|Home | About | Journals | Submit | Contact Us | Français|
The cardiac neural crest cells (cNCCs) and the second heart field (SHF) play key roles in development of the cardiac outflow tract (OFT) for establishment of completely separated pulmonary and systemic circulations in vertebrates. A neurovascular guiding factor, Semaphorin 3c (Sema3c), is required for the development of the OFT, however, its regulation of the interaction between cNCCs and SHF remains to be determined. Here, we show that a Sema3c is a candidate that mediates interaction between cNCCs and the SHF during development of the OFT. Foxc1/c2 directly activates the transcription of Sema3c in the OFT, whereas, a hypomorph of Tbx1, a key SHF transcription factor, resulted in the ectopic expression of Sema3c in the pharyngeal arch region. Fgf8, a downstream secreted factor of Tbx1, inhibited the expression of Sema3c in cNCCs via activation of ERK1/2 signaling. Blocking of FGF8 caused ectopic expression of SEMA3C and a migration defect of cNCCs, resulting in abnormal chick pharyngeal arch development. These results suggest that proper spatio-temporal expression of Sema3c, regulated positively by Foxc1/c2 and negatively by the Tbx1-Fgf8 cascade, respectively, is essential for the interaction between cNCCs and the SHF that correctly navigates cNCCs towards the OFT, composed of SHF-derived cells.
The neural crest is a multipotent and transient migratory lineage that gives rise to various different cell types1. Neural crest cells present in the region from the mid-otic placode to the caudal limit of somite 3 are referred to as cardiac neural crest cells (cNCCs)2. cNCCs migrate ventrally to form the third, fourth, and sixth pharyngeal arch arteries. They migrate further into the cardiac outflow tract (OFT) to give rise to the septum which divides a single great vessel arising from the embryonic heart, the truncus arteriosus, into the aorta and pulmonary trunk3.
Cardiac progenitor cells from a second source, termed the second heart field (SHF), are required in addition to cNCCs for formation of the OFT. Progenitor cells derived from the SHF in the pharyngeal mesoderm migrate into the embryonic heart tube from both the arterial and venous poles where they subsequently construct the OFT and atria, respectively4–6. Numerous transcription factors, including Isl1, Foxc1, Foxc2, and Tbx1, and signaling molecules including Fgfs, Bmps, and Wnts, have been reported to play a role in development of the SHF7.
In a recent study, it was found that Wnt1-positive neural crest cells form part of the intrapericardial aortic trunk, whereas lineages positive for the SHF-marker transcription factor Six2 give rise to the intrapericardial pulmonary trunk. This suggests that the aortic and pulmonary trunks are derived from cNCCs and SHF progenitors, respectively8. Because cNCCs and the SHF are essential for proper development of the OFT9, understanding the mechanisms that regulate the development of cNCCs and their interactions with the SHF will allow these findings to be applied in clinical practice.
Class 3 semaphorins are secreted proteins that act as axon repellents or attractants, controlling the formation of neuronal connections. They are also expressed in non-neuronal tissues and regulate cardiac morphogenesis and angiogenesis10. Semaphorin 3c (Sema3c) is a neurovascular guiding molecule of the class 3 semaphorin family that is implicated in cardiac development. Mice lacking Sema3c exhibit interruption of the aortic arch and persistent truncus arteriosus; thus Sema3c is thought to regulate the development and migration of cNCCs11. During the early stages of heart development, Sema3c is expressed in the OFT as well as in the pharyngeal arch region that contains cardiac progenitor niches composed of SHF progenitor cells and cNCCs12. Previous studies showed that a transcription factor, Gata6, regulates the expression of Sema3c and its receptor plexin A2, during development of the OFT13, 14. Taken together, Sema3c signaling from the OFT may be essential for migration of cNCCs towards the OFT; however, regulation of Sema3c expression is yet to be elucidated.
Here, we show that Foxc1/Foxc2 and Tbx1-Fgf8 signaling are essential positive and negative regulators of Sema3c expression, respectively, and contribute to the development and migration of cNCCs for OFT formation during embryogenesis. Our results suggest that Sema3c is a key signaling molecule that mediates interaction between cNCCs and the SHF, which is implicated in proper septation of the OFT and is critical to establishing separate systemic and pulmonary circulation systems.
In order to delineate the molecular mechanisms that regulate Sema3c expression during cardiac development, we analyzed Sema3c genomic sequences and searched for cis-regulatory elements that control Sema3c expression. Our transgenic approaches revealed that a 0.7 kb genomic sequence in the Sema3c 5′-flanking region, which includes exon 1, could activate Sema3c expression in the OFT and pharyngeal arches at embryonic day (E) 10.5 (Fig. 1A–C: pm-691).
To localize the regulatory elements responsible for Sema3c expression in the embryonic OFT and pharyngeal arches, we constructed multiple genomic fragments of the Sema3c 5′ region and tested their ability to direct lacZ expression under a heterologous promoter (hsp68)15. F0 transgenic mouse embryos harboring a 0.2 kb transgene from −691 to −492 of the Sema3c 5′ region (pm-492) showed X-gal staining in the OFT and pharyngeal arches at E10.5, reminiscent of transgenic mouse embryos with pm-691 (Fig. 1A–C: pm-492). Comparison of this 0.2 kb DNA sequence between mice and humans revealed a highly conserved consensus binding site for the winged helix/forkhead (Fox) transcription factors (Fig. 1A). A mutation in this Fox binding site (pm-492m) abolished lacZ expression in the OFT and pharyngeal arches (Fig. 1A–C: pm-492m), suggesting that this Fox binding site is essential for Sema3c expression in the embryonic OFT.
To test whether Fox transcription factors could directly activate Sema3c expression through the identified Fox binding site, we performed luciferase assays using co-expression of Foxc1, Foxc2 and numerous transcription factors that are essential for heart development with the reporter construct including the localized Sema3c contiguous sequences (pm-711: from −711 to +68) (Fig. 1D). Our analyses showed that Foxc1 and Foxc2 could activate the pm-711 luciferase significantly at much higher levels than any of the other tested transcription factors could. (Fig. 1E and Supplementary Figure S1). Deletion of the regulatory element, including the Fox binding site (pm-329), resulted in significantly decreased transcriptional activity of the luciferase reporter by Foxc1 and Foxc2 (Fig. 1E). ChIP assay demonstrated that both Foxc1 and Foxc2 could directly bind to the DNA fragment including the identified Fox binding site (Fig. 1F). Next, we analyzed the in vivo expression pattern of Sema3c mRNA. At E10.5, Sema3c mRNA expression in the OFT was significantly decreased in Foxc1 or Foxc2 null embryos as compared to wild-type embryos (Fig. 1G and H).
To further explore the regulation of Sema3c, we assessed the role of Tbx1, a downstream target of Foxc1 and Foxc216–18. Tbx1 is expressed in the SHF, but not in cNCCs19–21, and is a major genetic modulator of cardiac OFT defects in 22q11.2 deletion syndrome (22q11DS)22–24. A previous study showed that Tbx1 null mice showed a significant reduction in Sema3c expression in the OFT region25, though the exact regulatory mechanism of Tbx1 on Sema3c is unclear. We investigated the role of Tbx1 on the OFT enhancer in the Sema3c 5′-flanking region and found that Tbx1 itself has no effects on the transcriptional activity of the OFT enhancer region (Supplementary Figure S1). However, overexpression of Tbx1 together with Foxc1 showed a significant increase in the transcriptional activation of the OFT enhancer, whereas overexpression of Tbx1 with Foxc2 showed milder synergistic interactions than with Foxc1. (Supplementary Figure S2A). Furthermore, Foxc1 displayed a direct interaction with Tbx1 in the co-immunoprecipitation assay (Supplementary Figure S2B). As there is no Tbx1-specific binding locus on the enhancer region, these findings suggest that Tbx1 acts as a transcriptional co-factor of Foxc1 and Foxc2 and promotes synergistic Sema3c expression in the OFT myocardium derived from SHF progenitor cells. Taken together, these results indicate that Foxc1 and Foxc2 can directly regulate Sema3c expression in the OFT through the Fox-binding site on the 5′-flanking sequence of Sema3c, and that Tbx1 synergistically enhances the function of Foxc1 via direct interactions and partially enhances Foxc2 indirectly during cardiac development.
We further validated the role of Tbx1 in the regulation of Sema3c expression using Tbx1 hypomorphic mice26. In a manner similar to a previous study with Tbx1 knockout mice25, Sema3c expression was reduced in the OFT region of Tbx1 hypomorphic mice (Supplementary Figure S3: Tbx1 neo/neo)26 compared to that in heterozygous littermates (Supplementary Figure S3: Tbx1 neo/+). Interestingly, mice hypomorphic for Tbx1 showed ectopic Sema3c expression in the pharyngeal arches region at E10.5 in comparison with wild-type mice (Supplementary Figure S3). Since lacZ transgene regulated by Sema3c intron 1 or 3′-flanking region showed βGal expression in pharyngeal arch region (Supplementary Figure S4)13, we generated a transgenic mouse line harboring the intron1 and the 3′-flanking genomic sequence of Sema3c (Sema3c int1/3′-lacZ tg; Fig. 2A) and crossed it into the Tbx1 hypomorphic background. LacZ expression in Sema3c int1/3′-lacZ transgenic (tg) embryos was consistent with endogenous Sema3c mRNA expression at E10.5, whereas lacZ expression was detected in large part of pharyngeal arch region in Tbx1 hypomorphic embryos (Fig. 2B). Coronal sections revealed that the lacZ was ectopically expressed in pharyngeal arches in Tbx1 hypomorphic embryos (Fig. 2C). βGal expression was coincident with the Sema3c expression pattern in the pharyngeal arch region as observed by immunofluorescence, both in wild-type and Tbx1 hypomorphic embryos (Fig. 2D). In wildtype embryos at E10.5, Sema3c was expressed in the mesoderm of the pharyngeal arch region, but not in cNCCs labeled with anti-AP2α antibody (Fig. 2E: Tbx1+/+). Intriguingly, in Tbx1 hypomorphic embryos at E10.5, cNCCs migrated less toward the ventral area of the pharyngeal arch region and aggregated around the dorsal arteries with remarkable ectopic expression of Sema3c, as demonstrated by the overlapping expression of Sema3c and AP2α (Fig. 2E: Tbx1 neo/neo). Moreover, at this stage, Sema3c was highly expressed in the Isl1-positive SHF progenitor cells in the wild-type embryos; in contrast, Sema3c was expressed in AP2α-positive cNCCs in the pharyngeal arch region along with a significant decrease of Isl1-positive cells due to developmental defects of SHF progenitor cells in Tbx1 hypomorphic embryos (Fig. 2F). A past study has shown that Sema3c attracts cNCCs and regulates their migration into the most proximal part of the OFT27. To test whether the abnormal upregulation of Sema3c may disturb cNCC migration, we utilized the primary culture of cNCCs and found that lentiviral-mediated Sema3c overexpression caused aggregation of cNCCs that could be rescued by Sema3c neutralizing antibodies (Supplementary Figure S5). These results suggest that Tbx1 expression in the SHF is essential to inhibit the ectopic expression of Sema3c in cNCCs migrating in the pharyngeal arch region, and that loss of Tbx1 function may lead to ectopic expression of Sema3c in cNCCs, which results in their abnormal migration and/or aggregation.
Given the importance of Tbx1 in OFT development, it may also regulate the expression of Sema3c in cNCCs. However, the expression of Tbx1 in the SHF and its absence in cNCCs16, 20 suggests the existence of some secreted factors from the SHF, downstream of Tbx1, which mediate the inhibitory regulation of Sema3c in cNCCs. Using a candidate approach, we focused on Fgf8 signaling that is considered a downstream target of Tbx1 or Foxc genes17, 26. Fgf8 is a secreted signaling protein expressed and functioning in the cardiac crescent and splanchnic mesoderm as well as in the pharyngeal endoderm and ectoderm during development of the pharyngeal arches and the OFT28–31. In the pharyngeal arch region, Fgf8 is thought to be secreted from SHF progenitor cells and may contribute to the interaction between SHF and cNCCs26. Furthermore, a recent report showed that Fgf8 was both chemotactic and chemokinetic for cNCCs32. To elucidate the role of Fgf8 in regulation of Sema3c expression during the development of cNCCs, we utilized the primary culture of cNCCs. Interestingly, Sema3c expression in primary cultured cNCCs was downregulated by treatment with Fgf8 (Fig. 3A). It has been reported that Fgf8 signaling transduced the signal via ERK and AKT pathway and that the ERK signaling contributed to cNCC migration32–34. The inhibition of ERK signaling caused significant upregulation of Sema3c in Fgf8-stimmulated cNCCs whereas inhibition of ALK signaling showed no effect on Sema3c expression (Fig. 3B). Furthermore, the phosphorylation of ERK1/2 was disturbed in cNCCs around the pharyngeal arch arteries of Tbx1 hypomorphic embryos as compared to wild-type littermates (Fig. 3C). These results indicate that Tbx1-Fgf8-ERK signaling negatively regulates the expression of Sema3c in cNCCs.
To validate the role of Fgf8 in the differentiation of cNCCs during cardiogenesis, we utilized a Wnt1-Cre/floxed-EGFP transgenic mouse line35, 36. Slight mechanical dissociation of the third through sixth pharyngeal arch regions in the Wnt1-Cre/floxed-EGFP transgenic mice resulted in the formation of EGFP-positive neural spheres (Supplementary Figure S6A). After seven days of incubation in differentiation media, the EGFP-positive neural spheres gave rise to multi-lineage derivatives including neurons, glia, and smooth muscle cells (Supplementary Figure S6B). This suggests that neural spheres derived from roughly the third to sixth pharyngeal arch regions maintain multi-lineage potential. However, the addition of Fgf8 did not affect the differentiation efficiency of cNCCs to the three lineage derivatives (Supplementary Figure S6C), suggesting that Fgf8 does not play a role in the differentiation of cNCCs.
To test whether blockage of Fgf8 signaling could cause migration defects in cNCCs with enhanced expression of SEMA3C, we injected Fgf8 neutralizing antibody between the 4th and the 6th pharyngeal arches in chick embryos at HH stage 17 in ovo. After 48hrs incubation, embryos injected with Fgf8 neutralizing antibody showed the ectopic expression of SEMA3C mRNA surrounding the 4th to 6th pharyngeal arches (Fig. 4A and B). Furthermore, chick-quail chimera analyses demonstrated that embryos injected with Fgf8 neutralizing antibody resulted in migration defects of cNCCs at HH stage 22 (Fig. 4C). Consistently, ink injections revealed hypoplasia of the 6th pharyngeal arch artery in chick embryos injected with Fgf8 neutralizing antibody compared to control embryos (Fig. 4D). Taken together, these results suggest that Tbx1-Fgf8 signaling inhibits the expression of SEMA3C in migrating cNCCs, resulting in proper guidance of their migration from the dorsal pharyngeal region to the OFT.
In this study, we demonstrate a molecular mechanism for development of proper OFT septation, where Sema3c is tightly regulated in a spatio-temporal fashion during the interaction between cNCCs and SHF. Sema3c expression in the OFT myocardium was shown to be directly regulated by Foxc1 and Foxc2 through conserved Fox-binding sites, where Tbx1 acts synergistically as a co-factor for Foxc proteins. Partial downregulation of Sema3c was consistently observed in Foxc1- or Foxc2 null embryos; this is probably because of their redundancy, as a previous report showed that Foxc1 and Foxc2 played similar dose-dependent roles in development of the OFT and aortic arch system37. We and others have previously shown that Foxc proteins also regulate the expression of Tbx1 in the SHF in a dose-dependent manner16–18. It was reported that the OFT defect resulting from disruption of the Sema3c locus in mice was reminiscent of that seen in mice with ablation of Tbx125, whereas the compound Foxc1/Foxc2 mutant mice showed severe hypoplasia in outflow tract and right ventricle17. These findings may suggest the essential upstream effect of Foxc1/Foxc2 on multiple factors in the OFT development, including Tbx1-Sema3c transcriptional cascade.
Proper cNCC migration is necessary for the septation of great arteries. Although genetic ablation studies have shown many potential factors involved in the development of cNCCs38, the mechanism for proper homing of cNCCs from the neural tube to the OFT is still largely unknown. Plein et al. reported that mice with Sema3c ablations in the neural crest showed disrupted septation in the OFT, although the mesenchymal cells were located in the OFT cushion39. These mesenchymal cells in the OFT cushion lacked Sema3c expression; however, expression was maintained in the proximal region of the OFT myocardium. It was also reported that Sema3c attracted cNCCs and regulated their migration to the extreme proximal region of the OFT27. In this study, we showed that downregulation of Tbx1 resulted in dysregulated expression of Sema3c in the pharyngeal arch region, causing abnormal aggregation and migration of cNCCs. We confirmed that the overexpression of Sema3c in cNCCs led to their aggregation in vitro. On the other hand, Sema3c was downregulated in the OFT region of Tbx1 hypomorphic mice; this observation is consistent with that of a previous report25. Taken together, these findings suggest a developmental model that Sema3c from the SHF-derived OFT myocardium plays a role as an attractant for cNCCs, that inhibition of Sema3c is necessary in cNCCs during their migration, and that the activation of Sema3c is required for normal aggregation of cNCCs in the proximal OFT region. Proper spatio-temporal expression of Sema3c may be essential for correct septation of the OFT.
Given the close apposition and function of cNCCs and the SHF in the development of the OFT, it is possible that behavior of cNCCs may be dependent on signals from the SHF. As for interaction of the two cardiac stem cell lineages, our findings suggest that Fgf8 from the SHF may diffuse distally to directly influence the development of cNCCs via activation of ERK signaling. A recent report showed that Fgf8 was both chemotactic and chemokinetic for cNCCs32. Fgf8 was also shown as a downstream effector of Tbx1 secreted from the SHF in the pharyngeal arch and OFT region26. The ablation of Fgf8 with a Tbx1Cre transgene resulted in OFT defects40. Suppression of ERK signal in zebrafish causes non-migration of cNCCs41. Taken together, our data suggest that inhibition of Sema3c in cNCCs by Fgf8 secreted from the SHF in regulation by Tbx1 may contribute to a proper condensation of the cNCC cluster in the OFT that establishes the septum.
In conclusion, this study implicates Foxc1, Foxc2, and Tbx1-Fgf8 signaling as essential regulators for Sema3c expression, contributing the temporal and spatial regulation in the development of cNCCs during OFT and aortic arch formation. We propose a model molecular mechanism in which Tbx1 restricts the expression of Sema3c in the SHF by blocking its ectopic expression in migrating cNCCs via Fgf8 signaling in the pharyngeal arch region during development of the OFT (Fig. 5). Foxc1 and Foxc2, together with Tbx1, may synergistically activate Sema3c expression in progenitor cells in the SHF in a process during their migration and differentiation into the OFT myocardium. This orchestrated Sema3c regulation may lead to a signal for the correct migration of cNCCs into the OFT to give rise to the septum. Our findings provide new insight into the interaction between cNCCs and the SHF, as well as into the etiology of congenital heart diseases involving OFT defects.
Tbx1 neo/+ mice were kindly provided by Dr. Deepak Srivastava26. Foxc1-/- and Foxc2-/- mice embryos were kindly provided by Dr. Tsutomu Kume37. Transgenic mice expressing Cre recombinase under control of the Wnt1 promoter/enhancer (Wnt1-Cre)35 were mated with EGFP reporter mice (CAG-CAT-EGFP)36 to obtain Wnt1-Cre/Floxed-EGFP double-transgenic mice. Adult wild-type mice were purchased for mating from Sankyo Labo Service (Tokyo, Japan). PCR analysis of tail and yolk sac DNA of the Tbx1-neo allele, foxc1 null and foxc2 null alleles have been previously described26, 37. All experimental procedures were approved by the ethics committee of Keio University and were in accordance with the Guide for the Care and Use of Laboratory Animals (U.S. National Institutes of Health).
Mouse Sema3c 5′-flanking regions (for mouse Sema3c promoter/enhancer-lacZ plasmids; pm-691, pm-492 and pm-492m) or intron1 and 3′-flanking regions (for mouse Sema3c int1/3′-lacZ tg plasmid) were generated from the DNA extract of mouse tail by PCR and different fragments of the Sema3c upstream, intron1 or 3′-flanking regions were once cloned into plasmid pBluescript (Stratagene), then subcloned into the hsp68lacZ reporter construct42 or pGL3-Basic (Promega, Madison, WI, USA). The PCR-based mutagenesis on this plasmid was performed using pfu DNA polymerase and a QuikChange site-directed mutagenesis kit (Stratagene). The primers used were 5′-GATTCAGTGTGGAAAGGTggGCATAATTACTGATTTCAG-3′ and 5′-CTGAAATCAGTAATTATGCccACCTTTCCACACTGAATC-3′. Restriction mapping and sequence analysis confirmed the presence of the mutation, and the insert was excised and cloned into the hsp68lacZ reporter plasmid. All constructed vectors were verified by sequencing.
DNA fragments were isolated from mouse Sema3c promoter/enhancer–lacZ plasmids (pm-691, pm-492 and pm-492m) or mouse Sema3c int1/3′-lacZ tg plasmid using SalI site. These fragments were microinjected into fertilized oocyte pronuclei, eggs were transferred into the oviducts of pseudopregnant females, and transgenic embryos were identified by PCR analysis for each promoter or enhancer–lacZ gene in yolk sac DNA. Embryos were harvested at E10.5 and stained for beta-galactosidase activity, as described previously13.
Fertilized chicken eggs were incubated at 38°C in a humidified incubator until the embryos reached stages HH1743. The eggs were windowed and prepared for following procedures. Goat anti-Fgf8 polyclonal antibody (Santa Cruz sc-6958) were diluted by phosphate-buffered saline (PBS(−)) to the concentration of 100μg/mL. The neutralizing antibody or PBS(−) (for control) was injected into pharyngeal arch region between 4th and 6th pharyngeal arch arteries of HH17 chick embryos. The treated embryos were cultured for 48hrs at 38°C, then dissected and fixed with 4% PFA.
India ink (Rotring, Germany or Kiwaguro, Sailor, Japan) was injected into a vitelline vein using a capillary pipette pulled to a fine diameter. The ink was ejected using gentle, positive pressure until it filled the intravascular compartment. The embryos were dissected from surrounding membranes and immersion fixed in neutral buffered formalin overnight, dehydrated, and cleared in benzyl benzoate:methylsalicylate (1:1).
Fertilized chicken eggs and Cortunix japonica quail eggs were used. They were incubated at 38°C in forced-draft incubators with a constant humidity until the embryos reached stages HH8–10. The eggs were windowed and prepared for microsurgery following the procedures previously described44, 45. The transplantation procedure always used chick embryos as hosts and quail embryos as donors, and was performed homotopically. Each presumptive arch region (3rd, 4th and 6th) of the cardiac neural crest was transplanted individually. Bilateral transplantations were done. India ink, diluted 1/10 in saline solution, was injected into the subgerminal cavity to allow visualization of the embryo. The vitelline membrane was torn over the area of cardiac neural crest to be excised. The neural folds containing premigratory neural crest cells from a chick embryo was removed at a defined level by using an electrolytically sharpened tungsten needle. The region was replaced with the corresponding neural folds from a quail embryo. After the microsurgery, the egg-shell windows were sealed by using cellophane tape and the eggs were returned to the incubator. Sham-operated embryos were produced by tearing the vitelline membrane with no further surgery. Control embryos were opened the window and counted the somite with no treatment. 100μg/mL of goat anti-Fgf8 polyclonal antibody (Santa Cruz) or PBS(−) (for control) was injected into pharyngeal arch region between 4th and 6th pharyngeal arch arteries of HH17 chimera embryos. The treated embryos were cultured for 48hrs at 38°C, then dissected and fixed with 4% PFA.
HH17 chicken embryos were harvested and the pharyngeal arch region between 3rd and 6th pharyngeal arch arteries was dissected out with resection of the heart. Pharyngeal arch region were collected into 1.5ml tubes with 1ml of Trizol reagent (Invitrogen) and immediately frozen in liquid nitrogen and stored at −80°C. Total RNA extraction was performed according to the manufacturer’s instructions.
For whole-mount immunostaining, embryos were fixed in 4% paraformaldehyde in PBS at 4°C for 1hr and rinsed in PBS at room temperature. The embryos were dehydrated in a graded series of methanol and stored at −20°C. Endogenous peroxidase activity was blocked with 6% H2O2 in methanol for 6hrs at room temperature and the embryos were rehydrated and blocked in PBSMT (2% non-fat milk, 1.0% Triton X-100, PBS) for 1hr twice at room temperature. The embryos were incubated with monoclonal antibody QCPN (Developmental Studies Hybridoma Bank), which targets a nuclear antigenic determinant that is specific to quail cells, in PBSMT for 2 days at 4°C while being agitated on a shaking platform. After extensive washes with PBSMT at 4°C six times (1hour each), the embryos were incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (Jackson ImmunoResearch Lab Inc, USA) in PBSMT (1:200 dilution) at 4°C overnight. Embryos were rinsed in PBSMT at 4°C six times (1hour each) and finally in PBST (1% Triton X-100, PBS) for 20minutes at room temperature. The peroxidase staining was performed by incubating embryos in 0.16mg/ml 3-3′-diaminobenzidine tera-hydrochroride (DAB, Sigma) in PBT for 1hr followed by the addition of H2O2 to the final concentration of 0.01%. The staining reaction was stopped by rinsing in PBT. The embryos were postfixed in 4% paraformaldehyde in PBS at 4°C for 1hr. For sectioning, embryos were fixed in 4% paraformaldehyde in PBS at 4°C overnight and rinsed in PBS and dehydrated in graded alcohols, cleared, and embedded in paraffin following the procedures previously described45. Sections were cut at 10μm and deparaffinized, rehydrated, blocked with 0.3% H2O2 in PBS. The sections were blocked with 2% bovine serum albumin in PBS and incubated with QCPN overnight at 4°C. After washing, they were incubated with biotin-conjugated goat anti-mouse IgG (Vector BA-9200) diluted 1:200 in 2% bovine serum albumin in PBS. The slides were incubated with ABC Elite kit (Vector) and then washed. After final washing, the slides were visualization with DAB.
Total RNA was extracted using the Trizol reagent (Invitrogen), and RT-PCR was performed as described previously46. For quantitative analysis of mouse Sema3c and chicken SEMA3C expression, the respective cDNA was used as the template in a TaqMan real-time PCR assay using the ABI Prism 7500 FAST sequence detection system (Applied Biosystems) according to the manufacturer’s instructions. All samples were run in triplicate. The data was normalized to mouse Gapdh or chicken SEMA3C expression. The TaqMan probes for mouse Sema3c, Gapdh, and chicken SEMA3C and GAPDH were Mm00443121_m1, Mm99999915_g1, Gg03364166_m1 and Gg03346984_g1, respectively.
Whole-mount in situ hybridization analyses of mouse and chick embryos were performed as described25. The mouse Sema3c and chicken SEMA3C RNA probes were used for in situ hybridization. RNA transcripts were synthesized using a digoxigenin (DIG)-labeling kit (RocheMannheim) according to the manufacturer’s protocol.
Mice embryos were fixed in 4% paraformaldehyde on ice for 1hr and embedded in OCT, frozen in dry iced hexane, and kept at −80°C until sectioned. Frozen sections (8μm) were blocked with 5% goat serum in PBS with 0.1% Tween20 for 1hr at room temperature and stained with following primary antibodies: rat monoclonal anti-Sema3 (1:50, R&D MAB1728), mouse monoclonal anti-AP2α clone 3B5 (1:100, Developmental studies hybridoma bank), rabbit polyclonal anti-β-galactosidase (1:1000, Rockland 200–4136) and rabbit polyclonal anti-Isl1 (1:100, Abcam ab20670). For phosphor-Erk1/2 staining, mice embryos were fixed in 4% paraformaldehyde on ice for 1hr and embedded in paraffin. Paraffin sections (7μm) were quenched by 3% H2O2 in methanol and were treated for 20min with Citrate Buffer, pH 6.0 (Sigma-Aldrich) using standard protocols. Samples were blocked with 5% goat serum in PBS with 0.1% Tween20 for 1hr at room temperature and stained with rabbit monoclonal anti-phospho-ERK1/2 clone D13.14.4E (1:100, Cell Signaling #4370) and mouse monoclonal anti-AP2α Clone 3B5 (1:40). Signals were amplified using TSA Biotin Systems (Perkin Elmer NEL700A001KT). Secondary antibodies were used as follows: anti-rabbit IgG (Alexa 647 A-21245), anti-mouse IgG (Alexa 555 A-21422), anti-mouse IgG (Alexa 488 A-11001), anti-rabbit IgG (Alexa 555 A21429), and anti-rat IgG (Alexa 488 A-11006, all from Molecular Probes), Biotinylated Goat anti-rabbit IgG (Vector BA-1000) and Streptavidin, Alexa Fluor 594 conjugate (Molecular Probes S-11227). Nuclear counterstaining was performed with Hoechst 33342 (10mg/ml, Sigma B2261). Image acquisition was performed on a universal fluorescence microscope (Axioskop 2 Plus; Carl Zeiss) and a confocal microscope (LSM 510 Meta; Carl Zeiss) and ZEN software (Carl Zeiss).
Pregnant wild type C57BL6 female mice were deeply anesthetized and sacrificed by cervical dislocation and embryos were collected at E9.5 into dissection medium. The neural tubes between the otic vesicles and sixth somites were cut out. The neural tubes were incubated in 1mg/mL of collagenase/dispase (Sigma-Aldrich) in PBS buffer at room temperature for 10minutes followed by removal of non-neural ectoderm and mesoderm tissues. The isolated neural tubes were placed into the wells of fibronectin (Sigma-Aldrich)-coated 24-well plates and cultured with Self-Renewal (SR) medium consisting of 12.5mL low glucose DMEM (GIBCO), 7.5mL Neurobasal Medium (GIBCO), 117μM retinoic acid (Sigma-Aldrich), 50mM 2-mercaptoethanol (GIBCO), 3.75mL Chick Embryo Extract (MP Biomedicals), 250μL N2 salt supplement (GIBCO), 500μL B27 supplement (GIBCO), 250μL penicillin-streptomycin (GIBCO), 20μg/mL human recombinant IGF1 (R&D #291-G1) and 20μg/mL human recombinant bFGF (Pepro Tech #100–18b)47. After 24hrs incubation at 37°C, the neural tubes were removed.
For lentiviral vector production, whereby 80% confluent 293T cells (Thermo Fisher Scientific) were transfected overnight with virus production vectors including pLenti-c-Myc-DKK Sema3c (Origene) or pLenti6-EGFP (control vector), pucMDG, and pCMVΔ8.9148 using Lipofectamine 2000 (Life technologies) and were subsequently purified, concentrated using Lenti-X Concentrator (Clontech), and frozen until use. Titers were measured by using qPCR Lentivirus Titration Kit (Applied Biological Materials) and Power SYBR Green PCR Master Mix (Applied Biosystems). For transduction, concentrated viral supernatant containing 7.5×106 transducing particles were added into the culture medium of cNCC for 24hrs with or without 1μg/mL rat anti-Sema3c monoclonal antibody (R&D MAB1728), upon which cNCCs’ phenotype was observed using a Leica DM IL LED microscope (Leica).
Mouse cNCCs isolated from neural tubes were seeded to fibronectin-coated 6 well plates and cultured in SR medium with 100ng/mL human/mouse recombinant FGF8 (R&D) with or without ERK1/2 inhibitor (SCH772984 Selleck Chemicals) or AKT inhibitor (MK-2206 Selleck Chemicals) for 3 days.
Pregnant Wnt1-Cre/Floxed-EGFP mice or wild-type C57BL6 mice were deeply anesthetized and sacrificed by cervical dislocation, and embryos were collected at E9.5 into PBS with 1% glucose. The pharyngeal arch region between 3rd and 6th pharyngeal arch arteries was dissected and removed with resection of the neural tube and heart. The samples were then collected into 1.5ml tubes with 1ml of a serum-free sphere-forming medium consisting of MHM medium [DMEM/F-12 (1:1) (GIBCO 12100-046/21700-075) supplemented with insulin (25mg/ml), transferrin (100mg/ml), progesterone (20nM), sodium selenate (30nM), putrescine (60nM) (all from Sigma-Aldrich)], recombinant human EGF (100ng/ml) (Pepro Tech #100-15), human FGF-basic (100ng/ml) (Pepro Tech #100-18b), and B27 (20ng/ml) (Invitrogen). The dissected tissue was mechanically dissociated by pipetting, and cells from the pharyngeal arches were seeded at 5×103 cells/ml into an ultra - low attachment 6-well plate (Corning) with serum-free sphere-forming medium46 and incubated at 37°C, with 5% CO2; half of the medium was changed after 4 days.
Spheres were plated on poly-D-lysin/laminin (Sigma P7405/Invitrogen 23017-015)-coated 8-well chamber slides (Iwaki 5732–008) for immunostaining and cultured for 7 days in the following differentiation medium: MHM medium with 1% FBS (Sigma-Aldrich) and 25ng/mL mouse Fgf8 (Peprotech) or PBS (control).
Cells were fixed in 4% PFA and stained with the following primary antibodies: anti- β-III tubulin (mouse IgG2b, 1:500, Sigma T8660), anti- αSMA (mouse IgG2a, 1:1000, Sigma A2547), and anti-GFAP (rat IgG, 1:500, Invitrogen 13-0300). The secondary antibodies used for staining were the following: anti-mouse IgG2b (Alexa 488 A-21141), anti-mouse IgG2a (Alexa 350 A-21130), and anti-rat IgG (Alexa 555 A21434). The stained samples were observed under a universal fluorescence microscope (Axioskop 2 Plus; Carl Zeiss).
For the in vivo ChIP experiments, extracts were prepared from approximate 1×107 of P19CL6 cells49 which were transfected by FLAG-tagged Foxc1 or Foxc2 expression vectors 48hours before harvest. For the ChIP assays, the ChIP-IT Express Chromatin Immunoprecipitation Kits (Active Motif) were used according to manufacturer’s protocol. Primers in PCR reactions were 5′-ACAAGCTTCGCTGGACTGAT-3′ and 5′-TCTCGGCTCACTCTCTGGTT-3′ (−148+43). The amplified region corresponded to the mouse Sema3c promoter, which encompasses the FOX-binding site.
COS1 cells were transiently transfected by FLAG-tagged Foxc1/Foxc2 or Myc-tagged Tbx1 expression vectors with Lipofetctamine LTX and Plus reagent (Invitrogen), collected after 48h in lysis buffer (20mM Tris-HCl pH8.0, 137mM NaCl, 1% NP-40, 2mM EDTA, 1μg/mL Leupeptin, 1mM PMSF, 1μM pepA, and 50μM E-64), incubated with anti-c-Myc antibody (mouse IgG1κ, clone 9E10, 6mg/mL, BioLegend) and protein G Sepharose beads (GE healthcare), immunoblotted with monoclonal anti-Myc (BioLegend, 1:1000) or polyclonal anti-DDDDK-tag (Rabbit IgG, 1:1000, MBL, PM020) antibodies, and detected by the SuperSignal West Pico Chemiluminescent Substrate (Pierce).
HeLa cells (ATCC) were transfected using Lipofectamine LTX and Plus reagent with 400ng reporter vector, 800ng expression vectors, and 0.25ng pSV-Rluc internal control vector13. Luciferase activity was measured 48h after transient transfection according to the manufacturer’s instructions. Two duplicate, three independent experiments were performed.
HeLa cells were cotransfected with 400ng of several SEMA3C promoter–pGL3 vectors and 0.25ng pSV-Rluc internal control vector, together with 800ng pcDNA3.1, Foxc1 or Foxc2 expression vector. The luciferase assay was performed as described previously13.
For luciferase assays, all experiments were performed at least in triplicate and data are reported as normalized relative light units (fold activation) together with the SEM. For promoter activity assays, all experiments were performed at least in triplicate and data are reported as the ratio of normalized relative light units for coexpression with Foxc1 or Foxc2 to that with mock (pcDNA3.1). An unpaired two-tailed t-test was used to calculate significant differences between two groups. If the data were normally distributed, a one-way ANOVA was used and multiple comparison correction analysis was performed using Tukey’s Post Hoc HSD test. A p-value of<0.05 was considered statistically significant.
The authors are grateful to Daniel Bernstein for his critical reading, Deepak Srivastava for providing Tbx1 hypomorphic mice, Brian Black and Sylvia Evans for providing Mef2c and Isl1 expression vector, respectively. We also thank Masae Morishima, Reina Ishizaki, Takatoshi Tsuchihashi, and Ayami Kuramochi for technical assistance with mouse bleeding and genotyping, Keiko Uchida and Kazuki Yamamoto for technical support. This work was supported by MEXT KAKENHI, Grant-in-Aid for Scientific Research (B) and Grant-in-Aid for Young Scientists (B), and by Brain/MINDS project from AMED, Japan.
All authors have read and approved the manuscript. K.K. and H.Y. designed the study and wrote the manuscript; K.K. performed all experiments; S.S. performed the isolation and differentiation of cNCCs; S.-M.T. generated chick-quail chimeras and performed whole mount immunostaining; S.-G.O. performed lentiviral experiments, H.T. provided plasmids; T.K. and H.O. provided transgenic mice and embryos; R.M. provided scientific advice; and H.Y. provided funding and supervised the entire research project.
The authors declare that they have no competing interests.
Electronic supplementary material
Supplementary information accompanies this paper at doi:10.1038/s41598-017-06964-9
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Kazuki Kodo, Email: email@example.com.
Hiroyuki Yamagishi, Email: pj.oiek@gamayh.