Developmental Biology 385 (2014) 200–210

Contents lists available at ScienceDirect

Developmental Biology

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

Cochleovestibular development is integrated with migratory neural crest cells

Lisa L. Sandell a,n, Naomi E. Butler Tjaden b,c, Amanda J. Barlow d, Paul A. Trainor b,c a University of Louisville, Department of Molecular, Cellular and Craniofacial Biology, Louisville, KY 40201, USA b Stowers Institute for Medical Research, Kansas City, MO 64110, USA c Department of Anatomy and Cell Biology, University of Kansas Medical Center, Kansas City, KS 66160, USA d Department of Surgery, University of Wisconsin, Madison, WI 53792, USA article info abstract

Article history: The cochleovestibular (CV) nerve, which connects the to the brain, is the nerve that enables the Received 28 August 2013 of and balance. The aim of this study was to document the morphological development of Received in revised form the mouse CV nerve with respect to the two embryonic cells types that produce it, specifically, the otic 1 November 2013 vesicle-derived progenitors that give rise to neurons, and the neural crest cell (NCC) progenitors that give Accepted 8 November 2013 rise to glia. Otic tissues of mouse embryos carrying NCC lineage reporter transgenes were whole mount Available online 16 November 2013 immunostained to identify neurons and NCC. Serial optical sections were collected by confocal Keywords: microscopy and were compiled to render the three dimensional (3D) structure of the developing CV Ear nerve. Spatial organization of the NCC and developing neurons suggest that neuronal and glial Otic populations of the CV nerve develop in tandem from early stages of nerve formation. NCC form a Neural crest sheath surrounding the CV ganglia and central axons. NCC are also closely associated with neurites Neuron projecting peripherally during formation of the vestibular and cochlear . Physical ablation of NCC Craniofacial Axon in chick embryos demonstrates that survival or regeneration of even a few individual NCC from ectopic Cochlea positions in the hindbrain results in central projection of axons precisely following ectopic pathways Vestibular made by regenerating NCC. & 2014 Elsevier Inc. All rights reserved.

Introduction et al., 2011)). The CV nerve originates initially during embryogen- esis as a simple unified entity that progressively develops into an During embryogenesis the vertebrate inner ear begins as a elaborate branching structure (Kopecky et al., 2012; Streeter, simple flat otic placode that invaginates to become a spherical 1906). In mammals, the mature CV nerve is subdivided into epithelial sac known as the otic vesicle (reviewed in (Bok et al., distinct nerve trunks; a superior and inferior , 2007; Chen and Streit, 2013; Groves and Fekete, 2012; Ladher which together innervate the , and a cochlear et al., 2010; Schlosser, 2010)). This vesicle grows and remodels to nerve, which innervates the auditory organ. form a complex structure of interlinked compartments known as All peripheral nerves are composed of neurons supported by the membranous labyrinth (Kopecky et al., 2012; Streeter, 1906). In glial cells, and both cell types are important for nerve function mammals these compartments consist of a spiraling cochlea, (reviewed in (Hanani, 2005)). Peripheral nerve glia include which is the auditory organ, a saccule, utricle and three semicir- Schwann cells, which support neuronal axons and neurites, and cular canals, which together form the vestibular system. These satellite cells, which surround and myelinate ganglionic neuronal structures contains patches of mechano-sensory receptors that cell bodies. As one of the cranial peripheral nerves, the CV nerve is detect sound or gravity and motion, each of which is connected to likewise composed of both neurons and glia (Rosenbluth, 1962). the brain via the cochleovestibular (CV) nerve (reviewed in These two cell types arise developmentally from distinct sources, (Appler and Goodrich, 2011; Fekete and Campero, 2007; Yang the glial cells being derived from neural crest cell (NCC) progeni- tors (D’Amico-Martel and Noden, 1983; Harrison, 1924; Yntema, 1943b), while the neurons originate almost exclusively from the Abbreviations: 3D, 3-dimensional; CV, cochleovestibular; E, embryonic day; GFP, otic placode (Breuskin et al., 2010; D’Amico-Martel and Noden, Green Fluorescent Protein; NCC, neural crest cells; P, postnatal day; R, rhombomere 1983; van Campenhout, 1935), with the exception of rare neurons n Correspondence to: University of Louisville, Department of Molecular, Cellular of the vestibular ganglia. The exclusive placodal derivation of CV and Craniofacial Biology, School of Dentistry, Room 341, Louisville, KY 40201-2042, USA. neurons makes the nerve somewhat unique among cranial sen- E-mail address: [email protected] (L.L. Sandell). sory nerves, most of which contain neurons derived from two

0012-1606/$ - see front matter & 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ydbio.2013.11.009 L.L. Sandell et al. / Developmental Biology 385 (2014) 200–210 201 sources, the sensory placodes and NCC (D’Amico-Martel and the , but delamination of neuroblasts and formation of Noden, 1983). While experiments in amphibian, avian, and mam- the is not impaired (Coppola et al., 2010). malian systems have each indicated CV neurons derive almost Despite the many important insights regarding cranial sensory exclusively from placode progenitors, a recent study of transgenic nerve development that have been made, owing to the dynamic lineage reporter mice has posed a contradictory scenario wherein remodeling and morphological complexity of the embryonic inner NCC or a related population of migratory neuroepithelial cells give ear, many details of CV nerve formation remain obscure. To rise to a significant fraction of CV neurons and also to cells of the understand development of neuronal and glial progenitors in the otic epithelium (Freyer et al., 2011). embryonic CV nerve, we examined immunostained inner ears of The neuronal cells of the CV nerve develop by delaminating mouse transgenic NCC lineage reporter embryos and chick from the otic vesicle, proliferating and aggregating to form the CV embryos by confocal microscopy, compiling multiple optical sec- ganglion (Altman and Bayer, 1982; Carney and Silver, 1983). As tions into virtual 3D renderings of developing CV nerves. By this development proceeds CV neurons project central axons to the method we gain insight into the coordinated development of otic hindbrain and project peripheral neurites to sensory targets vesicle-derived neurons and NCC-derived glial cells of the developing within the otic epithelium (reviewed in (Fritzsch, CV nerve. 2003)). The glial cells of the CV nerve derive from NCC that emigrate from the hindbrain at the level of rhombomere 4 (R4) (D’Amico-Martel and Noden, 1983). Based on dye labeling analysis, Materials and methods CV neurons extend peripheral neurites to the developing vestib- ular sensory epithelium as early as embryonic day 11.5 (E11.5) and Mice central axons to the hindbrain as early as E12.5 (Fritzsch, 2003; Matei et al., 2005). Maturation of CV neurons, as measured by final Mouse strains utilized in this study included the following: “ ” fi cell division, indicates vestibular and cochlear neurons mature at Z/EG ,ofcial name, Tg(CAG-Bgeo/GFP)21Lbe/J, Jax stock # embryonic day E11.5 and E13.5, respectively (Matei et al., 2005; 003920; “ ” fi Ruben, 1967), earlier than the associated Schwann cell progenitors, R26R ,ofcial name, FVB.129S4(B6)-Gt(ROSA)26Sortm1Sor/J, which continue dividing up to the time of birth (Ruben, 1967). Jax stock # 009427, “ ” fi A growing body of evidence suggests that interactions between Wnt1Cre ,ofcial name, Tg(Wnt1-cre)11Rth Tg(Wnt1-GAL4) glial and neuronal progenitors may be important for development 11Rth/J, Jax stock #003829, “ ” of the CV nerve. In mouse, disturbance of ERBB2, a receptor that 6.5Pax3Cre , novel stable transgenic Cre driver line in FVB/NJ mediates neuron–glia interactions via the ligand Neuregulin 1 background. Transgene contains 6.5 kb NotI-HindIII fragment of (reviewed in (Corfas et al., 2004)) compromises development of genomic regulatory DNA 5' of mouse Pax3 start site. the CV nerve. Complete loss of ErbB2 severely disrupts formation of (Lee et al., 1995) but is lethal at E10.5 owing to Images to emphasize embryo morphology cardiac defect. ErbB2 mutants rescued to later stages exhibit abnormalities in development of the CV nerve, including altered In order to enhance visualization of morphology of embryos migration of neuronal cell bodies, abnormal targeting of peripheral stained for β-galactosidase activity, color images of β-galactosidase neurites and reduced neuron number (Morris et al., 2006). stained embryos were overlain onto greyscale images of same Whereas the phenotype of ErbB2 mutants indicates that migration embryos stained for DAPI to label all cell nuclei. Imaging of whole and targeting of CV neurons depends upon signaling interactions embryos stained with DAPI or other nuclear fluorescent dye with NCC glial progenitors, the relatively normal development of reveals details of embryo morphology not visible with white light, CV nerves in embryos lacking Sox10, a transcription factor impor- as previously described (Sandell et al., 2012). Overlay of color tant for development of peripheral glial NCC, suggests otherwise images of β-galactosidase stain embryos with greyscale image of (Breuskin et al., 2010). The reason for the differing effect of ErbB2 embryo morphology yields improved visualization of β- mutation versus Sox10 mutation is not yet clear, but, because Sox10 galactosidase signal relative to embryonic structures. mutation results in loss of NCC glial progenitors after E10.5, the normal growth and guidance of CV neurons in those mutants may Chick NCC ablation indicate that critical interactions occur earlier. That interactions between NCC glial progenitors and neurons Fertile chicken eggs (Gallus gallus domesticus) were prelimina- are important for development of cranial nerves is supported also rily incubated at 37 1C to allow embryos to develop to HH stage by chick and zebrafish studies in which NCC are eliminated or 8to9þ (6–9 somites). Eggs were then windowed and R4 NCC signaling interactions between NCC and neurons is blocked. were ablated by microsurgical removal of the neural tube from R3 Molecularly blocking Semaphorin/Neuropilin signaling in chick through R5. Operated embryos were re-sealed and allowed to disrupts NCC migratory pathways and impairs the inward move- continue development for an additional 24 or 48 h. ment of epibranchial placodal neurons (Osborne et al., 2005). In some studies ablation of NCC migration by physical or molecular Whole mount immunostain methods in chick results in reduced numbers of neuroblasts migrating from epibranchial ganglia and abnormal projection of For E10.5 mouse and chick, specimens were immunostained as central axons (Begbie and Graham, 2001; Freter et al., 2013; whole embryo. For E11.5 mouse inner ear, embryo head specimens Yntema, 1944), although other studies reported NCC removal did were bisected sagittally prior to immunostain to permit penetra- not disrupt formation of ganglia (Begbie et al., 1999). In zebrafish tion of antibody and wash solutions. For E12.5 specimens, inner also elimination of specific sub-populations of cranial NCC disrupts ears were isolated by dissection. Embryo specimens were fixed in formation of the epibranchial nerves (Culbertson et al., 2011). 4% paraformaldehyde (formaldehyde) phosphate buffered saline Visualization of early developmental association between neu- (PBS), overnight at 4 1C, then washed in PBS and transferred ronal and glial progenitors of the facial ganglion in mouse and (through a graded series) into 100% methanol. Tissues and chick indicates that NCC form a corridor surrounding placodal embryos to be immunostained were permeabilized in Dent’s neuroblasts (Freter et al., 2013). In mouse, genetic ablation of NCC bleach (MeOH:DMSO:30%H2O2, 4:1:1) for 2 h at room tempera- causes some abnormalities in growth of peripheral projections of ture, then washed with 100% methanol and transferred (through a 202 L.L. Sandell et al. / Developmental Biology 385 (2014) 200–210 graded series) into PBS. Blocking and antibody hybridization were grease. Confocal images were captured on an upright Zeiss LSM510 performed in a buffer of 0.1 M Tris pH7.5/0.15 M NaCl. Non-specific Pascal equipped with a 405 nm laser. For each specimen, a z-stack antibody binding was blocked by 2 h incubation with gentle of images was collected. rocking in buffer with 0.5% Perkin Elmer Block Reagent (FP1020), or alternatively 3% Bovine Serum Albumen. Primary antibody hybridization was performed in same blocking buffer overnight 3D image rendering at 4 1C. Following primary antibody hybridization specimens were washed 5 1 h in PBS at room temperature. Secondary antibody Confocal z-stacks were rendered with the 3D image processing hybridization was performed in blocking buffer overnight at 4 1C. software IMARIS (Bitplane AG). For optically cropped images, Secondary antibody was removed by 3 20 min washes in PBS at individual image planes were manually cropped to omit non-ear room temperature. Specimens were stained with DAPI dilactate, tissues and to retain only the area of interest. The z-stack of 10 μg/ml in PBS, 10 min or longer at room temperature. Specimens cropped images was then rendered to reveal the 3D structure of were then transferred (through a graded series) into 100% the developing otic vesicle and CV nerve. For visualization of methanol. ganglion shape fluorescence signal was rendered as solid surfaces to generate shadows to enhance visualization of ganglion Antibodies morphology.

ISL1 antibody, 1/50, (39.4D5, Developmental Studies Hybridoma Bank); Results βIII neuronal tubulin antibody (TUJ1), 1/500, (MMS-435P or PRB-435P, Covance); Imaging NCC in ear development GFP antibody, 1/500, (A6455 Invitrogen/Molecular Probes); HNK-1, 1/5, (supernatant from cultured HNK-1 cells, American In order to assess the distribution of NCC during morphogen- Type Culture Collection); esis of the CV nerve we examined embryos in which NCC were Alexa Fluor 488, 1/300, (A21206 or A21042, Invitrogen/Mole- marked by Cre recombinase activation of lacZ or GFP using the cular Probes);and R26R (Soriano, 1999) or Z/EG (Novak et al., 2000) reporter strains. Alexa Fluor 546, 1/300, (A11030, A10040, 10040, Invitrogen/ Wnt1 regulatory sequences drive expression in dorsal neural tube Molecular Probes). cells (Echelard et al., 1994), and the Wnt1Cre transgene, in combination with the R26R reporter irreversibly marks all NCC Whole mount confocal imaging of immunostained embryos and ear and derivatives by expression of lacZ (Chai et al., 2000). Consistent tissues with many previous studies characterizing Wnt1Cre;R26R embryos, using this combination of Cre driver and reporter we Tissue or embryos specimens that had been immunostained observe extensive β-galactosidase activity in NCC and derivatives and dehydrated as described above were prepared for clearing by throughout the pharyngeal arches (Fig. 1A). Similarly, the Wnt1Cre equilibrating through a second wash of 100% methanol to ensure can be combined with the Z/EG reporter, which allows NCC to be complete dehydration. Specimens were then cleared with BABB visualized in conjunction with different cell types and molecular (Benzyl Alcohol:Benzyl Benzoate, 1:2). Glass imaging wells were markers by double immunostaining for GFP and other molecules prepared by coating a Teflon O-ring with vacuum grease and such as βIII neuronal tubulin, which labels neuronal axons and affixing the ring to a glass depression slide. BABB cleared speci- neurites (Fig. 1C). The Wnt1Cre driver works well to visualize all mens were placed into prepared wells with the aid of UV light NCC, however, because NCC contribute so extensively to craniofa- under a fluorescent stereo microscope to visualize the DAPI signal. cial development, with this pan-NCC Cre driver it can be difficult Once specimens were place, the well was filled with BABB and a to visualize morphogenesis of specific tissues and structures in the coverslip was adhered to the top of the O-ring with vacuum head and face.

Fig. 1. Mouse Cre driver;reporter combinations used to visualize NCC in CV nerve development. (A and C) Embryos carrying Wnt1Cre;R26R (A) or Wnt1Cre;Z/EG (C) have all NCC and derivatives marked by expression of lacZ or GFP, respectively. Single arrowhead indicates second pharyngeal arch with all NCC labeled by Wnt1Cre-mediated recombination. (B and D) Within the cranial region, embryos carrying 6.5Pax3Cre;R26R (B) or 6.5Pax3Cre;Z/EG (D) preferentially label NCC emanating from R4 with β- galactosidase activity or GFP, respectively. The R4 NCC labeling is mosaic. Double arrowhead indicates second pharyngeal arch with mosaic labeling of R4 NCC. Hypaxial muscle progenitors are also labeled by this Cre driver (not shown). (A and B) Staining for β-galactosidase activity reveals NCC and derivatives labeled by Cre mediated recombination. Color image of β-galactosidase signal is overlain onto grayscale image of fluorescent DAPI signal to emphasize embryo morphology. (C, D) Double immunostain for GFP (green)and neuronal βIII tubulin (red) reveals NCC and neurons. DAPI labeling of all cell nuclei reveals pharyngeal arch morphology. Yellow box in (A) indicates approximate region of images shown in panels (C) and (D). Scale bar in (C) and (D)¼150 μm. L.L. Sandell et al. / Developmental Biology 385 (2014) 200–210 203

In order to examine the contribution of NCC within the context are distinct (Fig. 2C). Examination of an individual optical slices of CV nerve formation we generated a transgenic Cre driver mouse reveals that, for the CV ganglion, NCC are observed primarily on line that, within the cranial region, preferentially marks NCC and the exterior surface of the aggregated neuronal cell bodies derivatives relevant for development of the ear (Fig. 1B). The new (Fig. 2D). Rare NCC are present intercalated between neuronal cell transgene, herein 6.5Pax3Cre, contains a 6.5 kb region of geno- bodies of the CV ganglion adjacent to areas of the otic vesicle that mic DNA 5′ of the Pax3 start. The region has been previously express low level of ISL1, indicative of regions of delaminating shown to contain within it two distinct enhancer elements, one neuroblasts (Fig. 2D). driving expression in a subset of NCC (Milewski et al., 2004), and In order to gain understanding of the morphology of the one driving expression in hypaxial muscle progenitors (Brown developing CV ganglion, we utilized 3D image processing software et al., 2005). Similar to the previously described transgenes, our Imaris to crop individual image slices to optically isolate the CV new 6.5Pax3Cre transgene marks a subset of NCC and hypaxial ganglion, the facial ganglion, and the otic vesicle, and to render muscle progenitors. Relevant for this study, the 6.5Pax3Cre trans- fluorescence signal as a solid surface with shadows. Compilation of gene preferentially marks NCC emigrating from the R4 region that such images aids in visualizing the 3D organization of the ear and migrate into the second pharyngeal arch, but labels only rare associated structures. Optical isolation and surface rendering of cranial NCC cells from anterior or posterior axial positions in the the CV ganglion, facial ganglion and otic vesicle of an E10.5 head. Staining 6.5Pax3Cre;R26R embryos for β-galactosidase activ- Wnt1Cre;Z/EG embryo immunostained for GFP (NCC) and ISL1 ity reveals mosaic labeling of NCC in the stream that fills the (neurons) allows the visualization of the two ganglia relative to second pharyngeal arch (Fig. 1B). The pattern of mosaic NCC the second arch NCC stream and the otic vesicle (Fig. 3A and B). labeling within the second arch stream, and the absence of The CV ganglion appears as a saddle-shaped structure that anterior and posterior cranial NCC, allows visualization of NCC “straddles” the anterior portion of the otic vesicle, with the facial contributing to development of the ear and CV nerve indepen- ganglion nested within the anterior portion of the saddle. Two dently of other cranial NCC derivatives (Fig. 1D). small processes project from the ventral edge of the CV ganglion extending medially and laterally over the anterior ventral region of Second arch NCC stream envelops embryonic CV ganglion the otic vesicle (Fig. 3B–D').

In order to assess the spatial relationship between NCC and otic Peripheral neurite extension of the CV ganglion vesicle-derived neuroblasts of the developing CV ganglion, we examined whole mount Wnt1Cre:Z/EG embryos immunostained In order to understand the relationship between NCC and the for GFP (identifying NCC) and ISL1 (Islet 1 transcription factor centrally projecting axons and peripherally projecting neurites of expressed in neurons, identifying neuronal cell bodies of the CV the developing CV neurons, we performed whole mount immu- ganglion (Begbie et al., 2002; Ikeya et al., 1997)). We imaged nostaining of Wnt1Cre;Z/EG and 6.5Pax3Cre;Z/EG embryos, staining immunostained embryos by confocal microscopy, collecting com- for GFP (NCC) and βIII neuronal tubulin (neuronal projections). At plete z-stacks of images through the region of interest. The z-stack E10.5, immunostaining for βIII neuronal tubulin reveals that images were examined as individual slices or were compiled and peripheral projections of the CV nerve can be detected and the rendered to convey the 3D morphology of the developing CV early fibers are organized as a lattice with longitudinal fibers and ganglion. crossing fibers (Fig. 4A and B). Immunostaining for GFP in We imaged embryos at E10.5 in the region of the otic vesicle, Wnt1Cre:Z/EG embryos, in which all NCC express GFP, reveals that including the hindbrain at R4 and the proximal portion of the first NCC are integrated throughout the entire CV nerve and are and second pharyngeal arches (Fig. 2A). ISL1 staining of this region intercalated between the lattice-like neuronal projections at E10.5 reveals the facial nerve ganglion and CV ganglion are (Fig. 4B). NCC are positioned within the junctions of the develop- nested directly anterior to the otic vesicle (Fig. 2B and C). The path ing neuronal lattice and appear at the terminal extent of the of the R4 NCC stream appears to emerge from the neural tube, projecting neurite branches with neuronal filopodia extending a travels directly around the nested ganglia, and continues into the short way beyond the most peripheral NCC cells. second pharyngeal arch (Fig. 2B and C). In the region of the nested By optically isolating the otic vesicle and developing CV nerve ganglia the spatial organization of the R4 NCC stream is tightly at sequential stages of embryogenesis it is possible to understand correlated with the ganglia with no significant NCC presence more the 3D morphogenesis of the developing CV nerve (Fig. 5A–D). than one or two cell diameters distant from the ganglia (Fig. 2B Immunostaining a Wnt1Cre:Z/EG embryo for GFP and βIII neuronal and C). Although nested, the facial ganglion and the CV ganglion tubulin reveals that at E10.5 the peripheral neurites of the

Fig. 2. NCC stream co-localizes with nested facial and CV ganglia in E10.5 mouse embryos. (A) Whole embryo stained with DAPI and imaged on fluorescent stereo microscope. Yellow box corresponds to region of imaged in (2B and 3A). (B) Wnt1Cre;Z/EG embryo double immunostained with GFP for NCC and ISL1 for neuronal cell bodies. DAPI stain indicates all cell nuclei. Nested facial and CV ganglia are visible anterior to the otic vesicle. Stream of NCC and derivatives emanates from R4, surrounds the nested ganglia, and fills the second pharyngeal arch. (C) NCC stream emanating from R4 envelopes nested facial and CV ganglion neurons as a sleeve. (D) Rare NCC are visible within the aggregated neuronal cell bodies of the CV ganglion, primarily near regions of the otic vesicle with low level ISL1 signal indicative of delaminating neuroblasts. Dotted lines in (C, D) indicate epithelium of otic vesicle. Arrowhead, region of otic vesicle of delaminating neuroblasts indicated by low level ISL1 signal; dotted lines, otic vesicle epithelium; CVG, cochleovestibular ganglion; FG, facial ganglion; HB, hindbrain; ovl, otic vesicle lumen; PA2, pharyngeal arch 2. 204 L.L. Sandell et al. / Developmental Biology 385 (2014) 200–210

Fig. 3. Early CV ganglion resembles a saddle straddling ventral otic vesicle. Solid surface rendering of ganglion fluorescence signal and optical cropping of region of interest allows visualization of CV ganglion morphology of E10.5 embryo. ISL1 fluorescence (neuronal cell bodies) is rendered as solid surface to allow visualization of E10.5 mouse CV ganglion. (A) Otic region as indicated by yellow box in (Fig. 2A). Solid surface rendering of ISL1 signal indicates nested facial and CV ganglion straddle ventral otic vesicle at proximal region of second pharyngeal arch. Uncropped GFP immunostain of Wnt1Cre;Z/EG signal reveals NCC and derivatives filling the first and second pharyngeal arches. DAPI signal is cropped to define position of otic vesicle epithelium. (B) Lateral view of surface-rendered ganglion surfaces reveal two prongs extending across ventral otic vesicle. (C–C') Ventral view of rendered ganglia cell bodies reveals nested organization of facial and CV ganglia and ventral prongs of CV ganglion that extend medially and laterally. (D–D') Dorsal view of surface rendered ganglia reveals prongs of CV ganglion. Scale bars¼100 μm with respect to individual image planes, CVG, CV ganglion; FG, facial ganglion; HB, hindbrain; lp, lateral projection; mp, medial projection; OV, otic vesicle; PA1, pharyngeal arch 1; PA2, pharyngeal arch 2.

Fig. 4. CV Axons and neurites form a lattice with NCC intercalated in interstitial spaces at early stages of CV nerve development. (A) Whole mount E10.5 mouse head stained with βIII neuronal tubulin to visualize central axons and peripheral neurites of cranial nerves. Yellow box indicates region imaged in panel B. (B) CV nerve of Wnt1Cre;Z/EG embryo immunostained for βIII neuronal tubulin (neurons) and GFP (NCC) indicates peripheral vestibular neurites are organized as a lattice with NCC intercalated between the spaces of the lattice and at distal-most extent of neurite projections. cvn, terminal peripheral neurites of CV nerve; F/CV, facial nerve/CV nerve; GP, glossopharyngeal nerve, TG, ; V, vagal nerve. developing CV nerve project as two short branches, the future Optical isolation of otic tissues from 6.5Pax3Cre:Z/EG embryos inferior vestibular nerve extending medially, and the future super- at E12.5 reveals further morphogenesis of the CV nerve. At this ior vestibular nerve extending laterally over the ventral anterior stage the extending peripheral neurites of the early portion of the otic vesicle (Fig. 5A; Supplementary material Movie are organized as a fan shape that projects ventrally, between the 1). The two branches of peripheral neurite projections parallel the medially situated inferior vestibular nerve and laterally extending two projections of ganglion cell bodies observed with ISL1 staining superior vestibular nerve (Fig. 5C and E, Supplementary material (Fig. 3A–D′). Between the peripheral projecting fibers of the future Movie 2). Mosaic labeling of R4 NCC in the 6.5Pax3Cre:Z/EG inferior vestibular nerve and superior vestibular nerve, the future embryos reveals NCC are present in all regions of the developing cochlear nerve is visible as a thicker blunt mass with few or no fine CV nerve and are densely associated with the terminal edge of the neurites projections at this stage (Fig. 5A). extending neurites of the developing cochlear nerve (Fig. 5C and Optical isolation of similarly labeled E11.5 6.5Pax3Cre:Z/EG D). Fine neurite filopodia are visible beyond a dense wave front embryos, in which there is mosaic expression of GFP in R4 NCC, of NCC. reveals that the future inferior vestibular nerve elongates posteriorly In order to visualize the distribution of NCC in the CV nerve at medial to the otic vesicle between E10.5 and E11.5, while the future later stages of development we examined the inner ear of superior vestibular nerve extends dorso-laterally over the anterior 6.5Pax3Cre;R26R at postnatal day 0 (P0) (Fig. 5F). In these samples otic vesicle, the superior and inferior vestibular nerves appearing to β-galactosidase activity reveals that NCC are distributed through- embrace the otic vesicle at the narrow region between the pre- out the extent of CV nerve, including all portions of the spiral sumptive semi-circular canals and the future cochlea (Fig. 5B). At ganglion and cochlear nerve, the inferior vestibular nerve, and the this stage the future has thickened, extending superior vestibular nerve. No staining is observed within the otic ventrally and slightly anterior with a few fine neurite projections epithelium with this transgenic lineage reporter combination. as the first evidence of cochlear nerve fibers. Although the The temporal sequence of images presented here (Fig. 4B, 6.5Pax3Cre:Z/EG marks only a mosaic subset of R4 NCC, GFP labeled Fig. 5A–F) reveal the morphogenesis of CV nerve neurons and NCC are observed in all regions of the developing CV nerve. associated NCC-derived glia in relation to the remodeling of the L.L. Sandell et al. / Developmental Biology 385 (2014) 200–210 205

Fig. 5. Progression of CV nerve neurite outgrowth integrated with migratory NCC. (A–D) Lateral view of CV nerve and otic vesicle morphology visualized in isolation by optical cropping of individual image planes in confocal image stacks. Whole embryos or isolated inner ears were immunostained for GFP to identify NCC, and with βIII tubulin to identify central axons and peripheral neurites. DAPI staining reveals all cell nuclei within optically cropped otic vesicle. (A) whole CV nerve and otic vesicle of E10.5 Wnt1CRE;Z/EG, (B) whole CV nerve and otic vesicle of E11.5 6.5Pax3Cre;ZEG, (C) whole CV nerve and developing inner ear of E12.5 6.5Pax3Cre;Z/EG, (D) developing cochlea and fan of neurites from early spiral nerve of E12.5 6.5Pax3Cre;Z/EG with accompanying population of NCC at region of extending peripheral neurites. (E) Schematic representation of E12.5 CV nerve and otic vesicle as in panel (C) to aid in visualization of 3D organization. (F) β-galactosidase staining of inner ear of neonatal 6.5Pax3Cre;R26R pup reveals NCC evenly distributed throughout the CV nerve. Scale bars¼100 μm with respect to individual image planes. aa, anterior ampulla; aco, apical cochlea; ax, axonal projection to hindbrain; bco, basal cochlea; cn, cochlear nerve; cvg, cochleovestibular ganglion; dov, dorsal otic vesicle; hb, direction of hindbrain; ivn, inferior vestibular nerve; la, lateral ampulla; pa, posterior ampulla; pro, projecting cochlear neurites; sac, saccule; sg, spiral ganglion; svn, superior vestibular nerve; ut, utricle; vg, vestibular ganglia; vov, ventral otic vesicle.

Movie 1. E10.5 CV nerve visualized in isolation by optical cropping of individual Movie 2. E12.5 CV nerve visualized in isolation by optical cropping of individual image planes in confocal image stacks. Whole mount Wnt1CRE;Z/EG E10.5 embryo image planes in confocal image stacks. Whole mount 6.5Pax3Cre;Z/EG E12.5 inner immunostained for GFP (green) to identify NCC, and with βIII tubulin (red) to ear immunostained for GFP (green) to identify NCC, and with βIII tubulin (red) to identify central axons and peripheral neurites. DAPI staining (blue) reveals all cell identify central axons and peripheral neurites. DAPI staining (blue) reveals all cell nuclei. Optical cropping was performed on neuronal and NCC signal to isolate CV nuclei. Optical cropping was performed to isolate CV nerve and developing otic nerve.. A video clip is available online. Supplementary material related to this epithelium. A video clip is available online. Supplementary material related to this article can be found online at http://dx.doi.org/10.1016/j.ydbio.2013.11.009. article can be found online at http://dx.doi.org/10.1016/j.ydbio.2013.11.009. otic epithelium. We observe that peripheral neurite extension fibers encircle the otic vesicle at the junction between the begins as early as E10.5. At that stage the fibers of the future presumptive vestibular structures and the future cochlea. The inferior vestibular nerve, which will innervate the posterior crista cochlear nerve, emerging between the vestibular nerve branches, ampulla, extend medially, while the future superior vestibular begins to project peripheral neurites between E11.5 and E12.5, nerve, which will innervate the anterior and lateral cristae concomitant with the morphogenesis of the cochlea. Migratory ampullae and the utricle, extends laterally. Each branch of fibers NCC accompany projecting peripheral neurites at all stages and is associated with intercalated NCC. As development progresses, associate with all parts of the developed CV nerve. the inferior vestibular nerve fibers elongate posteriorly along the medial aspect of the otic vesicle as the superior vestibular nerve NCC guide formation of the CV nerve fibers extend dorso-laterally over the anterior aspect of the otic vesicle. As development progresses and the otic vesicle is remo- The visible association between NCC and CV neurons, and the deled into the membranous labyrinth, the two branches of nerve demonstrated role of NCC in central axon guidance of the facial 206 L.L. Sandell et al. / Developmental Biology 385 (2014) 200–210 nerve, prompted us to examine directly whether formation of the (Fig. 6A–C). In HH17 stage chick embryos, the central axons of the CV nerve is likewise dependent on the presence of migratory NCC. combined CV nerve and facial nerve appears as a robust trunk To that end we examined chick embryos in which otic NCC were extending from the combined ganglia medially to the hindbrain eliminated by physical ablation. Chick embryos in which the (Fig. 6B and C), consistent with previous analysis demonstrating presumptive R4 region of the hindbrain was removed at HH stage central axon projection of the VIIIth nerve beginning at day 2.5 8to9þ (26–30 h, 6–9 somites) were allowed to grow for 24 or (Fritzsch et al., 1993). At the anterior otic vesicle the peripheral 48 h following ablation. The embryos were immunostained to projection of the facial nerve bifurcates and extends ventrally from identify NCC, (using HNK-1 antibody) and neurons (anti-βIII the CV nerve, which can be detected as short neurites contacting neuronal tubulin antibody) to assess the development of the to the surface of the otic vesicle (Fig. 6B and C). CV nerve. Examining image planes through the CV nerve in non-ablated Confocal images of whole mount immunostained embryos control embryos at HH17 reveals a sheath of NCC enveloping the reveal that the CV nerve of a developing chick embryo is similar CV neurons similar to that observed in mouse, consistent with to its analogous structure in mouse, being positioned anterior- recent similar studies (Freter et al., 2013). In non-ablated control ventral to the otic vesicle, but appearing somewhat more embryos NCC surround the CV nerve forming a corridor from the tightly unified with facial nerve than its mammalian counterpart anterior otic vesicle to the hindbrain (Fig. 6D–F). In contrast, in

Fig. 6. Elimination of NCC disrupts outgrowth of central axons. Test of chick CV nerve development following ablation of R4 NCC. Neurons and NCC visualized by immunostain for βIII neuronal tubulin and HNK-1. (A–F) Normal HH stage 17 embryo with NCC intact not ablated. (A) Whole embryo visualized by DAPI labeling of all nuclei. Box indicates region imaged in other panels. (B) Otic region of normal non-ablated HH17 embryo stained for βIII tubulin and HNK-1 showing facial/CV nerves and otic vesicle. (C) Schematic representation of region imaged in panel (B) to aid in visualization of 3D organization. The CV nerve runs lateral–medial and slightly anterior-dorsal, connecting the laterally positioned otic vesicle with the medially positioned hindbrain at R4. (D–L) Similar views of CV nerve and otic vesicle as shown in (C) but optically cropped to eliminate facial nerve. (D–F) Non-ablated CV nerve of HH stage 17 control embryo. Trunk of central axons connects medial–anterior-dorsal to hindbrain co- localizing with NCC stream. (G–I) Otic region of embryo in which R4 NCC were ablated at HH stage 8 to 9þ followed by 24 h of incubation and growth to HH17. (G) NCC stream emanating from R4 is absent (hollow arrowhead). (H) Neuronal cells are present anterior to the otic vesicle as a rudimentary ganglion (gn), but lack central axons connecting to hindbrain (hollow arrowhead). (J–L) Following ablation of R4 NCC at HH stage 8 to 9þ and subsequent culture for 48 h to HH20 NCC regenerated from ectopic positions have migrated to otic region. In such embryos CV nerve forms central axons exactly matching pathway of regenerated ectopic NCC. (J) Rare NCC have migrated to the CV nerve from contralateral side (white asterisks), or from posterior region of the hindbrain (tailed arrows). (K) Central axons form (white asterisks and tailed arrows). (K) The central axons are co-localized with precisely along the migration pathways of regenerated ectopic NCC. Scale bars, 50 μm; hollow arrowhead, empty region normally occupied by migratory NCC from R4 and central axons from CV nerve; white asterisks, pathway of NCC regenerated from contralateral side; tailed arrows, pathway of NCC regenerated from region posterior to ablated section of hindbrain; A, anterior; ca, central axons; fn, facial nerve; gn, ganglion with no projections; hb, hindbrain; M, medial; OV otic vesicle; V, ventral. L.L. Sandell et al. / Developmental Biology 385 (2014) 200–210 207 embryos where R4 NCC had been removed at HH stage 8 to 9þ central axons, NCC interact intimately with peripheral projections followed by 24 h of growth to HH17 results in absence of the NCC of developing neurons. We also observe the early central and corridor between the CV ganglion and the hindbrain (Fig. 6G,I). peripheral projections are organized a lattice with NCC interca- When NCC are eliminated in this manner CV neurons are observed lated between the rungs of the lattice and at the terminal edges of as an isolated ganglion without central extensions to the hindbrain extending neurites. As the CV nerve grows and develops, NCC (Fig. 6H and I). continue their intimate localization with all parts of the nerve, Owing to developmental plasticity of NCC, ablation of R4 at HH migrating as a wave front in association with extending peripheral stage 8 to 9þ and subsequent growth for 48 h often resulted in neurites. regeneration of NCC from anterior or posterior axial levels (Fig. 6J–L). The early and extensive association of NCC with CV neurons Where such regeneration occurred CV nerve axons formed exactly indicates that CV nerve development should be considered as a and exclusively along the regenerated NCC migration pathways coordinated event between neuronal and glial progenitors. That (Fig. 6J–L). Even when the regenerated NCC were very few in development of cranial nerves requires coordination between number, central CV nerve axon projections form, localized precisely neuronal progenitors and NCC is supported by observations of within the path traversed by the regenerated NCC. In these cases the nerve abnormalities resulting from disruptions of NCC migration CV nerve axon destination corresponded to ectopic anterior or or disruptions of signaling between NCC and neurons. Specifically, posterior positions, or from the contralateral side of the neural tube, loss of Neuropilin-Semaphorin signaling results in abnormal from which the regenerated NCC arose. The ectopic axon pathways migration of cranial NCC, and causes corresponding abnormal corresponding to the regenerated NCC pathway provides further migration and positioning of cranial sensory neurons and nerve evidence that development of the CV nerve is dependent on fibers (Schwarz et al., 2008). Similarly, disrupting signaling interactions with migratory NCC from the hindbrain. between neuronal and glial progenitors also interferes with development of cranial nerves including the CV nerve. ERBB2, a receptor that mediates neuron–glia interactions via the ligand Discussion Neuregulin 1 (reviewed in (Corfas et al., 2004)) is required for proper development of the CV nerve. Loss of ErbB2 results in Development of CV neurons and associated NCC altered migration of CV neuronal cell bodies, abnormal targeting of CV peripheral neurites and reduced number of CV neurons (Morris This report describes our analysis of the 3D morphological et al., 2006). While these phenotypes indicate that NCC-derived development of the CV nerve with respect to neuronal progenitors glial progenitors influence guidance of peripheral neurites, there and glial progenitors, derived, respectively, from the otic vesicle are clearly other tissues and signaling interactions that are also and NCC. In terms of morphology, our observations of mammalian critical for formation of inner ear neuron migration and growth CV nerve formation are largely consistent with previous descrip- and targeting of central axons and peripheral neurites. For exam- tions of developing CV nerve shape (Kopecky et al., 2012; Streeter, ple, signaling by neurotrophins, which are expressed in the 1906; Yang et al., 2011), with added insights revealed by the use of sensory epithelia are also important for survival and guidance of molecular markers and lineage trace reporters. Notably, we find cochlear and vestibular peripheral neurites (Tessarollo et al., that neuronal progenitors and glial progenitors interact exten- 2004). sively at all stages of nerve development. The close association of NCC with CV neuroblasts is relevant for We detect central axon and peripheral neurite extensions interpreting previous studies of CV nerve formation and axon slightly earlier than has been reported by dye labeling studies, guidance. Much of our knowledge of CV nerve axon guidance is observing peripheral neurites across the ventral-anterior otic built on in vitro culture of chick CV ganglia (Ard et al., 1985; vesicle as early as E10.5. The initial peripheral neurites extend as Bianchi and Cohan, 1991; Hemond and Morest, 1992). Based on the two branches of fibers, the future inferior vestibular nerve pro- close association of NCC with the developing ganglia observed jecting medially, and the future superior vestibular nerve project- here, it is clear that in vitro explants of isolated CV ganglia would ing laterally. The future cochlear nerve begins to extend peripheral contain numerous associated NCC. neurites at E11.5, a day later than the vestibular nerves. As In addition to interactions between NCC and neurons in development progresses the peripheral neurites of the cochlear development of the CV nerve, the proximity of the R4 NCC stream nerve form a fan that spreads open concomitant with the exten- to the otic vesicle may be important for axial patterning of the sion of the cochlea. inner ear. Inner ear patterning has been elucidated in large part by Using the Wnt1-Cre and 6.5Pax3Cre drivers to lineage label NCC amphibian or chick experiments involving rotational transplants we observe that the NCC stream originating from R4 covers the of otic placode, otic vesicle or hindbrain (Bok et al., 2005; Harrison, nested facial and CV ganglia as a close-fitting sleeve and extends 1936; Wu et al., 1998). Rotational transplants of chick otic cup into the second pharyngeal arch. Association of NCC with the tissue at stages HH 10–12 (11 somites – 16 somites) indicate that developing CV ganglion has been described previously by 2D the anterior–posterior axis of the developing otic tissue is dictated section analysis (Breuskin et al., 2010; Davies, 2007), but the 3D by the host environment at that stage, and that the host influence rendering of multiple 2D image planes presented here allows the is not mediated by the neural tube (Bok et al., 2005). Because the sleeve-like association of the R4 NCC stream to be appreciated R4 NCC stream would be present at this stage, having emerged much more clearly. from the neural tube from HH stage 9 through 11 (7 somites – 13 The association of NCC as a sheath surrounding the combined somites) (Tosney, 1982), the presence of host NCC could account facial and CV ganglia is consistent with results of a similar recent for the anterior–posterior patterning influence on the transplanted study of chick and mice demonstrating that migratory NCC form tissue. A possible role for R4 NCC in patterning the inner ear is also corridors within which cranial sensory neurons migrate (Freter supported by genetic evidence. Hoxa2 is known to be important et al., 2013). In that study the NCC corridors surrounding the for anterior posterior patterning for both the hindbrain and second neuron progenitors of the combined facial and CV ganglia were arch NCC (Gendron-Maguire et al., 1993; Rijli et al., 1993). Muta- demonstrated to be important for facilitating migration of the tion of HOXA2 in humans can impact development of the inner ear neuroblasts and for accurate growth of central axons to the (Alasti et al., 2008), but, again, the effect is unlikely to be mediated hindbrain. Our observations reported here indicate that, in addi- by the hindbrain (Bok et al., 2005). Because mutation of Hoxa2 tion to forming a sleeve surrounding CV neuronal cell bodies and causes significant mis-patterning of R4 NCC it is possible that 208 L.L. Sandell et al. / Developmental Biology 385 (2014) 200–210 abnormal anterior–posterior specification of the NCC could be neurite extensions. The shape of the early cochlear nerve as a responsible for human HOXA2 inner ear defects. fan opening with the growth of the cochlea is consistent with the Analysis of Wnt1-Cre and 6.5Pax3Cre drivers in combination known maturation pattern of spiral ganglion neurons, wherein with R26R and Z/EG reporters does not indicate any significant NCC neurons at the base of the cochlea mature to their final cell contribution to the neuronal population of the CV nerve nor to the division before those at the apex (Koundakjian et al., 2007; otic epithelium, a result consistent with many previous studies Matei et al., 2005; Ruben, 1967). Importantly, the presence of (Breuskin et al., 2010; D’Amico-Martel and Noden, 1983; van migratory NCC in association with spiral ganglion neurons sug- Campenhout, 1935) but dissimilar from a recent report of trans- gests that NCC may play a role in guidance or growth of the genic mouse lineage trace analysis (Freyer et al., 2011). peripheral neuronal projections in a manner similar to their role in facilitating growth and targeting of central axons (Begbie and Early central axon and peripheral neurite extension integrated with Graham, 2001; Freter et al., 2013). NCC The issue of whether NCC guide peripheral nerve outgrowth has been examined previously with mixed results. Yntema For CV nerve development, the timing of projection of central demonstrated that NCC are required for migration of peripheral axons to the hindbrain and peripheral neurites to the sensory motor nerve fibers into the fore- and hind-limb of Amblystoma targets in the otic epithelium has been previously examined by (Yntema, 1943a). He also observed that ablation of cranial NCC neuronal dye tracing. With that method central axons are detected caused changes in the course of the trigeminal and facial nerves in as early as embryonic day 12.5 (E12.5) and peripheral neurites at chick (Yntema, 1944). Consistent with the notion that NCC and E11.5 (Fritzsch, 2003; Matei et al., 2005). Radiolabeling of terminal peripheral nerve outgrowth are linked, Noakes et al. observed that mitoses has also been used to investigate the temporal profile of Schwann cells migrate ahead of neuronal peripheral neurites in CV nerve development. Terminal mitoses staging indicates that CV the forelimb in chick (Noakes and Bennett, 1987), and that ablation Schwann cells reach maturity much later during embryogenesis of NCC blocked extension of motor neurons into the limb (Noakes than the corresponding neuronal cells with which they associate et al., 1988). (Altman and Bayer, 1982; Ruben, 1967). Analysis of mice with mutations of ErbB2 or ErbB3, in which The results of the present study, based on immunostaining for migration and survival of Schwann cell progenitors are impaired, βIII neuronal tubulin, demonstrate that central axon projection provides further evidence that Schwann cells progenitors play and and peripheral neurite extension can be detected beginning as important role in growth and formation of cranial sensory and early as E10.5, somewhat earlier than reported based on dye motor nerves (Erickson et al., 1997; Lee et al., 1995). For the labeling analysis. Moreover, immunostaining NCC lineage repor- cochlear nerve, loss of ErbB2 results in abnormal migration of ters in conjunction with immunolabeling of neuronal projections spiral ganglion neurons and aberrant projections of afferent indicates that, although the neuronal and glial cell populations peripheral neurites (Morris et al., 2006). Central axon projections may reach their terminal mitosis at different development stages, are targeted correctly. Thus, migration of NCC glial progenitors they are coordinately involved in the initial morphological devel- may not be important for central axon guidance. Alternatively, the opment of the nerve. Additionally, the resolution of the confocal normal central axon guidance may be integrated with early NCC 3D reconstruction images reveals the early central axonal trunk migration not disrupted by loss of ErbB2, which impairs Schwann and peripheral neurite branches of the E10.5 CV nerve are cell survival at later stages. organized as a lattice structure with NCC arranged between the Contradicting the model that NCC play a role in axon guidance interconnecting rungs of the neurite lattice. NCC are also posi- of peripheral nerves, genetic evidence from mice indicates that tioned at junctions near terminal extending filopodia. The position NCC are dispensable for the process. Genetic ablation of NCC using of NCC at junctions of neurite filopodia is reminiscent of historical Wnt1Cre to activate a Diphtheria toxin allele results in disorga- observations of Schwann cells at junctions of branching peripheral nized growth of nerves including the facial nerve, but the corda neurites in developing frog embryos (Harrison, 1924). These tympani and greater superficial petrosal components of the facial observations that NCC are present throughout the lattice of early nerve form relatively normally (Coppola et al., 2010). The prevail- CV nerve reinforces the idea that neuronal and glial components of ing view that Schwann cell progenitors do not guide peripheral CV nerve architecture are established concomitantly at early stages neurite outgrowth (Woodhoo and Sommer, 2008), is based in part of nerve development. on observations of Pax3splotch mutant mice, which lack Schwann From the images presented here it is unclear whether NCC cells in nerves of the sacral region but have correctly targeted precede or follow the peripheral neurite projections. The recent hindlimb motor nerves (Grim et al., 1992). The disparate evidence analysis of epibranchial placode neuron development by Freter et al. regarding the role of NCC-derived glia in guiding peripheral may provide some clue regarding this issue (Freter et al., 2013). In neurite outgrowth may reflect variation in guidance mechanisms that study of facial ganglion development, formation of the gang- between specific nerves types, difference in axial levels, or may lion and growth of central axons was demonstrated to depend upon reflect variation between organisms. Alternatively, differences may reciprocal interactions between NCC and epibranchial placode result from variation in method of NCC perturbation with respect neurons. Additionally, in vitro cultured placodal neurons were to the number of surviving or regenerating NCC. shown to extend processes better on a substrate of NCC than on a substrate of mesoderm. While the analysis did not distinguish Elimination of NCC disrupts outgrowth and targeting of central axons between central axons and peripheral neurites, the result suggests and peripheral neurites of the CV nerve that NCC may facilitate formation of axon and neurite projections by providing a favorable substrate for outgrowth. Owing to the close association of NCC with the developing CV neuronal cells in mice we assessed the effect of NCC ablation on Peripheral neurite extension of cochlear nerve accompanied by a axon outgrowth of the CV nerve in chick. Severe disruption of wave of NCC central axon and peripheral neurite outgrowth occurred following ablation of R4 NCC. The results recapitulate those reported for the One of the most striking observations from this analysis is that facial nerve (Begbie and Graham, 2001; Freter et al., 2013), and the early peripheral neurites of the cochlear nerve project fan-like demonstrate that axon growth for the CV nerve is likewise with NCC represented robustly at the wave front of terminal integrated with the stream of migratory NCC from R4. L.L. Sandell et al. / Developmental Biology 385 (2014) 200–210 209

Owing to the plasticity of NCC, physical ablation often resulted Bok, J., Chang, W., Wu, D.K., 2007. Patterning and morphogenesis of the vertebrate in migration of NCC regenerated from ectopic positions. In such inner ear. Int. J. Dev. Biol. 51, 521. Breuskin, I., Bodson, M., Thelen, N., Thiry, M., Borgs, L., Nguyen, L., Stolt, C., Wegner, cases, growth of axons was guided precisely along the pathways or M., Lefebvre, P.P., Malgrange, B., 2010. Glial but not neuronal development in corridors defined by the regenerating ectopic NCC. Similar results the cochleo‐ requires Sox10. J. Neurochem. 114, 1827–1839. have been observed for neurons of the Brown, C.B., Engleka, K.A., Wenning, J., Min, Lu, M., Epstein, J.A., 2005. Identification following genetic ablation of R4 NCC in mice (Chen et al., 2011). of a hypaxial somite enhancer element regulating Pax3 expression in migrating myoblasts and characterization of hypaxial muscle Cre transgenic mice. Genesis Importantly, we observe that CV nerve axons project and are 41, 202–209. guided to ectopic positions of the central nervous system when Carney, P.R., Silver, J., 1983. Studies on cell migration and axon guidance in the – very few, or even isolated individual NCC are present. In all cases, developing distal of the mouse. J. Comp. Neurol. 215, 359 369. Chai, Y., Jiang, X., Ito, Y., Bringas Jr., P., Han, J., Rowitch, D.H., Soriano, P., McMahon, A.P., CV axonal and neurite processes follow with perfect concordance Sucov, H.M., 2000. Fate of the mammalian cranial neural crest during tooth and the pathways of migratory NCC. The observation that isolated mandibular morphogenesis. Development 127, 1671–1679. regenerating NCC are sufficient to guide central axon growth Chen, J., Streit, A., 2013. Induction of the inner ear: stepwise specification of otic fate from multipotent progenitors. Hear. Res. 297, 3–12. suggests that experimental paradigms wherein even a few isolated Chen, Y., Takano-Maruyama, M., Gaufo, G.O., 2011. Plasticity of neural crest–placode NCC are present may not definitively test the role of NCC derived interaction in the developing visceral nervous system. Dev. Dyn. 240, glia in axon/neurite outgrowth and guidance. 1880–1888. Coppola, E., Rallu, M., Richard, J., Dufour, S., Riethmacher, D., Guillemot, F., Goridis, C., Brunet, J.-F., 2010. Epibranchial ganglia orchestrate the development of the cranial neurogenic crest. Proc. Natl. Acad. Sci. 107, 2066–2071. Conclusions Corfas, G., Velardez, M.O., Ko, C.-P., Ratner, N., Peles, E., 2004. Mechanisms and roles of Axon–Schwann cell interactions. J. Neurosci. 24, 9250–9260. The otic vesicle-derived neurons and NCC-derived glial popula- Culbertson, M.D., Lewis, Z.R., Nechiporuk, A.V., 2011. Chondrogenic and gliogenic subpopulations of neural crest play distinct roles during the assembly of tions of the CV nerve develop in tandem from early stages of CV epibranchial ganglia. PLoS ONE 6, e24443. nerve morphogenesis. NCC emerging from R4 are arranged as a D’Amico-Martel, A., Noden, D., 1983. Contributions of placodal and neural crest – sleeve that surrounds the nested CV and facial ganglia and the cells to avian cranial peripheral ganglia. Am. J. Anat. 166, 445 468. Davies, D., 2007. Temporal and spatial regulation of α6 integrin expression during axons projecting centrally from them. Early peripheral neurite the development of the cochlearvestibular ganglion. J. Comp. Neurol. 502, projections are organized as a lattice structure with NCC glial 673–682. progenitors intercalated throughout. As CV nerve morphogenesis Echelard, Y., Vassileva, G., McMahon, A.P., 1994. Cis-acting regulatory sequences governing Wnt-1 expression in the developing mouse CNS. Development 120, progresses, NCC continue to be robustly associated with periph- 2213–2224. erally projecting neurites of the vestibular and cochlear nerves. Erickson, S.L., O’Shea, K.S., Ghaboosi, N., Loverro, L., Frantz, G., Bauer, M., Lu, L.H., Perturbation of R4 NCC impairs formation of CV central axon Moore, M.W., 1997. ErbB3 is required for normal cerebellar and cardiac development: a comparison with ErbB2-and heregulin-deficient mice. Devel- projections, with central axons forming precisely along aberrant opment 124, 4999–5011. migration pathways of ectopically regenerating NCC. Together Fekete, D.M., Campero, A.M., 2007. Axon guidance in the inner ear. Int. J. Dev. Biol. these results support a model wherein the CV nerve, as well as 51, 549–556. other cranial sensory nerves, develops by integrated interactions Freter, S., Fleenor, S.J., Freter, R., Liu, K.J., Begbie, J., 2013. Cranial neural crest cells form corridors prefiguring sensory neuroblast migration. Development 140, between placodally-derived neuronal progenitors and NCC- 3595–3600. derived glial progenitors. Freyer, L., Aggarwal, V., Morrow, B.E., 2011. Dual embryonic origin of the mammalian otic vesicle forming the inner ear. Development 138, 5403–5414. Fritzsch, B., 2003. Development of inner ear afferent connections: forming primary neurons and connecting them to the developing sensory epithelia. Brain Res. Acknowledgment Bull. 60, 423–433. Fritzsch, B., Christensen, M.A., Nichols, D.H., 1993. Fiber pathways and positional Grant Support: Research in the Sandell laboratory is supported changes in efferent perikarya of 2.5-to 7-day chick embryos as revealed with dil and dextran amiens. J. Neurobiol. 24, 1481–1499. by the University of Louisville and by P20 RR017702 to Dr. Robert Gendron-Maguire, M., Mallo, M., Zhang, M., Gridley, T., 1993. Hoxa-2 mutant mice M. Greene, Birth Defects Center, University of Louisville, Louisville, exhibit homeotic transformation of skeletal elements derived from cranial KY. Research in the Trainor laboratory is supported by the Stowers neural crest. Cell 75, 1317–1331. Grim, M., Halata, Z., Franz, T., 1992. Schwann cells are not required for guidance of Institute for Medical Research and the National Institute of Dental motor nerves in the hindlimb in Splotch mutant mouse embryos. Anat. and Craniofacial Research (R01 DE 016082). The Microscopy Suite Embryol. 186, 311–318. at the Cardiovascular Innovation Institute in Louisville, KY is Groves, A.K., Fekete, D.M., 2012. Shaping sound in space: the regulation of inner ear patterning. Development 139, 245–257. supported by GM103507. Hanani, M., 2005. Satellite glial cells in sensory ganglia: from form to function. Brain Res. Rev. 48, 457–476. References Harrison, R.G., 1924. Neuroblast versus sheath cell in the development of peripheral nerves. J. Comp. Neurol. 37, 123–205. Harrison, R.G., 1936. Relations of symmetry in the developing ear of amblystoma Alasti, F., Sadeghi, A., Sanati, M.H., Farhadi, M., Stollar, E., Somers, T., Van Camp, G., punctatum. Proc. Natl. Acad. Sci. 22, 238–247. 2008. A mutation in HOXA2 is responsible for autosomal-recessive microtia in Hemond, S.G., Morest, D.K., 1992. Tropic effects of otic epithelium on cochleo- an Iranian family. Am. J. Hum. Genet. 82, 982–991. vestibular ganglion fiber growth in vitro. Anat. Rec. 232, 273–284. Altman, J., Bayer, S., 1982. Development of the and Related Ikeya, M., Lee, S.M.K., Johnson, J.E., McMahon, A.P., Takada, S., 1997. Wnt signalling Nuclei in the Rat. Springer-Verlag, Berlin. required for expansion of neural crest and CNS progenitors. Nature 389, Appler, J.M., Goodrich, L.V., 2011. Connecting the ear to the brain: molecular 966–970. mechanisms of auditory circuit assembly. Prog. Neurobiol. 93, 488–508. Kopecky, B., Johnson, S., Schmitz, H., Santi, P., Fritzsch, B., 2012. Scanning thin-sheet Ard, M.D., Morest, D.K., Hauger, S.H., 1985. Trophic interactions between the laser imaging microscopy elucidates details on mouse ear development. Dev. cochleovestibular ganglion of the chick embryo and its synaptic targets in Dyn.. culture. 16, 151–170. Koundakjian, E.J., Appler, J.L., Goodrich, L.V., 2007. Auditory neurons make stereo- Begbie, J., Ballivet, M., Graham, A., 2002. Early steps in the production of sensory typed wiring decisions before maturation of their targets. J. Neurosci. 27, neurons by the neurogenic placodes. Mol. Cell. Neurosci. 21, 502–511. 14078–14088. Begbie, J., Brunet, J.F., Rubenstein, J.L., Graham, A., 1999. Induction of the epibran- Ladher, R.K., O’Neill, P., Begbie, J., 2010. From shared lineage to distinct functions: chial placodes. Development 126, 895–902. the development of the inner ear and epibranchial placodes. Development 137, Begbie, J., Graham, A., 2001. Integration between the epibranchial placodes and the 1777–1785. hindbrain. Science 294, 595–598. Lee, K.-F., Simon, H., Chen, H., Bates, B., Hung, M.-C., Hauser, C., 1995. Requirement Bianchi, L.M., Cohan, C.S., 1991. Developmental regulation of a neurite-promoting for neuregulin receptor erbB2 in neural and cardiac development. Nature 378, factor influencing statoacoustic neurons. Dev. Brain Res. 64, 167–174. 394–398. Bok, J., Bronner-Fraser, M., Wu, D.K., 2005. Role of the hindbrain in dorsoventral but Matei, V., Pauley, S., Kaing, S., Rowitch, D., Beisel, K.W., Morris, K., Feng, F., Jones, K., not anteroposterior axial specification of the inner ear. Development 132, Lee, J., Fritzsch, B., 2005. Smaller inner ear sensory epithelia in Neurog1 null 2115–2124. mice are related to earlier cycle exit. Dev. Dyn. 234, 633–650. 210 L.L. Sandell et al. / Developmental Biology 385 (2014) 200–210

Milewski, R.C., Chi, N.C., Li, J., Brown, C., Lu, M.M., Epstein, J.A., 2004. Identification Schwarz, Q., Vieira, J.M., Howard, B., Eickholt, B.J., Ruhrberg, C., 2008. Neuropilin of minimal enhancer elements sufficient for Pax3 expression in neural crest 1 and 2 control cranial gangliogenesis and axon guidance through neural crest and implication of Tead2 as a regulator of Pax3. Development 131, 829–837. cells. Development 135, 1605–1613. Morris, J.K., Maklad, A., Hansen, L.A., Feng, F., Sorensen, C., Lee, K.-F., Macklin, W.B., Soriano, P., 1999. Generalized lacZ expression with the ROSA26 Cre reporter strain. Fritzsch, B., 2006. A disorganized innervation of the inner ear persists in the Nat. Genet. 21, 70–71. absence of ErbB2. Brain Res. 1091, 186–199. Streeter, G.L., 1906. On the development of the membranous labyrinth and the Noakes, P.G., Bennett, M.R., 1987. Growth of axons into developing muscles of the acoustic and facial nerves in the human embryo. Am. J. Anat. 6, 139–165. chick forelimb is preceded by cells that stain with Schwann cell antibodies. Tessarollo, L., Coppola, V., Fritzsch, B., 2004. NT-3 Replacement with brain-derived – J. Comp. Neurol. 259, 330 347. neurotrophic factor redirects vestibular nerve fibers to the cochlea. J. Neurosci. Noakes, P.G., Bennett, M.R., Stratford, J., 1988. Migration of schwann cells and axons 24, 2575–2584. into developing chick forelimb muscles following removal of either the neural Tosney, K., 1982. The segregation and early migration of cranial neural crest cells in tube or the neural crest. J. Comp. Neurol. 277, 214–233. the avian embryo. Dev. Biol. 89, 13–24. Novak, A., Guo, C., Yang, W., Nagy, A., Lobe, C.G., 2000. Z/EG, a double reporter van Campenhout, E., 1935. Experimental researches on the origin of the acoustic mouse line that expresses enhanced green fluorescent protein upon cre- ganglion in amphibian embryos. J. Exp. Zool. 72, 175–193. mediated excision. Genesis 28, 147–155. Woodhoo, A., Sommer, L., 2008. Development of the Schwann cell lineage: from the Osborne, N.J., Begbie, J., Chilton, J.K., Schmidt, H., Eickholt, B.J., 2005. Semaphorin/ neural crest to the myelinated nerve. Glia 56, 1481–1490. neuropilin signaling influences the positioning of migratory neural crest cells Wu, D.K., Nunes, F.D., Choo, D., 1998. Axial specification for sensory organs versus within the hindbrain region of the chick. Dev. Dyn. 232, 939–949. – Rijli, F.M., Mark, M., Lakkaraju, S., Dierich, A., Dolle, P., Chambon, P., 1993. non-sensory structures of the chicken inner ear. Development 125, 11 20. A homeotic transformation is generated in the rostral branchial region of the Yang, T., Kersigo, J., Jahan, I., Pan, N., Fritzsch, B., 2011. The molecular basis of head by disruption of Hoxa-2, which acts as a selector gene. Cell 75, 1333–1349. making spiral ganglion neurons and connecting them to hair cells of the organ – Rosenbluth, J., 1962. The fine structure of acoustic ganglia in the rat. J. Cell Biol. 12, of Corti. Hear. Res. 278, 21 33. fi 329–359. Yntema, C.L., 1943a. De cient efferent innervation of the extremities following Ruben, R.J., 1967. Development of the inner ear of the mouse: a radioautographic removal of neural crest in Amblystoma. J. Exp. Zool. 94, 319–349. study of terminal mitoses. Acta Oto-Laryngologica Suppl 220, 221. Yntema, C.L., 1943b. An experimental study on the origin of the sensory neurones Sandell, L.L., Kurosaka, H., Trainor, P.A., 2012. Whole mount nuclear fluorescent and sheath cells of the IXth and Xth cranial nerves in Amblystoma punctatum. imaging: convenient documentation of embryo morphology. Genesis 50, J. Exp. Zool. 92, 93–119. 844–850. Yntema, C.L., 1944. Experiments on the origin of the sensory ganglia of the facial Schlosser, G., 2010. Chapter four—making senses: development of vertebrate nerve in the chick. J. Comp. Neurol. 81, 147–167. cranial placodes. In: Kwang, J. (Ed.), International Review of Cell and Molecular Biology. Academic Press, San Diego, CA, pp. 129–234.