Abstract
Neural crest cells (NCCs) are essential embryonic progenitor cells that are unique to vertebrates and form a remarkably complex and coordinated system of highly motile cells. Migration of NCCs occurs along specific pathways within the embryo in response to both environmental cues and cell-cell interactions within the neural crest population. Here, we demonstrate a novel role for the putative Sonic hedgehog (Shh) receptor and cell adhesion regulator, cdon, in zebrafish neural crest migration. cdon is expressed in developing premigratory NCCs but is downregulated once the cells become migratory. Knockdown of cdon results in aberrant migration of trunk NCCs: crestin positive cells can emigrate out of the neural tube but stall shortly after the initiation of migration. Live cell imaging analysis demonstrates reduced directedness of migration, increased velocity and mispositioned cell protrusions. In addition, transplantation analysis suggests that cdon is required cell-autonomously for directed NCC migration in the trunk. Interestingly, N-cadherin is mislocalized following cdon knockdown suggesting that the role of cdon in NCCs is to regulate N-cadherin localization. Our results reveal a novel role for cdon in zebrafish neural crest migration, and suggest a mechanism by which Cdon is required to localize N-cadherin to the cell membrane in migratory NCCs for directed migration.
Keywords: neural crest, cell migration, zebrafish, shh signaling, adhesion
Introduction
During embryonic development, neural crest cells (NCCs) are specified at the neural- ectodermal boundary or neural plate border (NPB) by a series of inductive cues and combinatorial transcription factor regulation. Once specified, NCCs undergo dramatic changes in cell polarity and adhesion to undergo an epithelial-to-mesenchymal transition or EMT. They then migrate from the neural tube utilizing cues from the environment and from within the NCC population to reach their final differentiation point within the embryo. In zebrafish, NCCs first emigrate to a ‘neural crest staging area’ positioned dorsally in the neural keel/tube (Le Douarin, 1982; Raible and Eisen, 1996; Raible et al., 1992) and then exit to migrate along two major pathways in the trunk: a dorsolateral pathway under the epidermis and a ventromedial pathway along the anterior somite. The migration path is determined both by temporal exit from the dorsal neural tube as well as spatial cues from the environment (Kuriyama and Mayor, 2008). NCC migration requires the coordination of key cellular processes including changes in cell-cell adhesion, cell-matrix interactions, polarization of the cell, and reception and interpretation of various attractive and inhibitory cues (Theveneau and Mayor, 2012). One cell autonomous mechanism by which NCCs migrate collectively forward is by contact inhibition of locomotion (CIL), which functions in several ways to promote proper NCC migration (Carmona-Fontaine et al., 2008; Carmona-Fontaine et al., 2011; Theveneau et al., 2013). CIL is the process whereby cells at the leading edge of each NCC migratory stream are actively inhibited from moving backwards by the cells behind them, and cells that transiently contact one another are inhibited from adhering to one another, in sum allowing a group of NCCs to move collectively forward. N-cadherin (N-cad; also called Cdh2) mediates cell-cell adhesion and is required for cells to display CIL in Xenopus (Theveneau et al., 2013). Because N-cadherin is required for early development in mammals, the role of N-cadherin in NCC migration was not specifically analyzed in mouse null mutants (Monier-Gavelle and Duband, 1995) (Lele et al., 2002) nor in zebrafish (Lele et al., 2002).
In an effort to determine the molecular mechanisms of NCC migration, several NCC migratory guidance signals have been identified, including Ephrin and Semaphorin signaling as well as Sdf1(Cxcl12)/Cxcr4 signaling (Davy and Soriano, 2005; Gammill et al., 2006; Krull et al., 1997; Schwarz et al., 2009a; Schwarz et al., 2009b; Yu and Moens, 2005), (Belmadani et al., 2005; Olesnicky Killian et al., 2009; Theveneau et al.). However, the role of Shh signaling in neural crest migration is less well described although clearly a major factor in NCC biology. Naturally occurring mutations in humans and in mice where the Shh signaling pathway is disrupted result in severe holoprosencephaly (Hayhurst and McConnell, 2003). The loss of the Shh pathway signal transducer Smoothened specifically in NCCs (Izzi et al., 2011; Tenzen et al., 2006), or loss of total Shh signaling in later development results in craniofacial defects (Eberhart et al., 2006; Wada et al., 2005). In the zebrafish, knockdown of shh results in delayed formation of trunk NCCs populating the dorsal root ganglia (Ungos et al., 2003). In vitro studies support the idea that Shh mediates outgrowth and dispersion of NCCs on fibronectin (Testaz et al., 1999; Testaz et al., 2001), however, this interaction is independent of Patched/Smoothened and Gli transcription, suggesting another unknown Shh receptor or receptors must be required for this activity.
Cell-adhesion molecule-related/down-regulated by Oncogenes or cdon acts as a Shh receptor in many developmental contexts. cdon is a member of the neural cell adhesion molecule (N-CAM) family and has been identified as a Shh receptor, acting as a co-receptor with Patched1 (Izzi et al., 2011; Tenzen et al., 2006; Zhang et al., 2006), and in myoblast and neuronal differentiation, it binds N-cadherin and induces p38/MAPK signaling to direct cell differentiation and apoptosis (Lu and Krauss, 2010). Chick Cdon and the associated receptor BOC (Brother of Cdon) have been found to localize to long filopodial protrusions in Shh-responding cells and act as long-distance receptors that illicit a cell signaling response (Sanders et al., 2013). Additionally, Cdon human and mouse mutants display a microform holoprosencephaly phenotype, which is likely the result of deficient Shh signaling (Cole and Krauss, 2003), although it is not known if there are deficiencies in NCC development (Bae et al., 2011). Two recent studies suggest that cdon plays a role as a dependence receptor in NCC survival in the branchial arches of chick embryos and as a negative regulator of Shh in zebrafish optic cup development (Cardozo et al., 2014; Delloye-Bourgeois et al., 2014). Together, these studies suggest that cdon has multiple roles in development but its specific role in trunk NCCs remains unknown.
Here, we demonstrate the role of cdon in zebrafish trunk NCC migration. cdon is expressed in developing pre-migratory NCCs and loss of cdon leads to a decrease in ventral migration of trunk NCCs due to a loss of directed cell movement. cdon functions cell-autonomously to regulate directed NCC migration through proper localization of N-cadherin to the cell membrane, and is regulated by the Shh signaling pathway. Together, these data suggest a novel, Shh-dependent role for cdon in zebrafish NCC migration.
Materials and methods
Zebrafish maintenance and lines
Zebrafish were maintained according to Westerfield’s Zebrafish Book (Westerfield, 1993). Wildtype (WT) strains include AB, TAB, and EKK lines (ZIRC) and transgenic lines include sox10:eGFP (Dutton et al., 2008), tg(sox10:memRFP)vu234(Blasky et al., 2014) and tg(hsp70l:shha-eGFP) (Shen, et al. 2013). Developmental staging followed previously published standards (Kimmel et al., 1995). All experiments utilizing zebrafish are approved by UC Denver IACUC and conform to NIH regulatory standards of care and treatment.
In situ hybridization, immunofluorescence and Alcian Blue staining
Whole-mount RNA in situ hybridization (ISH) was performed as previously described (Thisse and Thisse, 1998). The cdon DIG clone for synthesizing the antisense probe was obtained by PCR from ZIRC clone cb507 using primers forward 5′-GGATCCATTAACCCTCACTAA-3′ and reverse 5′-TAATACGACTCACTATAGGG-3′. DIG-conjugated antisense probes were synthesized as previously published for the following genes: foxd3 (Powell et al., 2013), sox10 (Olesnicky et al., 2010), crestin (Rubinstein et al., 2000). Fluorescent ISH was performed as previously described (Pineda et al., 2006) and used the TSA Biotin System (Perkin Elmer NEL700A001KT) followed by Streptavidin Alexa Fluor 488 antibody (Invitrogen S11223) and the Sigma Fast Red kit (Sigma F4648). For immunofluorescence, embryos were fixed in 4% PFA for 1–2 hours at room temperature, embedded and cryo-sectioned at 12 μm. Immunofluorescent staining was performed using MNCD2 antibody (DSHB) with anti rat IgG secondary, anti-GFP antibody (A2131 Invitrogen) and DAPI (Invitrogen) and imaged via spinning-disk confocal. Alcian blue staining was performed as previously described (Johnson et al 2011).
Morpholino and mRNA injections and controls
We designed two morpholinos targeting the start site (ATG) and exon 2-intron 2 (e2i2) splice boundary of cdon to knockdown its expression. All morpholino controls were performed and no off target affects were observed (see below and Supplemental Figure 1, 2) and the ATG morpholino was used in subsequent experiments. Morpholino oligonucleotides (Gene Tools) were injected at the 1–2 cell stage with rhodamine dextran (Molecular Probes). Morpholinos include cdon ATG (5′-ATC TCA GGC CAC CGT CCT CCA TGA C -3′) and cdon E2I2 splice site (5′-ATCAGATGTGCTTACCTGCAATGGT -3′) injected at 3–6 ng and 7–14 ng respectively. Co-injection of cdon ATG MO (6ng) and p53 MO (2ng) 5′-GCG CCA TTG CTT TGC AAG AAT TG-3′ (Gene Tools, LLC; Robu et al. 2007) was performed. Verification of e2i2 mis-splicing was performed by RT-PCR as follows: mRNA from 5–10 pooled, 24 hpf embryos injected with 7–14 ng cdon E2i2 MO or 10 uninjected embryos was isolated using TRIZOL or the RNAeasy Micro Kit (Qiagen) and reverse transcribed into cDNA using the Superscript III First-Strand Synthesis System for RT-PCR (Invitrogen). Primers to amplify the cDNA to observe the splice defects are as follows: forward (5′-CACTAGATCCCAGTCACCATT-3′) and reverse (5′-AAATAAGGGCGACACTGCT -3′). In uninjected embryos, the expected PCR product is about 215 bp. Mis-splicing generates a fusion of e2i2 resulting in a 2.1 kb fragment (Supplemental Figure 1). Rescue experiments with cdon mRNA were performed by cloning full-length cdon into pCS2+ using the following primers: forward: (5′-GCC GCC ACC ATG GAG GAC GGT GGC CTG A-3′) and reverse: (5′-GAATTC TCAG ATC TCC TGC ACG GTG GG-3′). mRNA was then synthesized using the SP6 mMESSAGE mMACHINE kit (Ambion)
Live-imaging and transplants in transgenic embryos
sox10:eGFP embryos were injected with cdon morpholino as above and allowed to develop to 19–20 hpf. Embryos were imbedded in 1% agar with tricaine, mounted in a heated chamber and imaged by Zeiss Axio Observer inverted compound microscope using PerkinElmer UltraVIEW VoX Confocal System for 5–6 hours. Images were processed using Volocity Software (PerkinElmer) using deconvolution and contrast enhancement and assembled in QuickTime movies. Quantification of velocity and displacement were measured in 40 micron Z-stack tracked using Volocity v6.1.1. Statistical analysis was done in Excel followed by two-tailed Student’s t-test. For mosaic analysis, donor sox10:eGFP embryos were injected with 2.5% blue dextran (10,000 MW, lysine fixable, Invitrogen). At mid-blastula stages, embryos were dechorionated and approximately 25–50 cells were transplanted from the presumptive NPB of wildtype or morpholino injected sox10:eGFP embryos at blastomere stage to the presumptive NPB of sox10:RFP embryos in penicillin-streptomycin treated embryo media. Embryos were live-imaged at 30 hpf by fluorescent Olympus BX51WI compound microscope using a QImaging camera, and images were processed using Adobe Photoshop CC (Adobe Systems Inc).
Pharmacological treatments and qRT-PCR
At 10 hpf, manually dechorionated embryos were transferred into embryo media containing 10–50 μM cyclopamine in 100% EtOH (Sigma and Toronto Research Chemicals) or 10μl of 100% EtOH diluted in 10ml of embryo water as a carrier control and incubated at 28.5°C until 24 hpf. After one hour, embryos were fixed in 4% PFA and processed for ISH. qRT-PCR was performed as above. Primers for cdon (Dr3429263_m1), patched1/2 (Dr03118687_m1), gli1 (Dr0393669_m1) and internal control gene eIf1a (Dr3432748_m1) were used in TAQMAN assays on an AB700 Thermocycler (Life Technologies, CA).
Biological replicates, statistics and quantification
All experiments were performed with 3 or more biological replicates. In addition, qRT-PCR experiments contained at least 3 technical replicates per biological sample. For statistical analysis, Student’s t-test was used unless otherwise noted. Graphical error bars denote standard deviation or standard error of the mean as noted. All distance measurements were made using ImageJ. The distance between the dorsal edge of the embryo and the most ventral migrated NCC in each of the first 7 somites over the yolk extension was measured in μm. For fluorescent intensity analysis of N-cadherin expression, cells that had not migrated more than the midpoint of the neural tube were selected. Using plot profile in ImageJ, the midpoint of the cell was determined, and a line drawn through the center of the cell using the sox10-egfp positive cytoplasm to gate the membrane (ie at the edge of the gfp). A line scan of fluorescence intensity was then drawn (Figure 6). These scans were then normalized between 0–1 and a ratio calculated for membrane/total fluorescence, cytoplasm/total, and membrane/cytoplasm and display in graphical form in Figure 6. 14–16 cells were measured for both wildtype and morphants. For quantification of protrusions in Figure 3, the angle of each protrusion was measured in ImageJ over a 4 hour live cell imaging experiment using a membrane tagged sox10:mRFP. MATLAB was then used to bin the protrusion measurements into rose plots and to determine the statistical significance in the deviation from uniformity for each datatset.
Results
cdon is expressed in newly specified NCCs and during emigration and the initiation of migration
cdon is a potential cell-adhesion regulator and signaling receptor that is expressed in the mouse NPB and is downstream of the NPB specificer Prdm1a in zebrafish (Mulieri et al., 2000; Olesnicky et al., 2010). To determine if cdon may be involved in neural crest development, we analyzed the expression of cdon during zebrafish development. In situ hybridization shows that cdon is expressed specifically in the neural plate border (11 hpf) and at later stages (16–24 hpf) in the dorsal neural tube as well as diffusely in the developing somite (Figure 1A–D). We further examined if cdon is co-expressed with the NCC marker crestin by double fluorescent in situ hybridization (FISH). This analysis revealed that cdon mRNA (red) is co-expressed with the neural crest marker crestin (green) in the pre-migratory NCCs within the dorsal neural tube of the trunk but not migratory cells (Figure 1E, F). Cryosections of the trunk region of the double FISH embryos confirms the colocalization of cdon with crestin in the dorsal neural tube, in the neural crest migratory staging area (Figure 1G).
Figure 1. cdon is expressed in the NPB and premigratory trunk NCCs.
Dorsal view and anterior to the top (A), and lateral views, anterior to the left (B–F). ISH of 11, 16, 18, and 24 hpf stage wildtype embryos. (A–D) cdon is expressed in the neural plate border (NPB) and dorsal neural tube (arrows). (E, F) Lateral view confocal single optical plane micrograph of double fluorescent ISH of 24 hpf wildtype embryos for crestin (green) and cdon (red). cdon is expressed in premigratory NCCs in the trunk (yellow) but not in the migratory streams (arrows, E) and widely expressed in cranial region including the brain, eye, and pharyngeal arches (F). 14μm cryosections of double FISH embryos suggests that cdon and crestin colocalize in the dorsal neural tube/neural crest staging area in the trunk (G). b, brain; e, eye; NPB, neural plate border; NT, neural tube; pas, pharyngeal arches.
In the head region, cdon is expressed in premigratory NCCs in the dorsal brain and in the eye but not in crestin-positive migratory NCCs migrating ventrally into the pharyngeal arches at 24 hpf (Figure 1D, F, red). However, cdon is expressed in the pharyngeal arch mesenchyme once the NCCs reach their destination and populate the arches. These data suggest that cdon is expressed in premigratory NCCs and at the initiation of migration in the head and trunk region and, at the mRNA level, is downregulated prior to migration away from the trunk neural tube/keel.
cdon is not required for neural crest specification
After determining that cdon is expressed during neural crest specification and prior to NCC migration, we asked if cdon is required for neural crest specification by knockdown of expression with antisense morpholinos. Using a morpholino to the translational start site (cdon ATG), we observed that morphant embryos had no change in foxd3 and sox10 expression at 11–13 hpf compared to wildtype controls (Supplemental Figure 1), suggesting cdon does not play a role in NCC specification. In a small percentage of embryos (<5%) there is a slight expansion of the neural plate resulting in a wider appearance of the neural crest domain; nonetheless cdon morphant embryos display a normal degree of NCC specification.
cdon is required specifically for trunk neural crest migration
Interestingly, in the examination of later stage cdon morphants, we found significant defects in neural crest migration. While NCCs emigrate out of the neural keel, trunk NCC migration is severely impaired as NCCs appear to exit the dorsal neural tube, but appear to stall and do not migrate ventrally as wildtype NCCs do (Figure 2). At early stages of migration (16–18 hpf), crestin and sox10-expressing NCCs remain dorsal compared to wildtype (Figure 2A, E, C, G), accounting for the reduction of NCCs seen migrating (average distance migrated in wildtype=79.4 μm, distance in cdon MO=45.4 μm). Additionally, ectopic NCCs are observed clustered along the dorsal midline of cdon morphant embryos (Figure 2B, F). At 24 hpf, cdon morphant NCCs have not migrated as far from the dorsal midline and display a rounded morphology (Figure 2C, G, yellow line is the position of the horizontal myoseptum and the midpoint of ventral NCC migration). Quantification of the average distance that NCCs migrated ventrally was significantly less than wildtype uninjected embryos (Figure 2J, wildtype=146.9 μm, cdon MO=79.7 μm). While cdon morphant NCCs do stall in the dorsal aspect of the embryo, they are still viable, as no changes were observed with cleaved-caspase 3 staining in the trunk (data not shown). To assay the ability of NCCs to migrate long distances and form ventral derivatives we examined pigment cells and craniofacial cartilage. In the trunk, we determine that while melanocytes are specified normally in cdon morphants, they do not migrate to populate the ventral trunk as compared wildtype embryos 48 hpf (Figure 2D, H). Quantification suggests the number of melanocytes that populate the yolk extension is significantly decreased in cdon morphants (Figure 2J; average number of melanocytes over yolk extension in wildtype=30, cdon MO=6.7). In the cranial region, we observed the effects of cdon knockdown on craniofacial cartilage development and found no change in the formation and patterning of alcian blue stained cartilages in cdon morphants (Supplemental Figure 2A–I). In addition, at 26 hpf, dlx2a expression in the pharyngeal arches is also unchanged in cdon morphants as compared to uninjected controls (Supplemental Figure 2J–K). A small percentage of cdon morphants display some mild hypoplasia (16%) suggesting a slight developmental delay (Supplemental Figure 2B, E, H, L) and 3% of injected embryos have a moderate narrowing of the ethmoid plate cartilage (Supplemental Figure 2C, F, I, L). Together, these data suggest that cdon is dispensable for neural crest specification and cranial NCC migration, but is required for migration of trunk NCCs.
Figure 2. cdon is required for trunk neural crest migration.
Lateral views, anterior to the left except (B, F) which is a dorsal view with anterior to the left. (A–C, E–G) ISH for neural crest markers sox10 at 18 somites and crestin at 18 and 24 hpf in wildtype and cdon morphant embryos. (A–C) In wildtype embryos, NCCs migrate in streams ventrally from the neural tube, while (E–G) cdon morphant NCCs do not migrate far ventrally and appear to stall after migrating a short distance (distance migrated by the most ventral NCC in each of the first 7 migratory streams over the yolk extension; wildtype average= 146.9 μm, cdon MO average= 79.7μm; p=0.000009). In addition more NCCs are localized along the dorsal midline, suggesting a migratory defect (compared B to F; sox10: wildtype, n=40/41 normal migration where NCC streams migrate ventral to midline in at least 5 out of the first 7 streams above the yolk extension; cdon MO, n=33/39 with impaired migration; crestin: wildtype n=62/64 normal migration, cdon MO, n=21/25). (D, H) Imaging of melanocytes at 48 hpf displays a loss of ventral melanocytes in cdon morphants (average of 6.7 over yolk extension) compared to wildtype (average 30 over y.e., p=0.012), suggesting NCCs do not reach ventral positions. (I, J) Quantification of C, G and D, H.
To assure that the cdon morpholino is specific and the phenotypes observed were not due to non-specific cell death or off-target effects, we performed several control experiments to test the validity of the morpholinos used. We knocked down the expression of cdon with an ATG morpholino (as shown thus far) as well as a second splice-blocking morpholino targeted to the exon 2-intron 2 boundary of cdon, producing similar defects to the ATG morpholino in NCC migration and formation of ventral melanocytes (Supplemental Figure 3A–D, gel and quantification in E–G). In addition, a standard morpholino control injection was performed and we observed no phenotype in NCC migration (data not shown) and co-injection of the cdon ATG morpholino with a morpholino targeting p53 did not rescue the observed migratory defects (Supplemental Figure 3H–K, quantification in L). Additionally, to determine if cdon mRNA can rescue the NCC migration phenotype of the knockdown, we co-injected cdon MO with cdon mRNA and found a partial but significant rescue of the distance of ventrally migrated NCCs (Supplemental Figure 4). In sum, these data suggest that cdon has a specific role in directed NCC migration.
Cdon is required to regulate directedness of migration by the formation of stable and directed protrusions
To observe the behavior of NCCs in the absence of cdon, we performed live cell imaging in transgenic Tg(sox10:eGFP) embryos from 19–24 hpf, when trunk NCCs are actively migrating (Figure 3A, B; Supplemental Movie 1,2). While wildtype NCCs migrate in a stereotypic manner by migrating ventrally in narrow pathways along the somite in a directed fashion, the cdon morphant NCCs exhibit abnormal morphology and appear to be highly motile and less directed in getting to their destination (Figure 3A′, B′). Quantification of displacement of migratory cells suggests that the directedness of migration was significantly reduced in cdon morphants (Figure 3C), while the speed of cell migration (velocity in μm/second) is approximately twice as fast as wildtype NCCs (Figure 3D). In addition, several NCCs appear to exit the dorsal neural tube and delaminate but then stall in the migratory pathway and do not migrate further ventrally, supporting the idea that cdon is required for directed NCC migration.
Figure 3. The directedness of migration and cell protrusions is disrupted following cdon knockdown.
Live imaging of wildtype (A) and cdon morphant (B) sox10:eGFP embryos from 19–24 hpf shows failure of directed NCC migration in cdon morphants. A high-resolution image of migrating NCCs (at approximately 22 hpf) in wildtype and cdon morphants (insets to the right) displays abnormal cellular morphology and protrusions in cdon morphant embryos. (A′, B′) Inverse images of migrating tracks in both wildtype (A′) and morphant embryos (B′) displays a lack of directed migration along the anterior-posterior axis in cdon morphant embryos (n=6 cells for each), cells are in black and colored dashed lines represent a trace of each individual cell over the time period. (A″, B″) A plot of traces of same cells in A′ and B′ to illustrate the anterior-posterior movement and loss of directed migration in cdon morphant NCCs as compared to wildtype cells. (C, D) Quantification of Displacement index (C) and Velocity (D) using Volocity Software (PerkinElmer) suggests increased motility and decreased directedness in cdon morphant NCCs. Displacement is a measure of straight-line distance from the first position to the last position with a higher number indicative more distance between the first and last point. For wildtype cells (n=6) the displacement index was 58.9±28.27μm and in cdon MO (n=17) 18.68±14.45 μm; p=0.01. For velocity measurements, wildtype cells (n=6) were tracked at 0.0046±0.0015 μm/sec whereas cdon MO cells (n=17) tracked at 0.00931±0.0046 μm/sec; p=0.001. (E–G) Wildtype and cdon morphant NCC imaged in Tg(sox10:memRFP) over 120 minutes to determine the direction and number of cellular protrusions. (E) Wildtype cells have highly directed and stable protrusions (F). In the cdon morphant cells, the number and distribution of the protrusions is significantly increased as compared to wildtype cells (mean±sem = 4.4±0.81 vs 2.0±0.25, respectively; unpaired T-test, p = 0.0138), which gives the cells a more rounded appearance. (G) Rose plots show the distribution around a circle with 0° dorsal and 180° ventral. A circular T-test to quantify the directionality of the protrusions suggests that the wildtype cells are more directional then the cdon morphant NCCs and significantly departs from uniformity (length of red line, p<0.001) but the overall directionality (position of red line) is significantly different in the wildtype using the Rayleigh test of uniformity (p=1.21×10−8) and trending towards significance in the cdon morphant NCCs (p=0.021). Numbers in G are position around a circle from 0–360 and length of wedge indicates the number of protrusions (2, 4, 6) of a given cell.
To determine the role of cdon in forming stable and directed protrusions in NCCs as the migrate, required for directed NCC migration(Matthews et al., 2008), we imaged in a membrane tagged sox10:RFP line Tg(sox10:memRFP) which allowed us to visualize cellular protrusions in migrating cells. The number of cellular protrusions per cell is significantly increased in cdon morphant cells as compared to wildtype cells (mean ±sem = 4.4±0.81 vs 2.0±0.25, respectively; unpaired T- test, p = 0.0138). Moreover, the directionality of the protrusions was disrupted. cdon morphant cells extend short rapid protrusions in multiple directions while wildtype uninjected NCCs create stable ventrally positioned protrusions (Figure 3E–F Supplemental Movie 3, 4). A circular T-test to quantify the directionality of the protrusions suggests that the wildtype cells are more directional then the cdon morphant NCCs and significantly departs from uniformity (Figure 2 E, red line p<0.001). Together, these data suggest that cdon is required in premigratory NCCs for directed cell migration likely by regulating cell polarity and thereby the formation of stable and directed cell protrusions.
cdon is required cell-autonomously in trunk NCCs
To determine if cdon is required cell- or non-cell autonomously for NCC migration, we performed mosaic analysis by transplanting cdon morphant NCCs into a wildtype background. We injected cdon morpholino into Tg(sox10:eGFP) donor embryos and transplanted cells from the presumptive neural crest of donor embryos into the same region of uninjected tg(sox10:mRFP) host embryos. We then imaged embryos at 30 hpf, by which time the majority of wildtype (RFP+) host NCCs are within the migratory pathway (Figure 4B), while cdon morphant donor NCCs (GFP+) did not migrate from the dorsal neural tube within the trunk (Figure 4A, C). As mosaically transplanted wildtype cells had migrated to varying distances at the time of imaging, the number of NCCs migrated ventral to the neural tube was quantified in wildtype and cdon morphant transplanted cells (Figure 4F). cdon morphant cranial NCCs are able to migrate ventrally, supporting a specific role of cdon in trunk NCC migration (Figure 4G–I). Control experiments transplanting wildtype NCCs (sox10:eGFP donor cells) into either cdon morphant or wildtype hosts resulted in normal migration of donor cells along the host migratory pathway (Figure 4D–F), suggesting that cdon is a cell-autonomous regulator of trunk NCC migration but is not required within the migratory environment or in cranial NCC migration.
Figure 4. cdon is required cell-autonomously for NCC migration.
Lateral views of live transplanted embryos, anterior to the left. Both transplanted donor cells sox10:eGFP and wildtype sox10:RFP host NCCs are visualized in the same embryo. (A–C) Presumptive NCCs from cdon morphant sox10:eGFP embryos were transplanted into the presumptive neural plate border region of 4 hpf wildtype sox10:RFP host embryos and imaged at 30 hpf. Donor cells from cdon morphants (green) were unable to migrate in wildtype host background (red), suggesting that cdon is required cell autonomously for NCC migration. (B) WT into WT and (E) WT into cdon MO transplants were performed as controls (showing only transplanted cells in green). Transplanted donor wildtype NCCs in either a wildtype or cdon MO background migrate normally. The sox10:RFP expression in these embryos was too weak to visualize in live embryos as shown. (F) Quantification of the fraction of cells transplanted as compared to the total number migrating in all conditions. For the calculations, at least 5 embryos were used for quantification: cdon MO into wildtype, 34 trunk NCCs transplanted in 11 embryos, only 5 cells migrated ventrally beyond the dorsal neural tube, indicating that only 16.5% were able to migrate in a wildtype environment as compared to 97% of cdon morphant cranial NCCs (155 cells transplanted in 13 embryos). For WT into WT transplants, 28 transplanted trunk cells in 6 embryos and 72 cranial cells in 5 embryos migrated ventrally. For WT into MO, 30 transplanted trunk cells in 6 embryos and 108 cranial cells in 5 embryos. WT, wildtype uninjected embryo; p<0.05. (G–I) Donor cells from cdon morphants (green) were able to migrate in wildtype host background (red) in the cranial region, suggesting that cdon is required cell autonomously for trunk NCC migration but not cranial NCC migration.
cdon is regulated by the canonical Shh pathway
Because cdon is a Shh receptor, we examined the possibility that cdon functions within the Shh signaling pathway. We found by qRT-PCR and ISH that cdon expression is slightly but significantly increased in the dorsal neural tube following Shh inhibition with cyclopamine from 10–24 hpf, suggesting that cdon may function downstream of Shh signaling (Figure 5A–D, I). Consistent with this, overexpression of Shh by heat shock induction at 10 hpf using the Tg(hsp70l:shha-eGFP) zebrafish line (Shen et al., 2013) dramatically reduces cdon expression specifically within the dorsal neural tube/neural crest staging area (Figure 5E–H) suggesting that Shh may negatively regulate cdon expression. However, patched1/2 and gli1 expression are unchanged following cdon knockdown (Supplemental Figure 5) suggesting that cdon loss of function is not sufficient for alteration of canonical Shh signaling during trunk NCC migration. In addition, consistent with this and previous studies of cdon function, there is some disorganization of neural tube cells in cdon knockdown embryos (Cardozo et al., 2014). These data suggest that cdon is regulated by Shh perhaps to receive and interpret signals for NCC migration.
Figure 5. cdon mRNA expression is regulated by Shh.
cdon expression is slightly but significantly upregulated in the dorsal neural tube following Shh inhibition with cyclopamine as compared to control shown by ISH (A–D). Overexpression of Shh via heat shock results in loss of cdon expression. tg(hsp70l:shha-eGFP) embryos were heat shocked at 10 hpf and ISH was performed at 28 hpf for cdon expression. cdon is decreased within the dorsal trunk of Shh overexpressing embryos at 28 hpf (E–H; n=38 for heat shock; n=20 no heat shock). (I) qRT-PCR for cdon and patched 1/2 following Cyclopamine treatment show significant upregulation (p=0.0294) of cdon following Shh inhibition. Positive controls for Shh knockdown include patched1/2, a downstream readout of the canonical Shh signaling pathway, which is significantly reduced following cyclopamine treatment (p=0.0039). Gene expression was normalized to EF1a. Unpaired two tailed T-tests were performed in Graphpad (Prism).
cdon regulates N-cadherin localization in migrating NCCs
Another possible mechanism by which cdon functions to regulate NCC migration is by interacting with the adhesion protein N-cadherin (cdh2). To test this, we examined N-cadherin expression by immunofluorescence following cdon knockdown in Tg(sox10:eGFP) embryos. In wildtype embryos, N-cadherin localizes primarily to the cell surface of recently emigrated trunk NCCs, with expression resolving to the leading edge of the NCC membrane as migration proceeds ventrally along the neural tube (Figure 6A–C, G–I, N). In cdon morphants however, N-cadherin expression is seen both on the membrane and also diffusely throughout the cytoplasm of both recently emigrated and migratory NCCs and is no longer primarily localized to the cell membrane where it likely may act to regulate cell adhesion and/or protrusions (Figure 6D–F, J–L, O). Quantification of fluorescent intensity of N-cadherin in ImageJ using Plot Profile (Figure N, O) confirms that N-cadherin expression peaks at the leading edge cell membrane in migrating NCCs and this concentrated distribution is lost in cdon morphant NCCs. Profiles in cdon morphants were either at baseline levels across the cell or similar to that shown in Figure 5O, where expression is distributed throughout the cytoplasm. Total levels of N-cadherin expression were similar in wildtype and cdon morphant NCCs (data not shown). Ratios of fluorescent intensity of N-cadherin expression at the membrane or in the cytoplasm normalized to total intensity were then calculated and further demonstrate a shift in N-cadherin localization from the membrane to the cytoplasm in cdon morphant NCCs (Figure 5M). This suggests that Cdon regulates N-cadherin localization in trunk NCCs, possibly through binding N-cadherin and localizing it to the cell membrane, and that misregulation of N-cadherin localization results in misdirected NCC migration. While cdon mRNA is only expressed in premigratory cells, and quickly downregulated as cells start to migrate, we predict that the protein expression is maintained in NCCs throughout their migration. However, these data do suggest that there is already “priming” of migration before the cells leave the neural crest staging area, as polarity and adhesion proteins are relocalized for directed migration.
Figure 6. N-cadherin is mislocalized in cdon morphant NCCs.
Confocal images of cross-sectional immunofluorescent staining (12 μm sections, 1 μm optical sections) of the neural tube show N-cadherin localized to the cell surface of sox10:GFP-positive NCCs at 24 hpf (DAPI in blue, N-cadherin in red, sox10:GFP in green). (A–C) In wildtype embryos, N-cadherin is localized to the cell membrane leading edge, (D–F) while in cdon morphant NCCs, N-cadherin is evenly distributed in the cytoplasm. Insets in C, F show higher magnification of the triple labeled NCCs. (G–L) High magnification (63x) of a different group of neural crest cells in uninjected (G–I) and cdon MO injected (J–L) clearly demonstrating membrane localization in control neural crest cells while cdon knockdown cells have N-cadherin distributed evenly across the cell. (M–O) Fluorescent intensity of N-cadherin expression at the plasma membrane and cytoplasm was measured in ImageJ. Using plot profile, the midpoint of the cell was determined, and a line drawn through the center of the cell using the sox10-egfp positive cytoplasm to gate the membrane (ie at the edge of the gfp), shown as yellow line in inset, normalized from 0–1. From the line scan of fluorescence intensity, a line was then drawn down the center of the cell and ratios of membrane to total, cytoplasm to total, and membrane to cytoplasm were calculated for wildtype and cdon morphant NCCs; *p<0.05, **p<0.01. (N, O) Fluorescent intensity plot profile traces of migrating NCCs with ImageJ (NIH). In control cells, N-cadherin is expressed on the cell membrane and primarily on the leading edge membrane (n=13/16) while cdon MO cells show an even distribution across the cells (n=7/14) or a graded distribution (n= 7/14) but without correlation to leading verses trailing edges. NT, neural tube; *’s denote migratory NCCs, along with arrows labeling the leading edge of most ventral migrating cells. Scale bar is 12 μm in L.
Discussion
Here, we demonstrate a novel role for cdon in neural crest migration. Expression of cdon is initiated in the NPB prior to the 2-somite stage during neural crest specification. While it does not appear to play a role in specification, loss of cdon during early development impairs neural crest migration at later stages as defects in cdon morphants are not seen until NCCs are beginning to emigrate from the dorsal neural tube in the trunk. By in situ hybridization for neural crest markers, we find that NCCs are able to delaminate and begin migrating ventrally; however, they appear to stall just ventral to the neural tube, while wildtype NCCs continue migrating in streams to far ventral sites of differentiation. Indeed, when we analyze the appearance of neural crest derivatives such as melanocytes, this ventral-most population of NCCs never arrives to its final differentiation site on the ventral yolk. Interestingly, this defect is restricted primarily to trunk NCCs, suggesting a separate, cdon-independent mechanism for cranial NCC migration. This is not surprising as the migratory environment for cranial and trunk NCCs is likely very different and the mechanisms of directed cell migration appear to diverge in these two populations in zebrafish (Honjo and Eisen, 2005; Raible and Eisen, 1994, 1996; Schilling and Kimmel, 1994) as well as other vertebrates (Carmona-Fontaine et al., 2008; Gammill et al., 2006).
The role of cdon in NCC migration is most apparent when analyzing cell behaviors in live zebrafish embryos. Here, we see an increase in cell motility as evidenced by increased velocity and cell protrusions in cdon morphants. We also observe decreased directed cell migration by the conventional displacement measure, which could be due to defects in cell polarity. Together, this suggests that cdon may be playing multiple roles in cell migration, both by regulating cell motility and by regulating cell polarity that is necessary for directed migration. There are several potential cellular mechanisms that are regulated by cdon. One interesting possibility is that Cdon interacts directly with N-cadherin to regulate directed migration. In myogenesis, cdon has been shown to directly bind and ligate N-cadherin, which contributes to muscle cell differentiation (Gavard et al., 2004; Goichberg and Geiger, 1998; Lu and Krauss, 2010). As regulation of adherens junctions and cell-cell contact is essential for proper control of NCC EMT and migration, it is possible that cdon is involved in the regulation of N-cadherin localization and/or other cadherin proteins. It is apparent in cdon morphant NCCs that N-cadherin is mislocalized from the cell membrane to the cytoplasm during migration of trunk NCCs. This mislocalization of N-cadherin alone could be responsible for the phenotype observed in to cdon knockdown embryos, as loss of N-cadherin results in a very similar loss of directed migration phenotype in Xenopus NCCs (Theveneau et al., 2010). In addition, N-cadherin plays an important role in cell polarization as well (for example see (Dupin et al., 2009). As the Cdon protein can bind N-cadherin directly, it is conceivable that Cdon is required to regulate the localization of N-cadherin and possibly other cadherin proteins during NCC migration, perhaps by trafficking of cadherins to the cell membrane prior to migration. In support of this interaction, previous studies have suggested that N-cadherin and Cdon bind in cis at the plasma membrane of cells (Kang et al., 2003; Lu and Krauss, 2010), suggesting that Cdon and N-cadherin may bind at the membrane of premigratory NCCs. Another possibility is that Cdon is required to bind N-cadherin at the surface of adjacent neighboring NCCs to mediate cell-cell adhesion or contact-dependent signaling between premigratory NCCs within the dorsal neural tube. We have shown cdon functions cell autonomously in trunk NCC migration, however our analysis cannot differentiate which NCCs cdon is required in, the primary cell or a secondary neighboring migratory NCC. The misregulation of N-cadherin localization upon loss of cdon could contribute to premature delamination of NCCs within the dorsal neural tube, demonstrated by the increased motility and aberrant migration of NCCs dorsally and the ectopic movement of NCCs across the midline of the embryo that is observed in cdon morphant embryos. As the NCCs lose their cell-cell contacts and migrate within or dorsal to the neural tube, it is possible that they are no longer able to properly polarize or communicate with other NCCs and cells of the migratory environment, preventing the NCCs from migrating ventrally to their final destinations. Finally, the mislocalization of N-cadherin away from the membrane could cause a loss of downstream N-cadherin mediated signaling such as through p38/MAPK. The involvement of such signals during neural crest migration is still poorly understood and will be the focus of further investigation. Future experiments are required to addess the cellular localizaton and dynamics of Cdon protein in vivo to further elucidate the mechanism by which it regulates NCC motility.
Another role of cdon is in the Sonic hedgehog (Shh) signaling pathway. Cdon protein has been shown to bind directly to the Shh ligand, possibly acting as a co-receptor for Patched1 (Tenzen et al., 2006; Yao et al., 2006). In limb mesenchymal cells, Cdon localizes to microdomains within long, filopodial protrusions, which are thought to be important in the transport of Shh signaling components for long-range signaling events (Sanders et al., 2013). While the role of Shh signaling in neural crest migration is not well established in the trunk, it is required in zebrafish cranial NCC development (Delloye-Bourgeois et al., 2014; Eberhart et al., 2006; Wada et al., 2005). We observed regulation of cdon expression following manipulation of the Shh pathway via knockdown with cyclopamine or heat shock to overexpress Shh. This suggests that cdon can be regulated by Shh. It is possible then that Cdon is required to receive and interpret Shh signals during NCC migration and EMT, but this interaction remains unclear. Interestingly, there is some evidence that Shh may play a role in the migration of NCCs and some of their derivatives including cells of the dorsal root ganglia (Ungos et al., 2003). Furthermore, Shh plays a role in EMT of some cell types and provides positional cues in cranial and enteric NCCs (Mo et al., 2001; Ramalho-Santos et al., 2000) (Tobin et al., 2008)as well as mediating migration of NCCs on fibronectin (Testaz et al., 1999; Testaz et al., 2001). Because this activity does not require Patched or Smoothened, it is possible that the unknown receptor for this interaction is Cdon.
Cdon is a conserved cell-surface protein which is expressed in the developing mouse neural folds, NPB, and premigratory NCCs (Mulieri et al., 2000). Cdon−/− mice exhibit microform holoprosencephaly consisting of defects in some but not all cranial neural crest derivatives (Cole and Krauss, 2003), although it is difficult to determine the specific effects due to the severity of the holoprosencephalic defects (R. Krauss, personal communication). However trunk NCC formation, migration, and various derivatives were not specifically analyzed. Thus, it is possible that cdon has similar roles in mouse neural crest migration as it does in zebrafish. Recent evidence suggests that cdon functions as a dependence receptor for NCC survival in chick branchial arch development (Delloye-Bourgeois et al., 2014). In our study, we did not observe increased cell death or significant cranial neural crest phenotypes, although cdon morphants did on occasion display mild hypoplasia of craniofacial cartilages that is likely due to a developmental delay. Cdon has also recently been implicated as a decoy receptor for Shh in zebrafish optic vesicle development. cdon negatively regulates Shh signaling and also functions in a Patched-independent fashion to pattern the eye in a proximo-distal fashion (Cardozo et al., 2014). This dual function of cdon is interesting as we also believe that cdon plays multiple roles in NCC development: regulating cell adhesion and polarity, and signaling perhaps via the Shh pathway. In this way, cdon serves a multifunctional platform for cell migration. Additionally, there are several putative binding partners with Cdon, such as the related protein BOC, which have yet to be studied in the context of neural crest development. Interestingly, human CDON has been identified as a pro-apoptotic tumor suppressor which is downregulated in colon, kidney, lung, and breast cancer tumor cells of human patients (Delloye-Bourgeois et al., 2013). In prostate cancer models, Cdon was found to be upregulated and knockdown of Cdon impaired tumor cell invasion and survival (Hayashi et al., 2011). Mutations in Cdon are also associated with human holoprosencephaly, supporting its identified role in mouse craniofacial development (Bae et al., 2011). This suggests that some of the roles of Cdon are conserved across both zebrafish and mammals and that Cdon may be an important regulator of NCC development in humans as well. Our studies presented here demonstrate a novel role for cdon in embryonic neural crest migration and provide new insight into the mechanism of regulation of adhesion through N-cadherin in migratory NCCs, as well as potential insight into other migratory cells including cancer cells.
Supplementary Material
ISH for neural crest specification markers (A, B) foxd3 and (C, D) sox10 at 11 hpf (2 somites) and 13 hpf (4 somites) shows no change in expression between wildtype and cdon morphants, demonstrating that cdon does not regulate neural crest specification. n=37/38 for foxd3 and 47/48 for sox10.
Live imaging of sox10:eGFP positive NCCs from 19–24 hpf.
(A–I) Alcian blue staining of cartilage at 5 dpf shows normal formation of cartilage in uninjected wildtype and cdon morphant embryos. (B, E, H) In a small percentage of cdon morphants, mild hypoplasia was observed as well as (C, F, I) moderate ethmoid plate narrowing. (J, K) dlx2a expression, a marker of pharyngeal arch NCCs at 24 hpf is also not affected by cdon knockdown. (L) Embryos were classified by phenotype as exhibiting either normal formation, mild hypoplasia, or narrowing of the ethmoid plate. cdon morphants are mostly normal with a small percentage displaying some mild craniofacial defects.
(A, C, H, J) ISH for neural crest marker crestin and (B, D, I, K) melanocyte formation following cdon knockdown with splice blocking MO E2I2, or co-injected with p53 MO. In both, crestin expression and melanocyte formation is affected and is similar to the ATG MO shown in Figure 2. Distance migrated by e2i2 morphant NCCs of the first 7 somites over the yolk extension averaged 65.5μm compared to 108μm in wildtype, * p<0.0001, quantification in F; number of melanocytes over yolk extension=4.8 in e2i2 MO vs 22.8 in wildtype controls, *p=0.006, quantification in G. p53 MO was unable to rescue migration of cdon ATG MO NCCs (average distance migrated in p53+cdon MO=79.4μm compared to 143.7μm in p53 MO alone, p<0.0001, quantification in L.) (E) RT-PCR for splicing of cdon following injection of 6 or 12 ng of cdon MO shows multiple sized bands representing mis-splicing of cdon RNA. Wildtype band exhibiting normal splicing is amplified at 215 bp while mis-splicing and subsequent inclusion of intron 2 results in a band shift to 465–600 bp.
(A–C) Rescue of crestin expression at 24 hpf following cdon mRNA and MO co-injection. While injection of cdon ATG MO alone resulted in aberrant NCC migration (Figure 2), co-injection with 120–220 pg of cdon mRNA resulted in the development of a greater number of embryos that had migratory streams and do not display the cdon morphant phenotype (n=12/119 or 10% compared to 84% morphant phenotype observed in MO injected alone.) (A′–C′) Higher magnification of NCC migratory streams in all conditions. The partial rescue of the cdon morphant phenotype by co-injection with cdon mRNA suggests that the cdon ATG MO specifically targets cdon. (D) Quantification of the average distance migrated by NCCs: wildtype=125.4μm, cdon MO=78.8μm, cdon MO+120pg cdon mRNA=110μm, cdon MO+220pg cdon mRNA=111.5μm;* p<0.0001 by ANOVA.
(A–E) patched1/2 and gli1 expression following cdon knockdown. (F–I) ISH of cdon MO with patched1 in whole mount and 12 μm sections shows no change in patched1 expression. (J) Similarly, qRT-PCR for both patched1/2 and gli1 also shows no change in expression. Statistics were performed with a Students T-test, error bars are standard error of the mean (sem).
Live imaging of sox10:eGFP positive NCCs from 19–24 hpf in cdon MO embryos.
Live imaging of sox10:memRFP positive NCCs from 19–25 hpf. Movie shows 120 minutes from the start of the imaging period.
Live imaging of sox10:memRFP positive NCCs from 19–24 hpf in cdon morphant embryos.
Highlights.
cdon is expressed in neural crest precursors and the neural crest staging area
cdon is required for directed neural crest migration by regulating directed cell protrusions
cdon regulates the localization of N-cadherin, suggesting a mechanism for how directed migration is regulated
Acknowledgments
We thank past and present members of the Artinger lab, especially Christy Cortez Rossi for help with some of the initial expression analyses. We also thank Morgan Singleton for excellent fish care, Laura Hudish and Alex Blasky in the Appel lab for help with live cell imaging and Shh regents and analysis, Robert Krauss and Paola Bovolenta for cdon reagents and discussion of function, and Linda Barlow for insightful reading of this manuscript. MNCD2 antibody developed by M. Takeichi and H. Matsunami was obtained from the Developmental Studies Hybridoma Bank, created by the NICHD of the NIH and maintained at The University of Iowa, Department of Biology, Iowa City, IA 52242. This work is supported by NIDCR pre-doctoral fellowship F31DE022237 to D.R.P., and NIH grant R01DE017699 to K.B.A. and P30NS048154 to the UC Denver zebrafish core facility.
Footnotes
Conflict of Interest Statement
The authors declare no competing financial interests.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- Bae GU, Domene S, Roessler E, Schachter K, Kang JS, Muenke M, Krauss RS. Mutations in CDON, encoding a hedgehog receptor, result in holoprosencephaly and defective interactions with other hedgehog receptors. Am J Hum Genet. 2011;89:231–240. doi: 10.1016/j.ajhg.2011.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Belmadani A, Tran PB, Ren D, Assimacopoulos S, Grove EA, Miller RJ. The chemokine stromal cell-derived factor-1 regulates the migration of sensory neuron progenitors. J Neurosci. 2005;25:3995–4003. doi: 10.1523/JNEUROSCI.4631-04.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Blasky AJ, Pan L, Moens CB, Appel B. Pard3 regulates contact between neural crest cells and the timing of Schwann cell differentiation but is not essential for neural crest migration or myelination. Dev Dyn. 2014;243:1511–23. doi: 10.1002/dvdy.24172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cardozo MJ, Sanchez-Arrones L, Sandonis A, Sanchez-Camacho C, Gestri G, Wilson SW, Guerrero I, Bovolenta P. Cdon acts as a Hedgehog decoy receptor during proximal-distal patterning of the optic vesicle. Nat Commun. 2014;5:4272. doi: 10.1038/ncomms5272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Carmona-Fontaine C, Matthews HK, Kuriyama S, Moreno M, Dunn GA, Parsons M, Stern CD, Mayor R. Contact inhibition of locomotion in vivo controls neural crest directional migration. Nature. 2008;456:957–961. doi: 10.1038/nature07441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Carmona-Fontaine C, Theveneau E, Tzekou A, Tada M, Woods M, Page KM, Parsons M, Lambris JD, Mayor R. Complement fragment C3a controls mutual cell attraction during collective cell migration. Dev Cell. 2011;21:1026–1037. doi: 10.1016/j.devcel.2011.10.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Cole F, Krauss RS. Microform holoprosencephaly in mice that lack the Ig superfamily member Cdon. Curr Biol. 2003;13:411–415. doi: 10.1016/s0960-9822(03)00088-5. [DOI] [PubMed] [Google Scholar]
- Davy A, Soriano P. Ephrin signaling in vivo: look both ways. Dev Dyn. 2005;232:1–10. doi: 10.1002/dvdy.20200. [DOI] [PubMed] [Google Scholar]
- Delloye-Bourgeois C, Gibert B, Rama N, Delcros JG, Gadot N, Scoazec JY, Krauss R, Bernet A, Mehlen P. Sonic Hedgehog promotes tumor cell survival by inhibiting CDON pro-apoptotic activity. PLoS Biol. 2013;11:e1001623. doi: 10.1371/journal.pbio.1001623. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Delloye-Bourgeois C, Rama N, Brito J, Le Douarin N, Mehlen P. Sonic Hedgehog promotes the survival of neural crest cells by limiting apoptosis induced by the dependence receptor CDON during branchial arch development. Biochemical and biophysical research communications. 2014 Sep 26;452(3):655–60. doi: 10.1016/j.bbrc.2014.08.134. [DOI] [PubMed] [Google Scholar]
- Dupin I, Camand E, Etienne-Manneville S. Classical cadherins control nucleus and centrosome position and cell polarity. J Cell Biol. 2009;185:779–786. doi: 10.1083/jcb.200812034. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dutton JR, Antonellis A, Carney TJ, Rodrigues FS, Pavan WJ, Ward A, Kelsh RN. An evolutionarily conserved intronic region controls the spatiotemporal expression of the transcription factor Sox10. BMC Dev Biol. 2008;8:105. doi: 10.1186/1471-213X-8-105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eberhart JK, Swartz ME, Crump JG, Kimmel CB. Early Hedgehog signaling from neural to oral epithelium organizes anterior craniofacial development. Development. 2006;133:1069–1077. doi: 10.1242/dev.02281. [DOI] [PubMed] [Google Scholar]
- Gammill LS, Gonzalez C, Gu C, Bronner-Fraser M. Guidance of trunk neural crest migration requires neuropilin 2/semaphorin 3F signaling. Development. 2006;133:99–106. doi: 10.1242/dev.02187. [DOI] [PubMed] [Google Scholar]
- Gavard J, Marthiens V, Monnet C, Lambert M, Mege RM. N-cadherin activation substitutes for the cell contact control in cell cycle arrest and myogenic differentiation: involvement of p120 and beta-catenin. J Biol Chem. 2004;279:36795–36802. doi: 10.1074/jbc.M401705200. [DOI] [PubMed] [Google Scholar]
- Goichberg P, Geiger B. Direct involvement of N-cadherin-mediated signaling in muscle differentiation. Mol Biol Cell. 1998;9:3119–3131. doi: 10.1091/mbc.9.11.3119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hayashi T, Oue N, Sakamoto N, Anami K, Oo HZ, Sentani K, Ohara S, Teishima J, Matsubara A, Yasui W. Identification of transmembrane protein in prostate cancer by the Escherichia coli ampicillin secretion trap: expression of CDON is involved in tumor cell growth and invasion. Pathobiology. 2011;78:277–284. doi: 10.1159/000329588. [DOI] [PubMed] [Google Scholar]
- Hayhurst M, McConnell SK. Mouse models of holoprosencephaly. Curr Opin Neurol. 2003;16:135–141. doi: 10.1097/01.wco.0000063761.15877.40. [DOI] [PubMed] [Google Scholar]
- Honjo Y, Eisen JS. Slow muscle regulates the pattern of trunk neural crest migration in zebrafish. Development. 2005;132:4461–4470. doi: 10.1242/dev.02026. [DOI] [PubMed] [Google Scholar]
- Izzi L, Levesque M, Morin S, Laniel D, Wilkes BC, Mille F, Krauss RS, McMahon AP, Allen BL, Charron F. Boc and Gas1 each form distinct Shh receptor complexes with Ptch1 and are required for Shh-mediated cell proliferation. Dev Cell. 2011;20:788–801. doi: 10.1016/j.devcel.2011.04.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kang JS, Feinleib JL, Knox S, Ketteringham MA, Krauss RS. Promyogenic members of the Ig and cadherin families associate to positively regulate differentiation. Proc Natl Acad Sci U S A. 2003;100:3989–3994. doi: 10.1073/pnas.0736565100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF. Stages of embryonic development of the zebrafish. Dev Dyn. 1995;203:253–310. doi: 10.1002/aja.1002030302. [DOI] [PubMed] [Google Scholar]
- Krull CE, Lansford R, Gale NW, Collazo A, Marcelle C, Yancopoulos GD, Fraser SE, Bronner-Fraser M. Interactions of Eph-related receptors and ligands confer rostrocaudal pattern to trunk neural crest migration. Curr Biol. 1997;7:571–580. doi: 10.1016/s0960-9822(06)00256-9. [DOI] [PubMed] [Google Scholar]
- Kuriyama S, Mayor R. Molecular analysis of neural crest migration. Philos Trans R Soc Lond B Biol Sci. 2008;363:1349–1362. doi: 10.1098/rstb.2007.2252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Le Douarin NM. The neural crest. Cambridge Univ Press; New York: 1982. [Google Scholar]
- Lele Z, Folchert A, Concha M, Rauch GJ, Geisler R, Rosa F, Wilson SW, Hammerschmidt M, Bally-Cuif L. parachute/n-cadherin is required for morphogenesis and maintained integrity of the zebrafish neural tube. Development. 2002;129:3281–3294. doi: 10.1242/dev.129.14.3281. [DOI] [PubMed] [Google Scholar]
- Lu M, Krauss RS. N-cadherin ligation, but not Sonic hedgehog binding, initiates Cdo-dependent p38alpha/beta MAPK signaling in skeletal myoblasts. Proc Natl Acad Sci U S A. 2010;107:4212–4217. doi: 10.1073/pnas.0908883107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Matthews HK, Marchant L, Carmona-Fontaine C, Kuriyama S, Larrain J, Holt MR, Parsons M, Mayor R. Directional migration of neural crest cells in vivo is regulated by Syndecan-4/Rac1 and non-canonical Wnt signaling/RhoA. Development. 2008;135:1771–1780. doi: 10.1242/dev.017350. [DOI] [PubMed] [Google Scholar]
- Mo R, Kim JH, Zhang J, Chiang C, Hui CC, Kim PC. Anorectal malformations caused by defects in sonic hedgehog signaling. Am J Pathol. 2001;159:765–774. doi: 10.1016/S0002-9440(10)61747-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Monier-Gavelle F, Duband JL. Control of N-cadherin-mediated intercellular adhesion in migrating neural crest cells in vitro. J Cell Sci. 1995;108 (Pt 12):3839–3853. doi: 10.1242/jcs.108.12.3839. [DOI] [PubMed] [Google Scholar]
- Mulieri PJ, Okada A, Sassoon DA, McConnell SK, Krauss RS. Developmental expression pattern of the cdo gene. Dev Dyn. 2000;219:40–49. doi: 10.1002/1097-0177(2000)9999:9999<::AID-DVDY1032>3.0.CO;2-M. [DOI] [PubMed] [Google Scholar]
- Olesnicky E, Hernandez-Lagunas L, Artinger KB. prdm1a Regulates sox10 and islet1 in the development of neural crest and Rohon-Beard sensory neurons. Genesis. 2010;48:656–666. doi: 10.1002/dvg.20673. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Olesnicky Killian EC, Birkholz DA, Artinger KB. A role for chemokine signaling in neural crest cell migration and craniofacial development. Dev Biol. 2009;333:161–172. doi: 10.1016/j.ydbio.2009.06.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pineda RH, Svoboda KR, Wright MA, Taylor AD, Novak AE, Gamse JT, Eisen JS, Ribera AB. Knockdown of Nav1.6a Na+ channels affects zebrafish motoneuron development. Development. 2006;133:3827–3836. doi: 10.1242/dev.02559. [DOI] [PubMed] [Google Scholar]
- Powell DR, Hernandez-Lagunas L, LaMonica K, Artinger KB. Prdm1a directly activates foxd3 and tfap2a in zebrafish neural crest specification. Development. 2013;140(16):3445–55. doi: 10.1242/dev.096164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Raible DW, Eisen JS. Restriction of neural crest cell fate in the trunk of the embryonic zebrafish. Development. 1994;120:495–503. doi: 10.1242/dev.120.3.495. [DOI] [PubMed] [Google Scholar]
- Raible DW, Eisen JS. Regulative interactions in zebrafish neural crest. Development. 1996;122:501–507. doi: 10.1242/dev.122.2.501. [DOI] [PubMed] [Google Scholar]
- Raible DW, Wood A, Hodsdon W, Henion PD, Weston JA, Eisen JS. Segregation and early dispersal of neural crest cells in the embryonic zebrafish. Dev Dyn. 1992;195:29–42. doi: 10.1002/aja.1001950104. [DOI] [PubMed] [Google Scholar]
- Ramalho-Santos M, Melton DA, McMahon AP. Hedgehog signals regulate multiple aspects of gastrointestinal development. Development. 2000;127:2763–2772. doi: 10.1242/dev.127.12.2763. [DOI] [PubMed] [Google Scholar]
- Rubinstein AL, Lee D, Luo R, Henion PD, Halpern ME. Genes dependent on zebrafish cyclops function identified by AFLP differential gene expression screen. Genesis. 2000;26:86–97. doi: 10.1002/(sici)1526-968x(200001)26:1<86::aid-gene11>3.0.co;2-q. [DOI] [PubMed] [Google Scholar]
- Sanders TA, Llagostera E, Barna M. Specialized filopodia direct long-range transport of SHH during vertebrate tissue patterning. Nature. 2013;497:628–632. doi: 10.1038/nature12157. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schilling TF, Kimmel CB. Segment and cell type lineage restrictions during pharyngeal arch development in the zebrafish embryo. Development. 1994;120:483–494. doi: 10.1242/dev.120.3.483. [DOI] [PubMed] [Google Scholar]
- Schwarz Q, Maden CH, Davidson K, Ruhrberg C. Neuropilin-mediated neural crest cell guidance is essential to organise sensory neurons into segmented dorsal root ganglia. Development. 2009a;136:1785–1789. doi: 10.1242/dev.034322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Schwarz Q, Maden CH, Vieira JM, Ruhrberg C. Neuropilin 1 signaling guides neural crest cells to coordinate pathway choice with cell specification. Proc Natl Acad Sci U S A. 2009b;106:6164–6169. doi: 10.1073/pnas.0811521106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shen MC, Ozacar AT, Osgood M, Boeras C, Pink J, Thomas J, Kohtz JD, Karlstrom R. Heat-shock-mediated conditional regulation of hedgehog/gli signaling in zebrafish. Dev Dyn. 2013;242:539–549. doi: 10.1002/dvdy.23955. [DOI] [PubMed] [Google Scholar]
- Tenzen T, Allen BL, Cole F, Kang JS, Krauss RS, McMahon AP. The cell surface membrane proteins Cdo and Boc are components and targets of the Hedgehog signaling pathway and feedback network in mice. Dev Cell. 2006;10:647–656. doi: 10.1016/j.devcel.2006.04.004. [DOI] [PubMed] [Google Scholar]
- Testaz S, Delannet M, Duband J. Adhesion and migration of avian neural crest cells on fibronectin require the cooperating activities of multiple integrins of the (beta)1 and (beta)3 families. J Cell Sci. 1999;112 (Pt 24):4715–4728. doi: 10.1242/jcs.112.24.4715. [DOI] [PubMed] [Google Scholar]
- Testaz S, Jarov A, Williams KP, Ling LE, Koteliansky VE, Fournier-Thibault C, Duband JL. Sonic hedgehog restricts adhesion and migration of neural crest cells independently of the Patched- Smoothened-Gli signaling pathway. Proc Natl Acad Sci U S A. 2001;98:12521–12526. doi: 10.1073/pnas.221108698. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Theveneau E, Marchant L, Kuriyama S, Gull M, Moepps B, Parsons M, Mayor R. Collective chemotaxis requires contact-dependent cell polarity. Dev Cell. 2010;19:39–53. doi: 10.1016/j.devcel.2010.06.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Theveneau E, Mayor R. Neural crest delamination and migration: from epithelium-to-mesenchyme transition to collective cell migration. Dev Biol. 2012;366:34–54. doi: 10.1016/j.ydbio.2011.12.041. [DOI] [PubMed] [Google Scholar]
- Theveneau E, Steventon B, Scarpa E, Garcia S, Trepat X, Streit A, Mayor R. Chase-and-run between adjacent cell populations promotes directional collective migration. Nat Cell Biol. 2013;15:763–772. doi: 10.1038/ncb2772. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tobin JL, Di Franco M, Eichers E, May-Simera H, Garcia M, Yan J, Quinlan R, Justice MJ, Hennekam RC, Briscoe J, Tada M, Mayor R, Burns AJ, Lupski JR, Hammond P, Beales PL. Inhibition of neural crest migration underlies craniofacial dysmorphology and Hirschsprung’s disease in Bardet-Biedl syndrome. Proc Natl Acad Sci U S A. 2008;105:6714–6719. doi: 10.1073/pnas.0707057105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ungos JM, Karlstrom RO, Raible DW. Hedgehog signaling is directly required for the development of zebrafish dorsal root ganglia neurons. Development. 2003;130:5351–5362. doi: 10.1242/dev.00722. [DOI] [PubMed] [Google Scholar]
- Wada N, Javidan Y, Nelson S, Carney TJ, Kelsh RN, Schilling TF. Hedgehog signaling is required for cranial neural crest morphogenesis and chondrogenesis at the midline in the zebrafish skull. Development. 2005;132:3977–3988. doi: 10.1242/dev.01943. [DOI] [PubMed] [Google Scholar]
- Westerfield M. The zebrafish book: a guide for the laboratory use of zebrafish (Brachydanio rerio) M. Westerfield; Eugene, OR: 1993. [Google Scholar]
- Yao S, Lum L, Beachy P. The ihog cell-surface proteins bind Hedgehog and mediate pathway activation. Cell. 2006;125:343–357. doi: 10.1016/j.cell.2006.02.040. [DOI] [PubMed] [Google Scholar]
- Yu HH, Moens CB. Semaphorin signaling guides cranial neural crest cell migration in zebrafish. Dev Biol. 2005;280:373–385. doi: 10.1016/j.ydbio.2005.01.029. [DOI] [PubMed] [Google Scholar]
- Zhang W, Kang JS, Cole F, Yi MJ, Krauss RS. Cdo functions at multiple points in the Sonic Hedgehog pathway, and Cdo-deficient mice accurately model human holoprosencephaly. Dev Cell. 2006;10:657–665. doi: 10.1016/j.devcel.2006.04.005. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
ISH for neural crest specification markers (A, B) foxd3 and (C, D) sox10 at 11 hpf (2 somites) and 13 hpf (4 somites) shows no change in expression between wildtype and cdon morphants, demonstrating that cdon does not regulate neural crest specification. n=37/38 for foxd3 and 47/48 for sox10.
Live imaging of sox10:eGFP positive NCCs from 19–24 hpf.
(A–I) Alcian blue staining of cartilage at 5 dpf shows normal formation of cartilage in uninjected wildtype and cdon morphant embryos. (B, E, H) In a small percentage of cdon morphants, mild hypoplasia was observed as well as (C, F, I) moderate ethmoid plate narrowing. (J, K) dlx2a expression, a marker of pharyngeal arch NCCs at 24 hpf is also not affected by cdon knockdown. (L) Embryos were classified by phenotype as exhibiting either normal formation, mild hypoplasia, or narrowing of the ethmoid plate. cdon morphants are mostly normal with a small percentage displaying some mild craniofacial defects.
(A, C, H, J) ISH for neural crest marker crestin and (B, D, I, K) melanocyte formation following cdon knockdown with splice blocking MO E2I2, or co-injected with p53 MO. In both, crestin expression and melanocyte formation is affected and is similar to the ATG MO shown in Figure 2. Distance migrated by e2i2 morphant NCCs of the first 7 somites over the yolk extension averaged 65.5μm compared to 108μm in wildtype, * p<0.0001, quantification in F; number of melanocytes over yolk extension=4.8 in e2i2 MO vs 22.8 in wildtype controls, *p=0.006, quantification in G. p53 MO was unable to rescue migration of cdon ATG MO NCCs (average distance migrated in p53+cdon MO=79.4μm compared to 143.7μm in p53 MO alone, p<0.0001, quantification in L.) (E) RT-PCR for splicing of cdon following injection of 6 or 12 ng of cdon MO shows multiple sized bands representing mis-splicing of cdon RNA. Wildtype band exhibiting normal splicing is amplified at 215 bp while mis-splicing and subsequent inclusion of intron 2 results in a band shift to 465–600 bp.
(A–C) Rescue of crestin expression at 24 hpf following cdon mRNA and MO co-injection. While injection of cdon ATG MO alone resulted in aberrant NCC migration (Figure 2), co-injection with 120–220 pg of cdon mRNA resulted in the development of a greater number of embryos that had migratory streams and do not display the cdon morphant phenotype (n=12/119 or 10% compared to 84% morphant phenotype observed in MO injected alone.) (A′–C′) Higher magnification of NCC migratory streams in all conditions. The partial rescue of the cdon morphant phenotype by co-injection with cdon mRNA suggests that the cdon ATG MO specifically targets cdon. (D) Quantification of the average distance migrated by NCCs: wildtype=125.4μm, cdon MO=78.8μm, cdon MO+120pg cdon mRNA=110μm, cdon MO+220pg cdon mRNA=111.5μm;* p<0.0001 by ANOVA.
(A–E) patched1/2 and gli1 expression following cdon knockdown. (F–I) ISH of cdon MO with patched1 in whole mount and 12 μm sections shows no change in patched1 expression. (J) Similarly, qRT-PCR for both patched1/2 and gli1 also shows no change in expression. Statistics were performed with a Students T-test, error bars are standard error of the mean (sem).
Live imaging of sox10:eGFP positive NCCs from 19–24 hpf in cdon MO embryos.
Live imaging of sox10:memRFP positive NCCs from 19–25 hpf. Movie shows 120 minutes from the start of the imaging period.
Live imaging of sox10:memRFP positive NCCs from 19–24 hpf in cdon morphant embryos.