents of brain maturation, but the mechanisms underlying theirmaintenance, recruitment and fate are only partially understood(Bally-Cuif and Hammerschmidt, 2003; Kageyama et al., 2005;Panchision and McKay, 2002). To approach these mechanisms, weare focusing on an evolutionarily conserved progenitor pool, the‘intervening zone’ (IZ), which adjoins the midbrain-hindbrainboundary (MHB) in the neural tube of all vertebrate embryos (Bally-Cuif et al., 1993; Geling et al., 2003; Stigloher et al., 2008; Vaage,1969) (Fig. 1A-C). Lineage tracing of the IZ pool in the zebrafishembryo demonstrates that it progressively contributes neurons to theentire midbrain-hindbrain domain (MH) (Tallafuss and Bally-Cuif,2003), a territory involved in a number of physiological andintegrative functions, including sensory processing, motor controland social behavior.

The multiple functional outputs of the MH require a diversity ofneuronal subtypes organized in a precise neuroanatomical pattern.Two properties of the IZ demonstrate its importance in organizingMH maturation. Along the anteroposterior (AP) axis, the temporalorder with which cells leave the IZ to populate the differentiatingMH correlates with the future spatial organization and subtype of the

neurons that they generate (Tallafuss and Bally-Cuif, 2003). Hence,the mechanisms controlling maintenance of the IZ along the AP axisare likely to influence the generation of MH neuronal subtypes.Along the dorsoventral (DV, initially mediolateral) axis, IZ cellsexhibit differences in their propensity to undergo neurogenesis,which correlate with the earlier maturation of the basal plate ascompared with the alar plate (Easter et al., 1994): at earlyneurogenesis stages, medial (future ventral) IZ (MIZ) cells are moreprone to undergo neurogenesis than lateral (future dorsal) IZ (LIZ)cells (Ninkovic et al., 2005).

The maintenance of IZ progenitors is controlled by E(Spl) bHLHtranscription factors, such as mouse Hes1 and Hes3 or zebrafishHer5 and Her11, which inhibit expression of proneural genes, suchas neurogenin 1 (neurog1) (Hatakeyama et al., 2004; Kageyama etal., 2005; Stigloher et al., 2008). The compound genetic ablation ofmouse Hes1 and Hes3 leads to the premature differentiation of theIZ and loss of MH neuronal identities (Hatakeyama et al., 2004;Hirata et al., 2001). her5 and her11 are expressed across the IZ andplay redundant roles in IZ maintenance (Ninkovic et al., 2005).Specifically, the IZ is sensitive to a total level of Her5+Her11, withdifferences along the mediolateral axis: three copies of her5 and/orher11 are sufficient to maintain the IZ, two copies are enough tomaintain the LIZ but not the MIZ (which transforms into a neurog1-positive zone) (Fig. 1B), and a lower amount leads to ectopicexpression of neurog1 across the LIZ as well (Ninkovic et al., 2005)(Fig. 1C). Hence, E(Spl) activity controls IZ maintenance and issensed differently by MIZ and LIZ cells.

Unlike classical E(Spl) genes, her5 and her11 are not activated byNotch signaling (Bae et al., 2005; Geling et al., 2004; Hans et al.,2004; Hatakeyama and Kageyama, 2006; Stigloher et al., 2008), andthe molecular cascades involving E(Spl) activity during IZformation remain unknown. Similarly, the mechanisms renderingMIZ cells less sensitive to E(Spl) activity than LIZ cells need to bediscovered. To reveal these mechanisms, we used sensitized

Gsk3β/PKA and Gli1 regulate the maintenance of neuralprogenitors at the midbrain-hindbrain boundary in concertwith E(Spl) factor activityJovica Ninkovic*,†, Christian Stigloher*, Christina Lillesaar* and Laure Bally-Cuif‡

Neuronal production in the midbrain-hindbrain domain (MH) of the vertebrate embryonic neural tube depends on a progenitorpool called the ‘intervening zone’ (IZ), located at the midbrain-hindbrain boundary. The progressive recruitment of IZ progenitorsalong the mediolateral (future dorsoventral) axis prefigures the earlier maturation of the MH basal plate. It also correlates with alower sensitivity of medial versus lateral IZ progenitors to the neurogenesis inhibition process that maintains the IZ pool. This role isperformed in zebrafish by the E(Spl) factors Her5 and Her11, but the molecular cascades cooperating with Her5/11, and thoseaccounting for their reduced effect in the medial IZ, remain unknown. We demonstrate here that the kinases Gsk3β and cAMP-dependent protein kinase A (PKA) are novel determinants of IZ formation and cooperate with E(Spl) activity in a dose-dependentmanner. Similar to E(Spl), we show that the activity of Gsk3β/PKA is sensed differently by medial versus lateral IZ progenitors.Furthermore, we identify the transcription factor Gli1, expressed in medial IZ cells, as an antagonist of E(Spl) and Gsk3β/PKA, anddemonstrate that the neurogenesis-promoting activity of Gli1 accounts for the reduced sensitivity of medial IZ progenitors toneurogenesis inhibitors and their increased propensity to differentiate. We also show that the expression and activity of Gli1 in thisprocess are, surprisingly, independent of Hedgehog signaling. Together, our results suggest a model in which the modulation ofE(Spl) and Gsk3β/PKA activities by Gli1 underlies the dynamic properties of IZ maintenance and recruitment.

KEY WORDS: E(Spl), Her5, Her11, Zebrafish, Midbrain-hindbrain boundary, Neural progenitor, Gsk3β, PKA, Gli1

Development 135, 3137-3148 (2008) doi:10.1242/dev.020479

Helmholtz Zentrum Muenchen, German Research Center for Environmental Health,Department of Zebrafish Neurogenetics, Institute of Developmental Genetics,Ingolstaedter Landstrasse 1, D-85764 Neuherberg, Germany.

*These authors contributed equally to this work†Present address: Helmholtz Zentrum Muenchen, German Research Center forEnvironmental Health, Institute of Stem Cell Research, Ingolstaedter Landstrasse 1,D-85764 Neuherberg, Germany‡Author for correspondence (e-mail: [email protected])

Accepted 16 July 2008 DEVELO

PMENT

3138

conditions in which the total level of E(Spl) activity is reduced, andwe tested the influence of several signaling cascades and signaltransduction pathways active in the anterior neural plate. We showthat the kinases Gsk3β and PKA are novel determinants of IZformation in zebrafish, and demonstrate that Gsk3β/PKA cooperatewith E(Spl) activity in a dose-dependent manner throughout the IZ.Similar to E(Spl) factors, Gsk3β and PKA activities are senseddifferentially along the mediolateral axis. We demonstrate that thetranscription factor Gli1, expressed within the MIZ and displayingneurogenic activity, accounts for this differential response.Surprisingly, we also show that this activity of Gli1 is independentof Hedgehog (Hh) signaling. Our results provide a molecularframework to help understand IZ maintenance and its propertiesalong the AP and DV axes.

MATERIALS AND METHODSZebrafish strains and transgenic linesEmbryos from wild-type (AB) or transgenic [hsp70:tcf3-GFPw26 (Lewis etal., 2004) and Tg(her5PAC:EGFP)ne1939 (Tallafuss and Bally-Cuif, 2003)]fish were staged according to Kimmel et al. (Kimmel et al., 1995).smoothened (smub641) (Varga et al., 2001) and Dfw5/w5 (Lekven et al., 2003)mutants were obtained by pairwise mating of heterozygous adult carriers.

In situ hybridization and immunocytochemistryIn situ hybridization experiments were performed as described(Hammerschmidt et al., 1996b; Ninkovic et al., 2005) with the followingprobes: her5 (Müller et al., 1996), neurog1 (Korzh et al., 1998), pax2.1 (pax2a– ZFIN) (Lun and Brand, 1998), gli1 [recloned using the primersrecommended by Thisse and Thisse (Thisse and Thisse, 2005) for proben°eu934], gli2 (Karlstrom et al., 2003), gli3 (Tyurina et al., 2005), wnt1(Molven et al., 1991), myoD (myod1 – ZFIN) (Weinberg et al., 1996), shh(Krauss et al., 1993) and EGFP (Clontech). Primary antibodies forimmunohistochemistry were rabbit anti-GFP (AMS Biotechnology Europe,TP401, 1/500), mouse anti-human neural protein HUC/HUD (ELAVL3/4 –HUGO) (MoBiTec, A-21271, 1/300), rat anti-BrdU (Abcam, 1/200) and rabbitanti-phospho-histone H3 (Upstate Biotechnology, 1/200), revealed usingFITC-, AF488- or Cy3-conjugated secondary antibodies (Invitrogen). BrdUimmunohistochemistry involved pretreatment with 3.3 M HCl for 30 minutesat room temperature, followed with two 15-minute washings in sodiumtetraborate buffer (0.1 M, pH 8.5). Embryos were scored and photographedunder a Zeiss Axioplan photomicroscope.

RNA, morpholino and gripNA injectionsCapped RNAs were synthesized using the Ambion mMessage mMachineKit following the supplier’s instructions and were injected at the one-cellstage: a constitutively active form of protein kinase A, PKA*(Hammerschmidt et al., 1996a) at 20 ng/μl; a dominant-negative form of thePKA regulatory subunit, dnReg (Strähle et al., 1997) at 50 ng/μl; andtruncated Patched (patchedΔloop2) (Briscoe et al., 2001) at 50-150 ng/μl.Morpholino antisense oligonucleotides (gli1MO) (Gene-Tools, OR) andgripNA antisense oligonucleotides (her11GripNA and her5GripNA) (ActiveMotif, Belgium) were injected into one-cell stage embryos at 0.5 mM.Sequences of antisense oligonucleotides are: her5GripNA, 5�-GGTTCGCTCATTTTGTG-3�; her11GripNAATG, 5�-ATTCGG TGTG -CTCTTCAT-3� (Ninkovic et al., 2005); and gli1MO, 5�-CCGACA -CACCCGCTACACCCACAGT-3� (Karlstrom et al., 2003).

Pharmacological treatmentsGlycogen synthase kinase 3 β (Gsk3β) inhibition was achieved by applyingLiCl or Gsk3β inhibitor III (2,4-dibenzyl-5-oxothiadiazolidine-3-thione,OTDZT) (Calbiochem, Germany). Lithium (0.3 M LiCl) treatments wereperformed at 28°C in embryo medium for 15 minutes, followed by threewashes in embryo medium and further development. OTDZT was diluted to1 mM in embryo medium and applied as with LiCl. Cyclopamine(BIOMOL, Germany) was applied at 100 μM to dechorionated embryosfrom the stages indicated until fixation for in situ hybridization. Controlswere soaked in carrier only.

Live confocal imaging of MIZ and LIZ precursor cellsTg(her5PAC:EGFP)ne1939 transgenic zebrafish embryos (Tallafuss andBally-Cuif, 2003) were injected with 50 ng/μl 2xlckmRFP capped RNA(Megason and Fraser, 2003) in two central cells at the eight-cell stage.Developing embryos were imaged live using an inverted laser-scanningconfocal microscope (LSM510Meta, Zeiss) as described (Distel and Köster,2007; Köster and Fraser, 2004). Development was followed at 28°C usinga 20� 0.5 Plan-Apochromat objective from shield until tailbud stage, whendistinct expression of eGFP was visible. Three-dimensional stacks werecollected every 5 minutes using excitation at 561 nm to record 2xlckmRFPprotein. The movies were stopped when the embryos reached 90% epibolyand the final six scans were continued manually, including excitation at 488nm to record GFP protein and locate the IZ. Subsequent image processingwas performed using LSM software (version 3.2 SP1.1, Zeiss) andPhotoshop (version 9.0, Adobe Systems). Tracking of cells was performedusing the manual tracking function in ImageJ (version 1.37a, WayneRasband, National Institutes of Health USA, http://rsb.info.nih.gov/ij/) withthe plug-in from Fabrice Cordeli (Institut Curie, Orsay, France).

BrdU labelingWild-type embryos were dechorionated and soaked in 10 mM BrdU, 15%DMSO in embryo medium for 10 minutes on ice. Embryos were thenwashed three times on ice for 10 minutes each with washing buffercontaining 15% DMSO, fixed immediately and processed for BrdUimmunohistochemistry.

RESULTSLiCl, but not canonical Wnt signaling, modulatesIZ neurogenesisSignaling pathways active along the IZ at the onset of neurogenesismight influence the neurogenic potential of MIZ and LIZ cells. Atlate gastrulation, expression of wnt1, wnt3a (wnt3l – ZFIN) and

RESEARCH ARTICLE Development 135 (18)

Fig. 1. Location and properties of the ‘intervening zone’ (IZ)progenitor pool. (A) Scheme of the zebrafish anterior neural plate atthe 3- to 5-somite stage. Dorsal view, anterior up. The IZ area (identicalto the expression domain of her5 and her11) is outlined in red and iscomposed of medial and lateral cell populations (MIZ and LIZ,respectively). The IZ separates early neurog1-positive proneural clusters:r2l, presumptive lateral neurons of rhombomere 2; r2MN, presumptivemotoneurons of rhombomere 2; vcc, ventro-caudal cluster.(B) Phenotype observed at the 3- to 5-somite stage (orientation as in A)when two doses of ‘Her5+Her11’ are lost (e.g. upon injection of her5or her11 antisense oligonucleotides): the MIZ is lost and replaced by aneurog1-positive proneural cluster (arrows), whereas the LIZ is intact.(C) Phenotype when four doses of ‘Her5+Her11’ are lost (e.g. uponinjection of her5 and her11 antisense oligonucleotides): neurog1expression also becomes induced within the LIZ (arrows). Adapted withpermission from Ninkovic et al (Ninkovic et al., 2005) and Tallafuss andBally-Cuif (Tallafuss and Bally-Cuif, 2003). D

EVELO

PMENT

wnt10b are confined to the IZ area, at high levels within the LIZ andlow or undetectable levels medially (Buckles et al., 2004; Lekven etal., 2003). Because Wnt signaling has been implicated in controllingneurogenesis in other instances (e.g. Megason and McMahon, 2002;Panhuysen et al., 2004; Zechner et al., 2003), in particular bysupporting the proliferating state, we tested whether it influencedMIZ and LIZ cells. We expected that Wnt would favor theprogenitor state and that higher Wnt signaling laterally wouldaccount for the higher sensitivity of LIZ cells to the inhibitionprovided by Her5/Her11.

To address this issue, we first tested whether the Wnt signalingactivator LiCl (Hedgepeth et al., 1997; Klein and Melton, 1996)affected IZ formation. Wild-type embryos were incubated in 0.3 MLiCl for 15 minutes at late gastrulation, then washed and allowed todevelop until the 3- to 5-somite stage, when they were analyzed forneurog1 expression. When LiCl was applied at 60% epiboly, theanterior neural plate was posteriorized, as in mutants withhyperactivated Wnt (Dorsky et al., 2003; Heisenberg et al., 2001;Kim et al., 2000; van de Water et al., 2001) (Fig. 2A). The IZ wasenlarged along the AP axis, but neurogenesis inhibition was notaffected (90% of cases, n=50) (Fig. 2A). By contrast, when LiCl wasapplied at 80% epiboly, prominent expression of neurog1 wasobserved ectopically within the MIZ (83% of cases, n=60) (Fig.2B,E,E�, arrows, compare with Fig. 2D,F,F�), with only a moderateeffect on AP patterning. This neurogenic phenotype was cruciallydependent on the stage of LiCl application: incubations at later timesdid not affect IZ formation (88% of cases, n=56) (Fig. 2C). Thus,LiCl applied at 80% epiboly induces the premature commitment ofMIZ progenitors towards differentiation.

Because LIZ cells are less prone to undergo neurogenesis thanMIZ cells, it is possible that LiCl modulates neurogenesisthroughout the IZ, but that its effects are only visible medially in thewild-type embryo. To test this, we further applied LiCl to embryosin which Her5 activity was decreased by injection of her5 antisenseoligonucleotides (her5GripNA). Here, neurog1 expression wasinduced by LiCl throughout the IZ (86.5% of cases, n=52) (notshown), a phenotype never obtained with her5 knockdown alone(Fig. 1B) (Ninkovic et al., 2005). Thus, in the presence of LiCl, IZcells are globally rendered more susceptible to undergoneurogenesis. her5 or her11 expression was not modified by LiCltreatment (96% of cases, n=30) (not shown). LiCl might rathermodulate Her5/Her11 protein activity, or act downstream or inparallel to these factors.

Importantly, the effect of LiCl is opposite to that expected ifendogenous Wnt signaling controlled the behavior of MIZ andLIZ cells: the expression of Wnt ligands in the LIZ would predictthat activated Wnt decreases the tendency of a cell todifferentiation. Hence, the neurogenic effects of LiCl might notbe linked to Wnt signaling activation. To directly test thisinterpretation, we assessed the effects of LiCl on blockade ofcanonical Wnt downstream of LiCl activity. We used hsp70l:tcf3-GFPw26 transgenic embryos, which express a truncated form ofTcf3a (Tcf7l1a – ZFIN) that acts as a dominant repressor ofcanonical Wnt upon heat shock (Lewis et al., 2004). We submittedthese embryos to a heat-shock pulse at 50% epiboly, followed 2hours later by LiCl treatment (at 80% epiboly), and analyzedneurog1 expression at the 3- to 5-somite stage. A 2-hour delayafter the heat-shock pulse is sufficient to significantly antagonize

3139RESEARCH ARTICLEMolecular control of midbrain-hindbrain boundary progenitors

Fig. 2. LiCl enhances neurogenesis in place ofthe IZ independently of canonical Wntsignaling and its effects are mimicked by Gsk3βinhibition. Zebrafish embryos analyzed at the 3- to5-somite stage by in situ hybridization and flat-mounted, anterior up. (A-F�) Treatment with LiCl.At 80% epiboly (B,E,E�), treatment induces ectopicneurog1 expression in place of the MIZ (arrows inB,E�). LiCl posteriorizes the anterior neural platewhen applied earlier (A), and is without effect whenapplied later (C). Bracket in A-D indicates the MIZ.(D,F,F�) Control. (G,H) Effect of LiCl applied at 80%epiboly (G, arrow) in transgenic hsp70l:tcf3-GFPw3

embryos expressing a dominant-negative Tcf3aupon heat shock. Expression of this transgene alonedoes not perturb IZ formation (H). (I-L) The Gsk3βinhibitor OTDZT triggers ectopic neurog1 expressionin place of the MIZ (I,I�, compare with J,J�), andwithin the LIZ when Her5 activity is reduced byher5GripNA injection (L, arrows, compare with K).

DEVELO

PMENT

3140

Wnt activity at gastrulation (Lewis et al., 2004); thus, Gsk3βinhibition by LiCl should occur while downstream components ofcanonical Wnt signaling were blocked. neurog1 expression wasstill strongly induced by LiCl across the MIZ (83% of cases,n=42) (Fig. 2, compare G with H), showing that LiCl is unlikelyto influence IZ neurogenesis via activation of Tcf-mediated Wntsignaling.

Further arguments speak against a role for endogenous Wntsignaling in attributing IZ mediolateral differences. Firstly, inTg(TOP:GFP)w25 transgenic embryos in which GFP expression isunder the control of Tcf-binding sites (Dorsky et al., 2002), wewere unable to reveal active canonical Wnt signaling in theanterior neural plate at the onset of neurogenesis (98% of cases,n�100), whereas GFP expression was obvious in active Wntsignaling domains such as the embryonic margin (not shown).Secondly, lowering Wnt signaling (e.g. in hsp70l:tcf3-GFPw26

transgenic embryos heat-shocked at 50% epiboly) did not changeneurog1 induction, neither by itself nor upon Her5 blockade (notshown). We conclude that canonical Wnt signaling is not involvedin modulating the tendency of IZ cells to undergo neurogenesis,and that LiCl affects IZ neurogenesis independently of canonicalWnt.

LiCl effect on neurogenesis control is mimicked byinhibiting Gsk3βLiCl effects are broad-ranged, but primarily inhibit Gsk3β activity(Berridge et al., 1989; Hedgepeth et al., 1997; Klein and Melton,1996). To confirm that LiCl-induced neurogenesis across the IZ wasmediated by Gsk3β blockade, we tested the effects of the selectiveGsk3β inhibitor OTDZT (Martinez et al., 2002). Similar to LiCltreatment, wild-type embryos were incubated in OTDZT at 80%epiboly for 15 minutes, and processed for neurog1 in situhybridization at the 3- to 5-somite stage. OTDZT induced ectopicneurog1 expression across the MIZ, mimicking the LiCl effect (87%of cases, n=38) (Fig. 2I-J�). Furthermore, as with LiCl, loweringHer5/Her11 dosage revealed that Gsk3β is active throughout the IZ(85% of cases, n=40) (Fig. 2K,L). Therefore, Gsk3β is a crucialelement modulating neurogenesis at the IZ and is required in vivofor the formation and early maintenance of the MIZ.

Activated PKA compensates for Gsk3β inhibitionor reduced E(Spl) activityIn addition to targeting canonical Wnt, Gsk3β is involved in anumber of signaling pathways, where it triggers enhancedphosphorylation of target proteins after these have been primed byphosphorylation via cAMP-dependent protein kinase A (PKA) (Jiaet al., 2002; Price and Kalderon, 2002; Zhang et al., 2005). To assesswhether such a process might be at play during IZ formation, wetested whether PKA activation influences the neurogenic effect ofinhibiting Gsk3β. As reported, only a few embryos (50% in ourcase) injected with capped mRNA encoding a constitutively activecatalytic subunit of PKA (PKA*) (Hammerschmidt et al., 1996a)develop a normal neural plate (Blader et al., 1997). Among these, allformed a normal IZ (Fig. 3, compare A,F with D,H). However,PKA* inhibited the neurogenic effect of LiCl, restoring the MIZwhen LiCl was applied to PKA*-injected embryos (88% of cases,n=61) (Fig. 3, compare C with B). Therefore, general activation ofPKA does not in itself expand the IZ, but it can compensate for theloss of Gsk3β function to maintain neurogenesis inhibition in thislocation.

We next addressed whether PKA activation could alsocompensate for a downregulation of E(Spl) function at the MHB, bytesting the effects of co-injecting her5GripNA and PKA* cappedRNA into wild-type embryos. Whereas blocking Her5 function withgripNAs lead to a loss of the MIZ in the vast majority of cases (92%of cases, n=26) (Fig. 3E), the co-expression of PKA* efficientlyrescued this phenotype and restored MIZ formation (66% of cases,n=50) (Fig. 3, compare G with H). Thus, general activation of PKAalso promotes neurogenesis inhibition downstream of, or in parallelto, Her5 function.

Activated PKA and Gsk3β act in concert withE(Spl) factors to permit IZ formationTo determine whether PKA is in itself required for IZ formation,we blocked its activity by injection of capped RNA encoding adominant-negative form of the PKA regulatory subunit, dnReg(Strähle et al., 1997). dnReg robustly prevents the transcription ofdownstream targets of the Hh pathway, such as spalt (sall1a –ZFIN), that require PKA for their expression (88% of cases, n=72)

RESEARCH ARTICLE Development 135 (18)

Fig. 3. Activated PKA can compensate forreduced Gsk3β or E(Spl) activity topromote IZ formation. Zebrafish embryosanalyzed at the 3- to 5-somite stage forneurog1 and pax2.1 and flat-mounted,anterior up. Bracket, MIZ; arrows indicateectopic neurog1 expression. (A-D) ActivatedPKA (PKA*, injected as capped RNA at theone-cell stage) does not affect IZ formation(compare A with D, see also F) and blocks LiCl-induced neurog1 expression (arrows) acrossthe MIZ (compare B with C). (E-H) Coinjectionof PKA* RNA and her5GripNA (G) blocks theneurogenic effect of her5GripNA alone (E,arrows) and restores the MIZ. (H) Control.

DEVELO

PMENT

(not shown). In embryos expressing dnReg, the MIZ was replacedby ectopic neurog1 expression (84% of cases, n=50) (Fig. 4,compare C with A). When Her5 activity was also inhibited by theco-injection of her5GripNA, ectopic neurog1 expression was alsodetectable within the LIZ (86% of cases, n=35) (Fig. 4, compareE with A-C). To demonstrate that these cells were truly inducedwithin the LIZ and did not originate from the neurog1-positive

r2L domain (see Fig. 1A), we counted the number of cells in r2Lupon co-injection of her5GripNA and dnReg. This was unchangedas compared with control embryos (Fig. 4I). Cell proliferation inthis domain, measured by the number of cells expressing the M-phase marker phospho-histone H3 (PH3), also remainedcomparable to control embryos (Fig. 4J). Hence, migration ofprecursors from the r2L area, possibly compensated for byincreased proliferation in r2L, is unlikely to account for theectopic neurog1-positive cells found within the LIZ. Rather, weconclude that neurog1 expression is ectopically induced in thislocation upon blocking Her5 and PKA activities. Thus, blockingPKA increases the propensity of all IZ cells to undergoneurogenesis, and PKA is crucially required in vivo for MIZformation.

Because blocking PKA or Gsk3β triggers identical phenotypes,we wondered whether these two factors act cooperatively, and so weconcomitantly inhibited both activities by treating dnReg-injectedembryos with OTDZT. As described above, when PKA and Gsk3βare manipulated separately, neurog1 expression is induced in placeof the MIZ, but the LIZ is unaffected (Fig. 2K, Fig. 4C,D). Bycontrast, co-inhibition of PKA and Gsk3β generated ectopicneurog1-positive cells within the LIZ (75% of cases, n=62) (Fig.4F). Again, cell counts revealed no change in the number ofneurog1-positive cells within the r2L domain (Fig. 4I), stronglyarguing for a bona fide ectopic induction of neurog1 expressionwithin the LIZ. These results suggest a dose-dependent process co-regulated by PKA and Gsk3β.

Together, these results place neurogenesis inhibition byGsk3β/PKA downstream of, or in parallel to, E(Spl) activity. Toresolve this issue, we tested whether increased E(Spl) activitywould compensate for decreased Gsk3β/PKA. We blocked PKAin Tg(her5PAC:EGFP)ne1939 transgenic fish (Tallafuss and Bally-Cuif, 2003), which carry one additional copy of her11 under thecontrol of its own regulatory elements, and thereby express threedoses of Her5/Her11 activity in a correct spatio-temporal manner(Ninkovic et al., 2005). This transgenic background permitsnormal MIZ formation in the absence of Her5 (Ninkovic et al.,2005). However, it proved insufficient to block ectopic neurog1activation and rescue the MIZ upon expression of dnReg (79% ofcases, n=56) (Fig. 4, compare H with G). dnReg had no effect onthe level of expression of her5 and her11, and we failed to detectputative PKA or Gsk3β phosphorylation sites on Her5 and Her11(not shown). Although we cannot exclude the possibility thathigher doses of Her5/Her11 could be effective, the mostparsimonious explanation for these results is that Gsk3β/PKA actdownstream of E(Spl) factors.

MIZ and LIZ cells differ intrinsically in a cell cycle-independent processCell-intrinsic components or local signaling cues could accountfor rendering the MIZ and LIZ different in their response to theGsk3β/PKA and Her5/Her11 pathways. To test for the relevanceof cell-intrinsic mechanisms, we first assessed whether MIZ andLIZ differ in lineage. Extensive cell exchanges across the midlineof the zebrafish embryonic neural tube have been documented, aswell as dorsoventral dispersion, after the 2-somite stage (Papanand Campos-Ortega, 1997), but not before (Woo and Fraser,1995). However, IZ precursors have not been specifically studiedin this context. To address this, we time-lapsed MIZ and LIZprecursors from the shield stage onwards using confocalmicroscopy in Tg(her5PAC:EGFP)ne1939 transgenic embryos, inwhich her5-expressing cells are also positive for GFP (Tallafuss

3141RESEARCH ARTICLEMolecular control of midbrain-hindbrain boundary progenitors

Fig. 4. Activated PKA is required for IZ formation in cooperationwith Gsk3β and E(Spl) activity. (A-H) Zebrafish embryos analyzed atthe 3- to 5-somite stage for neurog1 and pax2.1 and flat-mounted,anterior up. (A-F) High magnifications of one half of the neural plate;blue arrows indicate ectopic neurog1 expression, white arrowheadsindicate neurog1-free LIZ. (A-C,E) Expression of a dominant-negativeform of PKA (dnReg) induces neurog1 expression across the MIZ (C), asoccurs upon downregulating E(Spl) (B). Concomitant blockade of bothpathways leads to ectopic LIZ neurog1 expression (E, compare withB,C). (D,F) Inhibiting Gsk3β or PKA activity with OTDZT leads to loss ofthe MIZ (C,D, see also Fig. 2K,K�), whereas co-inhibition also inducesectopic neurog1 expression within the LIZ (F). (G,H) The effect ofblocking PKA remains unchanged in the Tg(her5PAC:EGFP) background(H, compare with C), in which one additional copy of the her11 gene isexpressed. (I,J) The number of neurog1-expressing cells (I) and of cellsin M phase (J) in the r2L domain is unchanged in embryos in whichHer5 and PKA blockade triggers ectopic LIZ neurog1 expression(lightest gray boxes). In I, several other conditions triggering ectopic LIZneurog1 expression are further compared (blockade of Gsk3β and PKA,medium gray box; blockade of Her5 and Him, dark gray box) with thesame result. Bars indicate standard errors; n, the number of embryosanalyzed. D

EVELO

PMENT

3142 RESEARCH ARTICLE Development 135 (18)

Fig. 5. MIZ and LIZ cells differ in lineage but not in proliferation characteristics. (A) The origin of cells located within the MIZ (outlined indark green), LIZ (outlined in pale green), r2M (outlined in purple) and r2L (outlined in pink) was tracked in live confocal movies. The cells werefollowed from shield (top left) to tailbud stage (bottom right) with membrane-localized 2xlckmRFP (red) in Tg(her5PAC:EGFP)ne1939 transgenicembryos. GFP (green) reveals the location of the IZ at tailbud stage (see t=175 and t=200 minutes). Confocal optical projections at 25-minuteintervals, in which some tracked cells are identified with colored dots and lines. The same color is used for the same cell on each panel, andindicates the location of the cell at each time. The line indicates the path that will be followed by this cell to reach its final position at t=200minutes. The tracking of individual cells shows that the four populations (MIZ, LIZ, r2M and r2L) maintain their relative positions throughout themovie. Landmarks: location of the notochord, white arrow; IZ, dots. Number of embryos analyzed is 4. The total number of cells tracked in themovie illustrated is 26. (B) Schematics illustrating the relative positions of the four cell populations (color-coded as above) at shield and tailbudstages. Dark gray, notochord; light gray, neuroectoderm; green, Tg(her5PAC:EGFP)ne1939 expression domain. (C) Comparison of cell cyclecharacteristics of MIZ and LIZ cells during the time of IZ formation. The percentage of cells incorporating BrdU among all MIZ or LIZ cells (DAPI) isidentical at all stages. D

EVELO

PMENT

and Bally-Cuif, 2003). With the help of a lineage reporter(2xlckmRFP) (Megason and Fraser, 2003) injected as cappedRNA into a subset of cells, by playing the movies backwards fromthe tailbud stage we could trace the origin of single cells thatcontribute to the MIZ and LIZ (as well as to the r2M and r2Ldomains). We observed that these precursor cell groups remainedspatially segregated at all times between the shield and tailbudstages (Fig. 5A,B). In particular, we did not observe anycontribution of MIZ precursors to the LIZ and vice versa. Thisraises the possibility that MIZ and LIZ precursors inherit differentdeterminants that might influence their sensitivity to neurogenesisinhibitors.

These determinants might act on cell proliferation, on the degreeof cellular commitment toward differentiation, or both. Because cellcycle kinetics influence the capacity of a cell to enter theneurogenesis process (Calegari and Huttner, 2003), we testedwhether MIZ and LIZ cells differed in their proliferationcharacteristics. Markers for the S (BrdU incorporation) and M (PH3,and mitotic figures) phases were analyzed together with the IZmarker her5 in triple-labeled preparations at 75% epiboly to tailbud.At these stages, there is still a tendency for cells to dividesynchronously (Kimmel et al., 1994) and thus the co-analysis of S-and M-phase markers on single specimens was important todistinguish between differences in cell cycle length and cell divisiontiming. Cell cycle kinetics, as well as the expression of p27Xic(cdkn1c – ZFIN), did not significantly differ between the MIZ andLIZ in any of the more than 50 embryos analyzed (Fig. 5C, and datanot shown). These observations suggest that the cell divisioncharacteristics of the MIZ and LIZ are comparable, and that thesepopulations differ in their commitment in a manner independent ofcell cycle control.

Gli1 counteracts the IZ neurogenesis inhibitionprocess and accounts for the differentialsensitivity of MIZ and LIZ cells to neurogenesisinhibitorsAmong commitment factors expressed within the early neural plate,transcription factors of the Gli family have been implicated inneurogenesis control in many systems (reviewed by Agathocleouset al., 2007; Ruiz et al., 2002). Furthermore, Gli factors have beenidentified as intracellular targets for the PKA/Gsk3β pair (Huangfuand Anderson, 2006; Riobo and Manning, 2007). gli1 appears to beexpressed differentially between the MIZ and LIZ: in agreementwith published data (Karlstrom et al., 2003), we observed that gli1is transcribed within the anterior neural plate following a clearmediolaterally decreasing gradient starting at 80% epiboly (Fig.6A,C,E, and data not shown). Double staining for GFP, identifyingthe IZ in Tg(her5PAC:EGFP)ne1939 embryos, further confirmed thatthis gradient is also found at the level of the IZ (Fig. 6B,D,F). Threeother Gli genes have been identified in zebrafish (Karlstrom et al.,2003; Ke et al., 2005; Tyurina et al., 2005). gli2b is not expressed atdetectable levels within the IZ area at these stages, whereasexpression of gli2a and gli3 is ubiquitous (not shown).

To address the function of Gli1 during IZ formation, we blockedgli1 translation by injection of a gli1 morpholino (MO) (Karlstromet al., 2003) into wild-type embryos. This did not alter the IZ area(82% of cases, n=71) (Fig. 6, compare H,L with J,N; Fig. 7, compareE with A), although it was efficient at blocking expression of the Hhtarget nkx2.1 along the ventral midline of the anterior neural tube(96% of cases, n=30) (not shown). However, blocking Gli1 totallyabolished the neurogenic effect of LiCl: in gli1 MO-injectedembryos subjected to LiCl treatment at 80% epiboly, the MIZformed normally (82% of cases, n=52) (Fig. 6, compare I with G).

3143RESEARCH ARTICLEMolecular control of midbrain-hindbrain boundary progenitors

Fig. 6. MIZ and LIZ cells differ in their expression of Gli1, which promotes neurogenesis and antagonizes PKA during IZ formation.(A-F) Expression of gli1 (blue) and Her5-GFP (green) in Tg(her5PAC:EGFP)ne1939 transgenic zebrafish embryos (flat-mounted, anterior up) shows thatgli1 is transcribed at higher levels in the MIZ (arrows) than in the LIZ. B,D,F are high-magnification views of the boxed areas in A,C,E respectively,viewed sequentially (from top to bottom) under bright field, fluorescence and simultaneous bright field and fluorescence illumination.(G-N) Embryos analyzed at the 3- to 5-somite stage for neurog1 and pax2.1 (color-coded) and flat-mounted, anterior up. Blue arrows indicateectopic neurog1 expression. (G-J) Blocking Gli1 activity does not affect IZ formation (H, compare with control J), but prevents the induction ofneurog1 expression across the MIZ that is normally triggered by LiCl (I, compare with G). (K-N) Blocking Gli1 activity is sufficient to rescue the MIZnormally lost upon PKA downregulation (M, compare with K). (N) Control. D

EVELO

PMENT

3144

Thus, blocking Gli1 function can increase neurogenesis inhibitionand rescue IZ formation in the absence of Gsk3β activity. Using asimilar approach, we tested whether blocking Gli1 couldcompensate for the lack of PKA activity, by co-injecting gli1 MOand dnReg capped RNA into wild-type embryos. The co-inhibitionof Gli1 and PKA abolished the neurogenic effect of blocking PKAalone and rescued MIZ formation (84% of cases, n=61) (Fig. 6,compare M with K). These results demonstrate that Gli1 exerts aneurogenesis-promoting activity that opposes the activity ofGsk3β/PKA. Under normal conditions, the activity of Gli1 is sub-threshold, such that neurogenesis inhibition is obtained. It becomesvisible in the absence of Gsk3β or PKA activity, when the repressionof Gli1 alone suffices to restore neurogenesis inhibition.

The co-regulation of IZ formation by Gsk3β/PKA and E(Spl)factors described above suggests that Gli1 function should alsoinfluence E(Spl) function. To verify this point, we reduced E(Spl)-mediated neurogenesis inhibition by injection of her5GripNA orher11GripNA into wild-type embryos, and we simultaneouslyblocked Gli1 by the co-injection of gli1 MO. The ectopicneurogenesis triggered by Her5 blockade was abolished by the lossof Gli1 function, restoring the IZ (89% of cases, n=72) (Fig. 7,compare F with B). Likewise, blocking Gli1 together with Her11rescued the MIZ as compared with the effect of blocking Her11alone (74% of cases, n=67) (Fig. 7, compare G with C). Thus, Gli1expression also acts antagonistically to E(Spl) activity by promotingIZ neurogenesis, an effect that becomes apparent when E(Spl)dosage is reduced.

The neurogenesis-promoting function of Gli1 and its MIZ-specificexpression make it a good candidate to enhance the neurogenesispotential of the MIZ. As described, blocking Gli1 activity increasesMIZ sensitivity to two copies (or fewer) of Her5/Her11 (Fig. 7F,G).In addition, in the absence of Gli1, we observed that both the MIZand LIZ concomitantly lost their responsiveness to a furtherdownregulation of Her5/Her11 to now upregulate neurog1 in anidentical manner (90% of cases, n=56) (Fig. 7H,H�). These resultsidentify Gli1 as a crucial element rendering MIZ and LIZ cellsdifferentially sensitive to E(Spl) factors.

IZ formation and Gli1 activity are not under thecontrol of Hh signalingWe next searched for mechanisms controlling Gli1 expression oractivity within the MIZ. Hh signaling is active in the presumptiveaxial mesoderm from 50% epiboly onwards (Ertzer et al., 2007;Krauss et al., 1993), and Gli1 behaves as a classical positiveactivator of Hh targets. Thus, we studied whether IZ formation andGli1 expression require Hh signaling. We blocked Hh signalingupstream of PKA/Gsk3β action by incubating wild-type embryos incyclopamine. Cyclopamine blocks the transmembrane proteinSmoothened (Chen et al., 2002), which normally initiatesintracellular Hh signaling events and in particular the inhibition ofGsk3β and PKA (Huangfu and Anderson, 2006; Riobo andManning, 2007). As expected, such treatment performed at 50%epiboly efficiently inhibited expression of myoD in adaxial cells(97% of cases, n=40) (Fig. 8, compare H with G) (Barresi et al.,2000; Chen et al., 2001) and of spalt in the anterior neural plate (notshown). Surprisingly, however, cyclopamine treatment did not affectgli1 expression in the IZ area at any stage (100% of cases, n>20 foreach stage) (Fig. 8A-F, and data not shown). Whatever the stage ofapplication, cyclopamine also did not affect IZ formation (Fig. 8,compare J-L with N). Because Hh signaling might promote Gli1activity rather than expression, we assessed whether cyclopaminedownregulated Gli1 function. For this, we first reduced E(Spl)activity using her5GripNA, and next applied cyclopamine at 50%epiboly. Whereas inhibiting Gli1 function in such cases led to arescue of the MIZ (Fig. 7F), cyclopamine was without effect andwas incapable of counteracting ectopic neurog1 induction (89% ofcases, n=75) (Fig. 8, compare M with I). We conclude thatcyclopamine treatment affects neither the expression nor thefunction of Gli1 at the IZ. Similar results were obtained in smub641

mutants, deficient in Smoothened function (Varga et al., 2001),analyzed for gli1 expression at 70% epiboly, tailbud and 3 somites(n>30 embryos in each case) (not shown). Finally, we also observedthat gli1 expression and IZ formation were unaffected whentransduction of Hh signaling was blocked upon overexpression of aHh-insensitive form of the receptor Patched, PatchedΔloop2

RESEARCH ARTICLE Development 135 (18)

Fig. 7. Gli1 accounts for the differential sensitivity of the MIZ and LIZ to E(Spl) neurogenesis inhibitors. (A-H�) Zebrafish embryosanalyzed at the 3- to 5-somite stage for neurog1 and pax2.1 (color-coded) and flat-mounted, anterior up. Brackets, MIZ; blue arrows, ectopicneurog1 expression. In the absence of Gli1, which in itself has no effect on neurog1 expression (compare E with A), MIZ cells are unaffected by theloss of Her5 or Her11 (F,G, compare with B,C). MIZ and LIZ cells become concomitantly affected by a further downregulation of E(Spl) activity (lossof both Her5 and Her11), when they both upregulate neurog1, in an identical manner (H,H�, compare with D). D

EVELO

PMENT

(Briscoe et al., 2001) (100% of cases, n>100) (Fig. 8, compare Owith P), even in the absence of Her5 function (not shown). Theseobservations strongly suggest that the neurogenic activity of Gli1 atthe IZ does not require Hh signaling.

DISCUSSIONThe IZ gives rise in spatio-temporal order to the large majority ofmidbrain-hindbrain neurons (Tallafuss and Bally-Cuif, 2003). Wepreviously demonstrated that E(Spl) activity is a crucialmechanism inhibiting neurogenesis to permit IZ formation(Geling et al., 2003; Geling et al., 2004; Ninkovic et al., 2005).We identify here a collaborative mechanism, relying on theactivity of the two kinases, PKA and Gsk3β. The E(Spl) andGsk3β/PKA pathways act in a dose-dependent manner, and oneof their activities is to oppose the neurogenic effect of thetranscription factor Gli1. Gli1, expressed medially, enhances thetendency of MIZ cells to undergo neurogenesis, a mechanism thatmight account for the earlier differentiation of basal versus alarplate neurons during MH development. Together, these resultshelp refine a molecular model for the sequential differentiation ofmidbrain-hindbrain neurons along the AP and mediolateral (DV)axes (Fig. 9).

Co-regulation of IZ formation by the Gsk3β/PKAand E(Spl) pathwaysAn important finding of our work is the identification of Gsk3βand PKA as new mediators of IZ formation. Both enzymesoften act sequentially on the same targets in vivo, PKA priming

target proteins for a subsequent phosphorylation by Gsk3β(Price and Kalderon, 2002), but alternative models areemerging in which PKA and Gsk3β independentlyphosphorylate the same target (Taurin et al., 2006). At the IZ,the enhanced effect of lowering both Gsk3β and PKA, and thefact that constitutive PKA activation can compensate forreduced Gsk3β, suggest the involvement of an unconventionaldose-dependent process incorporating the level of activity ofboth enzymes. PKA and Gsk3β might be rate-limiting for thefull phosphorylation necessary to functionally modify the sametarget. Alternatively, PKA and Gsk3β might act on distinctmolecular targets, cooperating in parallel pathways during IZformation. Our results do not allow us to distinguish betweenthese possibilities.

E(Spl) and Gsk3β/PKA act in an additive manner, and ourobservations support parallel activities of E(Spl) and Gsk3β/PKAon a dose-dependent process sensitive to the global level of thesepathways. Gsk3β/PKA and E(Spl) inhibition might converge ontocommon targets promoting neurogenesis inhibition, or haveparallel targets cooperating in the neurogenesis inhibition process.The direct targets of E(Spl) activity during IZ formation remainunknown, although it is known to downregulate expression ofneurog1, coe2 and p27Xic (cdkn1c – ZFIN) (Geling et al., 2004).Many targets inhibited by Gsk3β/PKA phosphorylation have beenidentified, including factors controlling cell cycle (e.g. N-myc1;Mycn) (Kenney et al., 2004; Mill et al., 2005), neurogenesis andcell differentiation (e.g. Neurogenin 2, XNeuroD, Xash1 andMash1) (Ma et al., 2008; Moore et al., 2002), or both (e.g. β-

3145RESEARCH ARTICLEMolecular control of midbrain-hindbrain boundary progenitors

Fig. 8. Gli1 expression and activity are notunder the control of Hh signaling during IZformation. (A-F) Expression of gli1 is unaffectedupon cyclopamine (A,C,E) or mock treatment (B,D,F)applied between the dome stage and the stage ofanalysis [70% epiboly (A,B), tailbud (C,D) and 3somites (E,F)]. (G,H) Expression of myoD (blue) andpax2.1 (red) on 6-somite zebrafish embryos (dorsalviews, anterior left) upon cyclopamine treatment (H)as compared with mock treatment (G). The samecyclopamine concentration was used as in A,C,E;treatment between 50% epiboly and 6 somites.myoD is strongly decreased in adaxial cells (H, whitearrowhead; compare with G, blue arrow).(I-N) Expression of neurog1 (blue) and pax2.1 (red)at the 3- to 5-somite stage (anterior up) isunaffected upon cyclopamine treatment applied atany stage between dome and 3-5 somites (J-L,compare with N). Cyclopmaine also does not rescuethe effects of her5GripNA injections (M, comparewith I), in contrast to the effect of blocking Gli1activity (see Fig. 7F). (O,P) Expression of neurog1(blue) and pax2.1 (red) at 3-5 somites is unaffectedby blockade of Hh signaling via overexpression ofPatchedΔloop (O, compare with control P).

DEVELO

PMENT

3146

catenin or Gli proteins) (reviewed by Frame and Cohen, 2001;Huangfu and Anderson, 2006; Riobo and Manning, 2007). Noneof our observations suggests a major role for cell proliferationcontrol in IZ formation, corroborating previous findings thatneurogenesis does not systematically follow cell cycle exit at theearly neural plate stage (Geling et al., 2003; Hardcastle andPapalopulu, 2000; Harris and Hartenstein, 1991). Rather, in thissystem, entry into the neurog1-positive state follows informationunlinked to cell cycle characteristics and we favor commitmentfactors as the main targets for the Gsk3β/PKA pathway.

There are target sites for Gsk3β on zebrafish NeuroD (Mooreet al., 2002) but not on Neurog1 (our observations) and, in the IZ,Gsk3β/PKA affects neurog1 transcription rather than its activity.We failed to identify Gsk3β or PKA target sites on Her5 andHer11 (not shown), but an indirect role of Gsk3β/PKA on theactivity of these factors cannot be excluded. Gli1 is an obviouscandidate target, as Gsk3β/PKA classically regulate the nucleartranslocation (Gli1) or processing (Gli2, Gli3) of Gli factors in amanner antagonized by Hh signaling (Huangfu and Anderson,2006; Riobo and Manning, 2007). However, Gsk3β/PKA alsoblock neurogenesis in the LIZ, where gli1 is not expressed.Hence, other direct targets of Gsk3β/PKA remain to be identifiedthat permit IZ formation.

Finally, as with E(Spl) factors, the upstream pathwaysinvolving Gsk3β/PKA at the IZ are currently unknown.Classically, Wnt and Hh inhibit Gsk3β and/or PKA to permitsignal transduction (reviewed by Hooper and Scott, 2005;Huangfu and Anderson, 2006; Logan and Nusse, 2004; Rioboand Manning, 2007). However, Wnt reporters remained silent inthe anterior neural plate at the stage of interest, and blockingWnt or Hh activities or ubiquitously enhancing PKA activity did

not expand the IZ. IZ formation might be achieved through aconstitutive activation of Gsk3β/PKA in concert with localcofactors, and/or through signaling pathways other than Wntand Hh. A number of upstream regulators have been identifiedfor these kinases – in particular, classical mitogenic pathwaysdriven by PI3 kinase and Akt (Dudek et al., 1997; Kenney et al.,2004). It will be important to test their influence during IZformation.

Gli1 activity renders MIZ cells less sensitive toE(Spl)- and Gsk3β/PKA-mediated neurogenesisinhibitiongli1 expression is concentrated medially and has a neurogenesis-promoting effect, albeit at sub-threshold levels. Significantly,blocking Gli1 activity abolishes the difference between MIZ andLIZ sensitivity towards E(Spl)-mediated neurogenesis inhibition.Although multiple additional blocks might still exist betweenneurog1 expression and the completion of the neuronaldifferentiation program, the mechanism uncovered here is likely tofacilitate the transition of MIZ cells towards neuronal productionand might be relevant to the earlier maturation of the midbrain basalplate that is observed in all vertebrates during development (Easteret al., 1994; Puelles et al., 1987; Ross et al., 1992; Wilson et al.,1990). Although E(Spl) and Gli1 activities have opposing effects onneurogenesis at the IZ, our triple knockdown experiments of Gli1,Her5 and Her11 demonstrate that Gli1 itself is dispensable forneurog1 expression in the absence of the inhibitory activity of E(Spl)factors. Gli1 activity might, however, be required in this context forthe progression of neuronal differentiation beyond neurog1expression, an issue that remains to be tested.

Gli1 exerts positive and negative effects on neurogenesisdepending on the cell type or state. Neurogenesis progression failswhen Gli1 function is blocked in the Xenopus neural plate orzebrafish retina (Masai et al., 2005; Nguyen et al., 2005). Bycontrast, downregulation of Gli1 prevents re-entry into S phase inthe chicken ventral neural tube (Cayuso et al., 2006). Overall,Gli1 might bring cells closer to differentiation, hence, pushingearly progenitors to amplify, and later progenitors to a final cellcycle. At the IZ, however, where an amplifying population has notbeen observed, an effect on the cell cycle is improbable and wepropose that Gli1 targets cell cycle-independent commitmentgenes. Further arguments supporting this hypothesis stem fromprevious analyses of zebrafish gli1 (detour) mutants (Karlstromet al., 1999). These mutants lack cranial motoneurons, includingnerve III (Chandrasekhar et al., 1999) [which is likely to originatefrom the MIZ (Tallafuss and Bally-Cuif, 2003) (K. Webb andC.S., unpublished)], but display no defects in neurogenesis at 18hours, pointing to a differentiation rather than an induction defect(Chandrasekhar et al., 1999). However, the lack of cranialmotoneurons of gli1 mutants can be mimicked by loss ofSmoothened function or by cyclopamine treatment (Vanderlaanet al., 2005), and hence might be a result of impaired Hhsignaling, which contrasts with our interpretation of Gli1regulation at the MIZ. In any case, both analyses point to a rolefor Gli1 in promoting differentiation in a manner independent ofcell cycle control.

Hh-independent regulation of Gli1Given the importance of Gli1 function in neurogenesis progression,it remains crucial to uncover the pathway(s) controlling Gli1expression in the MIZ. Gsk3β/PKA can downregulate Gli1(Huangfu and Anderson, 2006; Riobo and Manning, 2007) and

RESEARCH ARTICLE Development 135 (18)

Fig. 9. Model for IZ formation in zebrafish. One side of the neuralplate is represented (gray bar, the axial midline; red, MIZ and gli1expression domain; black, LIZ). For simplicity, Gsk3β/PKA are showntogether, although they might act in parallel. The E(Spl) and Gsk3β/PKApathways inhibit neurogenesis throughout the IZ in a dose-dependentand cooperative manner. Gli1, expressed in the MIZ, actsantagonistically and promotes neurogenesis. Gli1 activity might bepartially downregulated by Gsk3β/PKA. Responding to the sum of theseactivities, MIZ cells are more prone to commit towards neurogenesisthan are LIZ cells. Gli1Act, activated Gli1; Gli1P, phosphorylated Gli1.

DEVELO

PMENT

might do so during IZ formation. This is likely to occur at the post-transcriptional level, as gli1 transcript levels were not modified byGsk3β or PKA blockade (not shown). However, this downregulationof Gli1 by Gsk3β/PKA is only partial, because blocking Gli1activity has further phenotypic consequences.

A most surprising aspect of our study is that Gli1 expression andactivity at the IZ are not under the control of Hh signaling alone.Blocking Hh with cyclopamine, in smoothened (smu) mutants or byoverexpressing a Hh-insensitive form of Patched, affected neithergli1 expression nor IZ formation. Although surprising, thesefindings are in keeping with previous observations by Karlstrom etal. who reported maintenance of gli1 expression in the anteriorneural plate of smu mutants or cyclopamine-treated embryos(Karlstrom et al., 2003). Recently, a discrepancy between the effectsof smu mutations or cyclopamine and the effects of forskolin, whichactivates PKA, were noted during retinal neurogenesis in zebrafish:whereas forskolin strongly impaired neurogenesis progression, smumutant or cyclopamine-treated embryos exhibited only mild defects(Masai et al., 2005). Our results are comparable to theseobservations and, we believe, question the role of Hh, if any, in Gli1regulation across the IZ. In a few instances, activation of Gli1 (orGli2) has been reported to follow activation of the Erk pathway byFgf signaling (Brewster et al., 2000; Riobo et al., 2006a; Riobo etal., 2006b) or of the PI3K and Akt pathways by mitogens (Riobo etal., 2006b). It is possible that such an atypical, Hh-independentmechanism is involved at the midline of the anterior neural plate.Unravelling this process will be important to our understanding ofneurogenesis control and the cross-regulatory activities of Hhsignaling pathway components.

We are grateful to Dr Kenji Imai and to members of the Bally-Cuif laboratoryfor support and discussions; to Silvia Krutsch, Birgit Tannhäuser and StefanieTopp for technical support; to the R. Köster laboratory for help with time-lapsemicroscopy; to Dr William Norton for his careful and critical reading of themanuscript; and to Drs Rolf Karlstrom, Uwe Strähle and Masanari Takamiya forsharing knowledge and reagents on the Hedgehog pathway andneurogenesis. Work in the Bally-Cuif laboratory is funded by a junior groupgrant from the Volkswagen Association, the EU 6th Framework IntegratedProject ZF-Models (LSHC-CT-2003-503466), the Life Science Association (GSF2005/01), the Center for Protein Science-Munich (CIPSM), and a specialresearch grant from the Institut du Cerveau et de la Moelle épinière (ICM,Paris).

ReferencesAgathocleous, M., Locker, M., Harris, W. A. and Perron, M. (2007). A general

role of hedgehog in the regulation of proliferation. Cell Cycle 6, 156-159.Bae, Y. K., Shimizu, T. and Hibi, M. (2005). Patterning of proneuronal and inter-

proneuronal domains by hairy- and enhancer of split-related genes in zebrafishneuroectoderm. Development 132, 1375-1385.

Bally-Cuif, L. and Hammerschmidt, M. (2003). Induction and patterning ofneuronal development, and its connection to cell cycle control. Curr. Opin.Neurobiol. 13, 16-25.

Bally-Cuif, L., Goridis, C. and Santoni, M. J. (1993). The mouse NCAM genedisplays a biphasic expression pattern during neural tube development.Development 117, 543-552.

Barresi, M. J., Stickney, H. L. and Devoto, S. H. (2000). The zebrafish slow-muscle-omitted gene product is required for Hedgehog signal transduction andthe development of slow muscle identity. Development 127, 2189-2199.

Berridge, M. J., Downes, C. P. and Hanley, M. R. (1989). Neural anddevelopmental actions of lithium: a unifying hypothesis. Cell 59, 411-419.

Blader, P., Fischer, N., Gradwohl, G., Guillemont, F. and Strähle, U. (1997).The activity of neurogenin1 is controlled by local cues in the zebrafish embryo.Development 124, 4557-4569.

Brewster, R., Mullor, J. L. and Ruiz i Altaba, A. (2000). Gli2 functions in FGFsignaling during antero-posterior patterning. Development 127, 4395-4405.

Briscoe, J., Chen, Y., Jessell, T. M. and Struhl, G. (2001). A hedgehog-insensitiveform of patched provides evidence for direct long-range morphogen activity ofsonic hedgehog in the neural tube. Mol. Cell 7, 1279-1291.

Buckles, G. R., Thorpe, C. J., Ramel, M. C. and Lekven, A. C. (2004).Combinatorial Wnt control of zebrafish midbrain-hindbrain boundary formation.Mech. Dev. 121, 437-447.

Calegari, F. and Huttner, W. B. (2003). An inhibition of cyclin-dependent kinasesthat lengthens, but does not arrest, neuroepithelial cell cycle induces prematureneurogenesis. J. Cell Sci. 116, 4947-4955.

Cayuso, J., Ulloa, F., Cox, B., Briscoe, J. and Marti, E. (2006). The Sonichedgehog pathway independently controls the patterning, proliferation andsurvival of neuroepithelial cells by regulating Gli activity. Development 133, 517-528.

Chandrasekhar, A., Schauerte, H. E., Haffter, P. and Kuwada, J. Y. (1999). Thezebrafish detour gene is essential for cranial but not spinal motor neuroninduction. Development 126, 2727-2737.

Chen, J. K., Taipale, J., Cooper, M. K. and Beachy, P. A. (2002). Inhibition ofHedgehog signaling by direct binding of cyclopamine to Smoothened. GenesDev. 16, 2743-2748.

Chen, W., Burgess, S. and Hopkins, N. (2001). Analysis of the zebrafishsmoothened mutant reveals conserved and divergent functions of hedgehogactivity. Development 128, 2385-2396.

Distel, M. and Köster, R. (2007). In vivo time-lapse imaging of zebrafishembryonic development. CSH Protocols doi:10.1101/pdb.prot4816.

Dorsky, R. I., Sheldahl, L. C. and Moon, R. T. (2002). A transgenic Lef1/beta-catenin-dependent reporter is expressed in spatially restricted domainsthroughout zebrafish development. Dev. Biol. 241, 229-237.

Dorsky, R. I., Itoh, M., Moon, R. T. and Chitnis, A. (2003). Two tcf3 genescooperate to pattern the zebrafish brain. Development 130, 1937-1947.

Dudek, H., Datta, S. R., Franke, T. F., Birnbaum, M. J., Yao, R., Cooper, G. M.,Segal, R. A., Kaplan, D. R. and Greenberg, M. E. (1997). Regulation ofneuronal survival by the serine-threonine protein kinase Akt. Science 275, 661-665.

Easter, S. S., Jr, Burrill, J., Marcus, R. C., Ross, L., Taylor, J. S. H. and Wilson,S. W. (1994). Initial tract formation in the vertebrate brain. Prog. Brain Res. 102,79-93.

Ertzer, R., Muller, F., Hadzhiev, Y., Rathnam, S., Fischer, N., Rastegar, S. andSträhle, U. (2007). Cooperation of sonic hedgehog enhancers in midlineexpression. Dev. Biol. 301, 578-589.

Frame, S. and Cohen, P. (2001). GSK3 takes centre stage more than 20 yearsafter its discovery. Biochem. J. 359, 1-16.

Geling, A., Itoh, M., Tallafuss, A., Chapouton, P., Tannhäuser, B., Kuwada, J.Y., Chitnis, A. B. and Bally-Cuif, L. (2003). bHLH transcription factor Her5 linkspatterning to regional inhibition of neurogenesis at the midbrain-hindbrainboundary. Development 130, 1591-1604.

Geling, A., Plessy, C., Rastegar, S., Strähle, U. and Bally-Cuif, L. (2004). Her5acts as a prepattern factor that blocks neurogenin1 and coe2 expressionupstream of Notch to inhibit neurogenesis at the midbrain-hindbrain boundary.Development 131, 1993-2006.

Hammerschmidt, M., Bitgood, M. J. and McMahon, A. P. (1996a). Proteinkinase A is a common negative regulator of Hedgehog signaling in thevertebrate embryo. Genes Dev. 10, 647-658.

Hammerschmidt, M., Pelegri, F., Mullins, M. C., Kane, D. A., van Eeden, F. J.,Granato, M., Brand, M., Furutani-Seiki, M., Haffter, P., Heisenberg, C. P. etal. (1996b). dino and mercedes, two genes regulating dorsal development in thezebrafish embryo. Development 123, 95-102.

Hans, S., Scheer, N., Riedl, I. v. Weizsacker, E., Blader, P. and Campos-Ortega,J. A. (2004). her3, a zebrafish member of the hairy-E(spl) family, is repressed byNotch signalling. Development 131, 2957-2969.

Hardcastle, Z. and Papalopulu, N. (2000). Distinct effects of XBF-1 in regulatingthe cell cycle inhibitor p27(XIC1) and imparting a neural fate. Development 127,1303-1314.

Harris, W. A. and Hartenstein, V. (1991). Neuronal determination without celldivision in Xenopus embryos. Neuron 6, 499-515.

Hatakeyama, J. and Kageyama, R. (2006). Notch1 expression isspatiotemporally correlated with neurogenesis and negatively regulated byNotch1-independent Hes genes in the developing nervous system. Cereb. Cortex16 Suppl. 1, i132-i137.

Hatakeyama, J., Bessho, Y., Katoh, K., Ookawara, S., Fujioka, M., Guillemot,F. and Kageyama, R. (2004). Hes genes regulate size, shape and histogenesisof the nervous system by control of the timing of neural stem cell differentiation.Development 131, 5539-5550.

Hedgepeth, C. M., Conrad, L. J., Zhang, J., Huang, H. C., Lee, V. M. andKlein, P. S. (1997). Activation of the Wnt signaling pathway: a molecularmechanism for lithium action. Dev. Biol. 185, 82-91.

Heisenberg, C. P., Houart, C., Take-Uchi, M., Rauch, G. J., Young, N.,Coutinho, P., Masai, I., Caneparo, L., Concha, M. L., Geisler, R. et al. (2001).A mutation in the Gsk3-binding domain of zebrafish Masterblind/Axin1 leads toa fate transformation of telencephalon and eyes to diencephalon. Genes Dev.15, 1427-1434.

Hirata, H., Tomita, K., Bessho, Y. and Kageyama, R. (2001). Hes1 and Hes3regulate maintenance of the isthmic organizer and development of themid/hindbrain. EMBO J. 20, 4454-4466.

3147RESEARCH ARTICLEMolecular control of midbrain-hindbrain boundary progenitors

DEVELO

PMENT

3148

Hooper, J. E. and Scott, M. P. (2005). Communicating with Hedgehogs. Nat. Rev.Mol. Cell Biol. 6, 306-317.

Huangfu, D. and Anderson, K. V. (2006). Signaling from Smo to Ci/Gli:conservation and divergence of Hedgehog pathways from Drosophila tovertebrates. Development 133, 3-14.

Jia, J., Amanai, K., Wang, G., Tang, J., Wang, B. and Jiang, J. (2002).Shaggy/GSK3 antagonizes Hedgehog signalling by regulating Cubitusinterruptus. Nature 416, 548-552.

Kageyama, R., Ohtsuka, T., Hatakeyama, J. and Ohsawa, R. (2005). Roles ofbHLH genes in neural stem cell differentiation. Exp. Cell Res. 306, 343-348.

Karlstrom, R. O., Talbot, W. S. and Schier, A. F. (1999). Comparative syntenycloning of zebrafish you-too: mutations in the Hedgehog target gli2 affectventral forebrain patterning. Genes Dev. 13, 388-393.

Karlstrom, R. O., Tyurina, O. V., Kawakami, A., Nishioka, N., Talbot, W. S.,Sasaki, H. and Schier, A. F. (2003). Genetic analysis of zebrafish gli1 and gli2reveals divergent requirements for gli genes in vertebrate development.Development 130, 1549-1564.

Ke, Z., Emelyanov, A., Lim, S. E., Korzh, V. and Gong, Z. (2005). Expression ofa novel zebrafish zinc finger gene, gli2b, is affected in Hedgehog and Notchsignaling related mutants during embryonic development. Dev. Dyn. 232, 479-486.

Kenney, A. M., Widlund, H. R. and Rowitch, D. H. (2004). Hedgehog and PI-3kinase signaling converge on Nmyc1 to promote cell cycle progression incerebellar neuronal precursors. Development 131, 217-228.

Kim, C. H., Oda, T., Itoh, M., Jiang, D., Artinger, K. B., Chandrasekharappa,S. C., Driever, W. and Chitnis, A. B. (2000). Repressor activity of Headless/Tcf3is essential for vertebrate head formation. Nature 407, 913-916.

Kimmel, C. B., Warga, R. M. and Kane, D. A. (1994). Cell cycles and clonalstrings during formation of the zebrafish central nervous system. Development120, 265-276.

Kimmel, C. B., Ballard, W. W., Kimmel, S. R., Ullmann, B. and Schilling, T. F.(1995). Stages of embryonic development of the zebrafish. Dev. Dyn. 203, 253-310.

Klein, P. S. and Melton, D. A. (1996). A molecular mechanism for the effect oflithium on development. Proc. Natl. Acad. Sci. USA 93, 8455-8459.

Korzh, V., Sleptsova, I., Liao, J., He, J. and Gong, Z. (1998). Expression ofzebrafish bHLH genes ngn1 and nrd defines distinct stages of Neuraldifferentiation. Dev. Dyn. 213, 92-104.

Köster, R. W. and Fraser, S. E. (2004). Time-lapse microscopy of braindevelopment. Methods Cell Biol. 76, 207-235.

Krauss, S., Concordet, J. P. and Ingham, P. W. (1993). A functionally conservedhomolog of the Drosophila segment polarity gene hh is expressed in tissues withpolarizing activity in zebrafish embryos. Cell 75, 1431-1444.

Lekven, A. C., Buckles, G. R., Kostakis, N. and Moon, R. T. (2003). Wnt1 andwnt10b function redundantly at the zebrafish midbrain-hindbrain boundary.Dev. Biol. 254, 172-187.

Lewis, J. L., Bonner, J., Modrell, M., Ragland, J. W., Moon, R. T., Dorsky, R. I.and Raible, D. W. (2004). Reiterated Wnt signaling during zebrafish neuralcrest development. Development 131, 1299-1308.

Logan, C. Y. and Nusse, R. (2004). The Wnt signaling pathway in developmentand disease. Annu. Rev. Cell Dev. Biol. 20, 781-810.

Lun, K. and Brand, M. (1998). A series of no isthmus (noi) alleles of the zebrafishpax2.1 gene reveals multiple signaling events in development of the midbrain-hindbrain boundary. Development 125, 3049-3062.

Ma, Y. C., Song, M. R., Park, J. P., Henry Ho, H. Y., Hu, L., Kurtev, M. V., Zieg,J., Ma, Q., Pfaff, S. L. and Greenberg, M. E. (2008). Regulation of motorneuron specification by phosphorylation of neurogenin 2. Neuron 58, 65-77.

Martinez, A., Alonso, M., Castro, A., Perez, C. and Moreno, F. J. (2002). Firstnon-ATP competitive glycogen synthase kinase 3 beta (GSK-3beta) inhibitors:thiadiazolidinones (TDZD) as potential drugs for the treatment of Alzheimer’sdisease. J. Med. Chem. 45, 1292-1299.

Masai, I., Yamaguchi, M., Tonou-Fujimori, N., Komori, A. and Okamoto, H.(2005). The hedgehog-PKA pathway regulates two distinct steps of thedifferentiation of retinal ganglion cells: the cell-cycle exit of retinoblasts and theirneuronal maturation. Development 132, 1539-1553.

Megason, S. G. and McMahon, A. P. (2002). A mitogen gradient of dorsalmidline Wnts organizes growth in the CNS. Development 129, 2087-2098.

Megason, S. G. and Fraser, S. E. (2003). Digitizing life at the level of the cell:high-performance laser-scanning microscopy and image analysis for in totoimaging of development. Mech. Dev. 120, 1407-1420.

Mill, P., Mo, R., Hu, M. C., Dagnino, L., Rosenblum, N. D. and Hui, C. C.(2005). Shh controls epithelial proliferation via independent pathways thatconverge on N-Myc. Dev. Cell 9, 293-303.

Molven, A., Njolstad, P. R. and Fjose, A. (1991). Genomic structure andrestricted neural expression of the zebrafish wnt-1 (int-1) gene. EMBO J. 10,799-807.

Moore, K., Schneider, M. and Vetter, M. (2002). Posttranslational mechanismscontrol the timing of bHLH function and regulate retinal cell fate. Neuron 34,183-195.

Müller, M. v., Weizsäcker, E. and Campos-Ortega, J. A. (1996). Transcription ofa zebrafish gene of the hairy-Enhancer of split family delineates the midbrainanlage in the neural plate. Dev. Genes Evol. 206, 153-160.

Nguyen, V., Chokas, A. L., Stecca, B. and Altaba, A. R. (2005). Cooperativerequirement of the Gli proteins in neurogenesis. Development 132, 3267-3279.

Ninkovic, J., Tallafuss, A., Leucht, C., Topczewski, J., Tannhäuser, B., Solnica-Krezel, L. and Bally-Cuif, L. (2005). Inhibition of neurogenesis at the zebrafishmidbrain-hindbrain boundary by the combined and dose-dependent activity of anew hairy/E(spl) gene pair. Development 132, 75-88.

Panchision, D. M. and McKay, R. D. (2002). The control of neural stem cells bymorphogenic signals. Curr. Opin. Genet. Dev. 12, 478-487.

Panhuysen, M., Vogt-Weisenhorn, D., Blanquet, V., Brodski, C., Heinzmann,U., Beister, W. and Wurst, W. (2004). Effects of Wnt signaling on proliferationin the developing mid-hindbrain region. Mol. Cell. Neurosci. 26, 101-111.

Papan, C. and Campos-Ortega, J. A. (1997). A clonal analysis of spinal corddevelopment in the zebrafish. Dev. Genes Evol. 207, 71-81.

Price, M. A. and Kalderon, D. (2002). Proteolysis of the Hedgehog signalingeffector Cubitus interruptus requires phosphorylation by Glycogen SynthaseKinase 3 and Casein Kinase 1. Cell 108, 823-835.

Puelles, L., Amat, J. A. and Martinez-de-la-Torre, M. (1987). Segment-related,mosaic neurogenetic pattern in the forebrain and mesencephalon of early chickembryos: I. Topography of AChE-positive neuroblasts up to stage HH18. J.Comp. Neurol. 266, 247-268.

Riobo, N. A. and Manning, D. R. (2007). Pathways of signal transductionemployed by vertebrate Hedgehogs. Biochem. J. 403, 369-379.

Riobo, N. A., Haines, G. M. and Emerson, C. P., Jr (2006a). Protein kinase C-delta and mitogen-activated protein/extracellular signal-regulated kinase-1control GLI activation in hedgehog signaling. Cancer Res. 66, 839-845.

Riobo, N. A., Lu, K. and Emerson, C. P., Jr (2006b). Hedgehog signaltransduction: signal integration and cross talk in development and cancer. CellCycle 5, 1612-1615.

Ross, L., Parrett, T. and Easter, S. J. (1992). Axonogenesis and morphogenesis inthe embryonic zebrafish brain. J. Neurosci. 12, 467-482.

Ruiz, I. A. A., Palma, V. and Dahmane, N. (2002). Hedgehog-Gli signalling andthe growth of the brain. Nat. Rev. Neurosci. 3, 24-33.

Stigloher, C., Chapouton, P., Adolf, B. and Bally-Cuif, L. (2008). Identificationof neural progenitor pools by E(Spl) factors in the embryonic and adult brain.Brain Res. Bull. 75, 266-273.

Strähle, U., Fischer, N. and Blader, P. (1997). Expression and regulation of anetrin homologue in the zebrafish embryo. Mech. Dev. 62, 147-160.

Tallafuss, A. and Bally-Cuif, L. (2003). Tracing of her5 progeny in zebrafishtransgenics reveals the dynamics of midbrain-hindbrain neurogenesis andmaintenance. Development 130, 4307-4323.

Taurin, S., Sandbo, N., Qin, Y., Browning, D. and Dulin, N. O. (2006).Phosphorylation of beta-catenin by cyclic AMP-dependent protein kinase. J. Biol.Chem. 281, 9971-9976.

Thisse, B. and Thisse, C. (2005). High throughput expression analysis of ZF-Models consortium clones. ZFIN direct data submission.

Tyurina, O. V., Guner, B., Popova, E., Feng, J., Schier, A. F., Kohtz, J. D. andKarlstrom, R. O. (2005). Zebrafish Gli3 functions as both an activator and arepressor in Hedgehog signaling. Dev. Biol. 277, 537-556.

Vaage, S. (1969). Segmentation of the primitive neural tube in chick embryos.Ergebn. Anat. Entwickl. Gesch. 41, 1-88.

van de Water, S., van de Wetering, M., Joore, J., Esseling, J., Bink, R.,Clevers, H. and Zivkovic, D. (2001). Ectopic Wnt signal determines the eyelessphenotype of zebrafish masterblind mutant. Development 128, 3877-3888.

Vanderlaan, G., Tyurina, O. V., Karlstrom, R. O. and Chandrasekhar, A.(2005). Gli function is essential for motor neuron induction in zebrafish. Dev.Biol. 282, 550-570.

Varga, Z. M., Amores, A., Lewis, K. E., Yan, Y. L., Postlethwait, J. H., Eisen, J.S. and Westerfield, M. (2001). Zebrafish smoothened functions in ventralneural tube specification and axon tract formation. Development 128, 3497-3509.

Weinberg, E. S., Allende, M. L., Kelly, C. S., Abdelhamid, A., Murakami, T.,Andermann, P., Doerre, O. G., Grunwald, D. J. and Riggleman, B. (1996).Developmental regulation of zebrafish MyoD in wild-type, no tail and spadetailembryos. Development 122, 271-280.

Wilson, S., Ross, L., Parrett, T. and Easter, S. (1990). The development of asimple scaffold of axon tracts in the brain of the embryonic zebrafishBrachydanio rerio. Development 108, 121-145.

Woo, K. and Fraser, S. E. (1995). Order and coherence in the fate map of thezebrafish nervous system. Development 121, 2595-2609.

Zechner, D., Fujita, Y., Hulsken, J., Muller, T., Walther, I., Taketo, M.,Crenshaw, E. B., Birchmeier, W. and Birchmeier, C. (2003). beta-cateninsignals regulate cell growth and the balance between progenitor cell expansionand differentiation in the nervous system. Dev. Biol. 258, 406-418.

Zhang, W., Zhao, Y., Tong, C., Wang, G., Wang, B., Jia, J. and Jiang, J. (2005).Hedgehog-regulated Costal2-kinase complexes control phosphorylation andproteolytic processing of Cubitus interruptus. Dev. Cell 8, 267-278.

RESEARCH ARTICLE Development 135 (18)

DEVELO

PMENT