Bioenergetic metabolism is a key regulator of cellular function and signaling, but how it can instruct the behavior of cells and their fate during embryonic development remains largely unknown. Here, we investigated the role of glucose metabolism in the development of avian trunk neural crest cells (NCCs), a migratory stem cell population of the vertebrate embryo. We uncovered that trunk NCCs display glucose oxidation as a prominent metabolic phenotype, in contrast to what is seen for cranial NCCs, which instead rely on aerobic glycolysis. In addition, only one pathway downstream of glucose uptake is not sufficient for trunk NCC development. Indeed, glycolysis, mitochondrial respiration and the pentose phosphate pathway are all mobilized and integrated for the coordinated execution of diverse cellular programs, epithelial-to-mesenchymal transition, adhesion, locomotion, proliferation and differentiation, through regulation of specific gene expression. In the absence of glucose, the OXPHOS pathway fueled by pyruvate failed to promote trunk NCC adaptation to environmental stiffness, stemness maintenance and fate-decision making. These findings highlight the need for trunk NCCs to make the most of the glucose pathway potential to meet the high metabolic demands appropriate for their development.

Cellular metabolism, long regarded as a neutral factor during development, is now recognized for its instructing role in cell fate, which is achieved by regulating gene networks and cell signaling (Bhattacharya et al., 2021; Miyazawa and Aulehla, 2018; Pavlova and Thompson, 2016; Shyh-Chang et al., 2013; Zhang et al., 2018). The primary function of cellular metabolism is to provide cells with bioenergetic and biosynthetic supplies from nutrients, e.g. glucose, amino-acids and fatty acids (Zhu and Thompson, 2019). Glucose constitutes the major source for ATP production through two pathways occurring in distinct cellular compartments: glycolysis in the cytoplasm and oxidative phosphorylation (OXPHOS) in mitochondria. Glucose is also an important source of carbon for biomass production through multiple metabolic pathways branching from glycolysis. Among them, the pentose phosphate pathway (PPP) constitutes a major source of nucleotides and NADPH for anabolism (Patra and Hay, 2014).

Glucose metabolism differs among cells and depends on gene regulatory programs driving expression and activity of metabolic enzymes (Lempradl et al., 2015). Thus, in differentiated cells, it is usually catabolic with high production of ATP, as a result of entry of the glycolysis by-product pyruvate into mitochondria and OXPHOS. In contrast, in proliferating undifferentiated cells, as well as in cancer cells, glucose utilization is preferably oriented toward anabolism, and ATP production relies mostly on aerobic glycolysis fed by intense glucose uptake, with extracellular lactate production and low OXPHOS activity, a process known as the Warburg effect (Schell et al., 2017; Vander Heiden et al., 2009). Glucose metabolism in cells also varies in time and space to cope with environmental changes (Johnson et al., 2003; Miyazawa and Aulehla, 2018; Perestrelo et al., 2018; Shyh-Chang et al., 2013; Zhang et al., 2018). Notably, a shift from aerobic glycolysis to OXPHOS occurs in adult stem cells leaving quiescence in order to differentiate and in induced pluripotent stem cells exiting pluripotency (Gu et al., 2016; Ito and Suda, 2014; Perestrelo et al., 2018; Shyh-Chang et al., 2013).

Neural crest cells (NCCs) constitute one of the most spectacular populations of stem cells of the vertebrate embryo (Bronner, 2018; Erickson et al., 2023; Trainor, 2013). These cells appear early all along the embryonic rostrocaudal axis in the dorsal neural tube (NT). Through epithelial-to-mesenchymal transition (EMT), NCCs delaminate and disperse away from the NT to distal sites where they differentiate. Owing to their partition into different populations (cranial, cardiac, vagal, trunk and sacral) along the axis, NCCs give rise to numerous cell types: skeletal and supportive tissues, endocrine cells, cardiac septum and enteric ganglia, as well as peripheral neurons, glia, and melanocytes. Throughout development, NCCs are exposed to different environments and cues impacting on their gene networks, cellular features, and fate (Barriga and Mayor, 2019; Chevalier et al., 2016; Duband et al., 2015; Martik and Bronner, 2017; Soldatov et al., 2019). To date, nutrient requirements and their utilization in energy production throughout NCC development remain largely unknown. Recently, a study on avian cranial NCC delamination reported a metabolic shift toward aerobic glycolysis at onset of migration (Bhattacharya et al., 2020), suggesting that cranial NCCs share metabolic properties with cancer cells. However, the other NCC populations along the embryonic axis, notably trunk NCCs, differ strikingly from cranial NCCs in their gene-regulatory network deployed and interaction with their environment (Li et al., 2019; Rothstein and Simoes-Costa, 2023; Scully et al., 2016; Simões-Costa and Bronner, 2015; Simões-Costa et al., 2012; Soldatov et al., 2019; Théveneau et al., 2007), and might therefore exhibit different metabolic features, possibly accounting for differences in their migratory pattern and ultimate fate.

Here, using a combination of in vivo and in vitro strategies, we investigated the role of glucose metabolism in quail trunk NCC development. We explored which metabolic pathways are recruited for their bioenergetic and biosynthetic supplies and to what extent NCC metabolic activity can drive their behavior and fate. Our results indicate that trunk NCCs necessitate glucose metabolism for development and display a metabolic signature characteristic of glucose oxidation, unlike cranial NCCs. We also show that, although necessary, the OXPHOS pathway does not suffice for trunk NCC development, which instead involves massive mobilization of all the metabolic pathways downstream of glucose uptake, i.e. glycolysis, OXPHOS and PPP, cooperating together. Finally, our data reveal that metabolic pathways activated by nutrient inputs can drive fate decisions in NCCs and raise the intriguing possibility that NCC metabolic activity might contribute to the identity and segregation of the different populations.

Glucose metabolism is required for trunk NCC development in vivo

To determine whether glucose metabolism is involved in trunk NCC development, we first analyzed the expression patterns of key players of glycolysis at the time of their migration (Fig. S1A) by in situ hybridization (ISH) on whole-mount quail embryos and on sections. mRNAs for the glucose transporter Glut-1, for phosphofructokinase PFKP (PFK), and for phosphoglycerate kinase-1 PGK-1, were all expressed throughout the NT at NCC pre-migratory, delamination and early migration stages (Fig. 1A). All mRNAs were evenly distributed along the NT dorsoventral axis and were also detectable in both pre-migratory and early migrating NCCs (Fig. 1B; Fig. S1B). Consistent with this, incubation of embryos with the fluorescent glucose analog 2-NDBG as an estimate of glucose uptake revealed a staining pattern in the NT matching that of Glut-1 mRNA, but which was more pronounced at the NCC migration level (Fig. 1A).

Fig. 1.

Glucose metabolism is required for trunk NCC development. (A) Whole-mount views of the trunk of HH14 quail embryos. Left, hybridized with probes for Glut-1, PFK, PGK-1, Foxd-3 and Sox-10, anterior to the top. Foxd-3 and Sox-10 mark the axial levels of NCC pre-migration (Foxd-3-low, Sox-10-low), delamination (Foxd-3-high, Sox-10-low), and migration (Foxd3-high, Sox-10-high) stages. Right, fluorescent glucose analog (2-NDBG) uptake with the intensity shown by pseudocolor image (16 colors; higher levels in red–yellow) using ImageJ (n=5). (B) Cross-sections through the last somites of whole-mount ISH for Glut-1, PFK, PGK-1 and Sox-10 showing the dorsal NT and delaminating (arrowhead) and early-migrating (arrow) NCCs. The corresponding global views of the Glut-1, PFK and PGK-1 sections are shown in Fig. S1B. (C) Normalized levels of Snail-2, Foxd-3 and Sox-10 mRNAs in the trunk of embryos 5 h after 2-DG or vehicle (no-drug) injection measured by qRT-PCR. Measurements were undertaken in triplicate per embryo and gene, analyzed by two-way ANOVA with Tukey post test relative to the no-drug condition. Results are mean±s.e.m. (D) Whole-mount views of the trunk of embryos 5 h, 24 h and 48 h after 2-DG or vehicle injection, hybridized with probes for Foxd-3 and Sox-10 or immunolabeled for HNK-1. Vertical bars delineate the axial levels where NCC development is defective. Arrowheads, arrows and double arrows point at delaminating, migrating and differentiating NCCs, respectively. The corresponding global views of the embryos are shown in Fig. S2A. (E) Cross-sections through the caudal (left), mid (center), and anterior (right) trunk of embryos 24 h after 2-DG or vehicle injection and processed for whole-mount ISH for Sox-10. Arrows point at NCCs. (F) Overall aspect of trunk NT explants in 5 h and 24 h cultures in medium without glucose and pyruvate (no Gluc-no Pyr), with 5 mM glucose, 1 mM pyruvate, both, and in glucose and pyruvate medium with 2-DG. (G) Scatter plot with mean±s.e.m. of NCC outgrowth area at 5 h and 24 h in conditions as in F, analyzed using one-way ANOVA followed by Dunnett's multiple comparison test relative to the condition with glucose and pyruvate. (H) Distance covered over time by the NCC migratory front in explants as in F and measured using video-microscopy. t0 corresponds to onset of recording 2–4 h after initiation of culture. 2-DG was applied immediately before recording. Front progression was analyzed using a custom GUI written in MATLAB. Values are mean±s.e.m. Data are from at least three independent experiments, and the images in A, B, D, E and F are representative images from these indepedent experiments. (n) indicates the number of embryos or explants analyzed. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; ns, not significantly different, P>0.05. Scale bars: 200 µm (A,D); 100 µm (B,E,F). ao, aorta; dmt, dermamyotome; ec, ectoderm; nt, neural tube; pm, unsegmented paraxial mesoderm; sc, sclerotome; so, somite.

Fig. 1.

Glucose metabolism is required for trunk NCC development. (A) Whole-mount views of the trunk of HH14 quail embryos. Left, hybridized with probes for Glut-1, PFK, PGK-1, Foxd-3 and Sox-10, anterior to the top. Foxd-3 and Sox-10 mark the axial levels of NCC pre-migration (Foxd-3-low, Sox-10-low), delamination (Foxd-3-high, Sox-10-low), and migration (Foxd3-high, Sox-10-high) stages. Right, fluorescent glucose analog (2-NDBG) uptake with the intensity shown by pseudocolor image (16 colors; higher levels in red–yellow) using ImageJ (n=5). (B) Cross-sections through the last somites of whole-mount ISH for Glut-1, PFK, PGK-1 and Sox-10 showing the dorsal NT and delaminating (arrowhead) and early-migrating (arrow) NCCs. The corresponding global views of the Glut-1, PFK and PGK-1 sections are shown in Fig. S1B. (C) Normalized levels of Snail-2, Foxd-3 and Sox-10 mRNAs in the trunk of embryos 5 h after 2-DG or vehicle (no-drug) injection measured by qRT-PCR. Measurements were undertaken in triplicate per embryo and gene, analyzed by two-way ANOVA with Tukey post test relative to the no-drug condition. Results are mean±s.e.m. (D) Whole-mount views of the trunk of embryos 5 h, 24 h and 48 h after 2-DG or vehicle injection, hybridized with probes for Foxd-3 and Sox-10 or immunolabeled for HNK-1. Vertical bars delineate the axial levels where NCC development is defective. Arrowheads, arrows and double arrows point at delaminating, migrating and differentiating NCCs, respectively. The corresponding global views of the embryos are shown in Fig. S2A. (E) Cross-sections through the caudal (left), mid (center), and anterior (right) trunk of embryos 24 h after 2-DG or vehicle injection and processed for whole-mount ISH for Sox-10. Arrows point at NCCs. (F) Overall aspect of trunk NT explants in 5 h and 24 h cultures in medium without glucose and pyruvate (no Gluc-no Pyr), with 5 mM glucose, 1 mM pyruvate, both, and in glucose and pyruvate medium with 2-DG. (G) Scatter plot with mean±s.e.m. of NCC outgrowth area at 5 h and 24 h in conditions as in F, analyzed using one-way ANOVA followed by Dunnett's multiple comparison test relative to the condition with glucose and pyruvate. (H) Distance covered over time by the NCC migratory front in explants as in F and measured using video-microscopy. t0 corresponds to onset of recording 2–4 h after initiation of culture. 2-DG was applied immediately before recording. Front progression was analyzed using a custom GUI written in MATLAB. Values are mean±s.e.m. Data are from at least three independent experiments, and the images in A, B, D, E and F are representative images from these indepedent experiments. (n) indicates the number of embryos or explants analyzed. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; ns, not significantly different, P>0.05. Scale bars: 200 µm (A,D); 100 µm (B,E,F). ao, aorta; dmt, dermamyotome; ec, ectoderm; nt, neural tube; pm, unsegmented paraxial mesoderm; sc, sclerotome; so, somite.

To assess the role of glucose metabolism in trunk NCC development, we used an in vivo loss-of-function strategy (Fig. S1C). 2-deoxyglucose (2-DG), a potent inhibitor of the whole glycolytic cascade (Fig. S1A), was injected in ovo in the caudal region of quail embryos at stages preceding onset of migration, and NCC development was subsequently followed during delamination using Snail-2 and Foxd-3 markers (Duband, 2010; Lukoseviciute et al., 2018; Nieto, 2002), during migration using Sox-10 (Schock and LaBonne, 2020), and later during formation of neural derivatives using HNK-1 (Vincent and Thiery, 1984). First, quantitative real-time RT-PCR (qRT-PCR) analyses of the caudal region 5 h post-injection revealed a robust repression of Snail-2, Foxd-3 and Sox-10 following 2-DG treatment (Fig. 1C). 2-DG also caused a delay in NCC delamination and provoked severe migration defects in a reproducible manner, ultimately resulting in an aberrant patterning of the sensory and sympathetic ganglia (Fig. 1D; Fig. S2A,B). Detailed spatiotemporal analyses of migrating NCC distribution along the NT after 24 h showed that cells were missing from the top of the NT in the caudal trunk, trapped dorsally instead of following the dermamyotome ventral side in the mid-trunk, or were absent from their target sites in the periaortic territory in the anterior trunk (Fig. 1E). Moreover, whereas at 10 mM 2-DG, no overt morphological perturbations of the embryo were noticed (Fig. 1D; Fig. S2A,B), at 25 mM we observed severe defects in both the NT and paraxial mesoderm after 24 h, ultimately causing tail truncation (Fig. S2A–C), as also reported previously (Oginuma et al., 2017). These results establish that in vivo glucose metabolism is necessary for efficient NCC dispersion and for patterning of their neural derivatives.

Glucose metabolism is required for trunk NCC dispersion in in vitro culture

To verify that alterations of NCC development did not merely result from indirect effects on the patterning and signaling activity of neighboring tissues, and to better appreciate the NCC response to glucose, we used a classical in vitro culture approach (Fig. S1D) in which NCCs were generated from isolated trunk NT explants in defined conditions favoring delamination, migration and differentiation, with timing and kinetics comparable to those observed in vivo (Duband et al., 2020). First, we analyzed whether glucose was necessary for NCC dispersion, and at what concentration it was the most effective, by culturing trunk NT explants in DMEM containing or not containing glucose and supplemented with glutamine and only 1% serum to minimize the contribution of exogenous growth factors. The behavior of the NCC population was characterized over time during delamination and migration, focusing on migratory front progression, aspect, size and cell density of the outgrowth, as well as individual cell morphology. In absence of glucose, a few NCCs were capable of initiating migration, but they rapidly failed to disperse away over time and showed a poorly spread morphology and low density (Fig. 1F–H; Fig. S3A,B). In the presence of glucose, in contrast, the NT rapidly produced large and dense NCC outgrowths expanding fast during the first 24 h and with numerous well-spread cells (Fig. 1F–H; Fig. S3A,B), as observed in classical serum-rich media (Dupin et al., 2018; Rovasio et al., 1983; Santiago and Erickson, 2002). Glucose favored NCC spreading at all concentrations tested, but maximal NCC outward progression, with robust and sustained individual cell velocity, high persistence and density, was reached at 5 mM (Fig. S3C–G), which was the concentration used for the rest of the study.

Because pyruvate provided as a nutrient can be an alternative entry point to fuel OXPHOS independently of glucose (Fig. S1A), we compared its ability to support NCC development to that of glucose. In pyruvate, NCC dispersion occurred almost normally at culture onset, but it decreased strongly after 5–8 h, with cells exhibiting elongated spindle-shaped morphologies (Fig. 1F; Fig. S3A). Pyruvate was systematically less efficient than glucose at promoting NCC dispersion, with significantly reduced NCC outgrowths, cell densities and outward progressions at all concentrations tested (Fig. 1F–H; Fig. S3A–D). Maximal NCC dispersion and individual cell velocity were reached at 1–2 mM, whereas persistence was greater for concentrations ≥2 mM (Fig. S3E,G). Furthermore, pyruvate did not support NCC spreading over time at 0.1 mM, whereas it induced extensive NCC flattening at high concentrations (Fig. S3C). Therefore, throughout the study, pyruvate was used at 1 mM.

To identify possible additive or cooperating effects of glucose and pyruvate, we analyzed responses of NT explants to both compounds used in combination. We found that NCC outgrowth aspect, progression and cell morphologies were virtually indistinguishable from those observed in glucose alone (Fig. 1F–H; Fig. S3B), thereby establishing the predominant role of glucose in supporting NCC development. Finally, we analyzed the effect of 2-DG on NCC dispersion in culture. In its presence, NCCs initially segregated normally from the NT, and their outgrowth area and progression were close to those of untreated explants (Fig. 1F,G), but cells were less dense, with a spindle shape, as observed in pyruvate. Then, expansion of the population stalled almost completely (Fig. 1H), and NCCs started to round up, resulting in sparse cells occupying a limited area around the NT (Fig. 1F,G; Fig. S5A,B).

Finally, we investigated the expression of the glucose transporter Glut-1 and glycolytic enzymes in NCCs in culture. Glut-1 was expressed similarly on the surface of NCCs in glucose or in pyruvate (Fig. S1E), consistent with their capacity to adsorb glucose under both conditions, as judged on 2-NDBG uptake (Fig. S1E). Glut-1 and PFK mRNAs were low in NCCs and more strongly expressed in the NT (Fig. S1F), and they were both upregulated in the presence of glucose compared to pyruvate (Fig. S1F,G). In contrast, PGK-1 was strongly expressed in both the NT and migrating NCCs and was not affected by nutrient supplies (Fig. S1F).

Together, these results show that, like in vivo, trunk NCCs cultured in vitro can metabolize glucose using the glycolytic machinery and that glucose metabolism is necessary for their efficient dispersion. Moreover, pyruvate, a by-product of glycolysis, cannot be substituted for glucose for efficient migration.

Trunk NCCs display a typical OXPHOS signature during dispersion

To identify the metabolic pathways downstream to glucose uptake that are involved in trunk NCC dispersion in vitro, we performed metabolic profiling of NT explants by measuring mitochondrial respiration and glycolysis using a Seahorse XF analyzer and by quantifying the amount of ATP produced and lactate released at the end of the Seahorse assay (Fig. 2). Analyses of the bioenergetic profiles of individual explants exposed to glucose, pyruvate or a combination of both, associated with MitoStress assays, revealed that, in all conditions at 5 h after initiation of migration, a great majority of the explants display an OXPHOS signature characterized by a high oxygen consumption rate (OCR) and low extracellular acidification rate (ECAR) (Fig. 2A,B). In addition, no change in metabolic activity was observed after 24 h (Fig. 2B). As previously reported for chick cranial NCCs (Bhattacharya et al., 2020), quail cranial NT explants at 5 h exhibited a typical aerobic glycolysis profile characterized by a lower OCR and a high ECAR (Fig. 2B). Interestingly, in glucose, greater values of OCR and ECAR (Fig. 2A,D) and ATP production (Fig. 2F) were observed than in pyruvate. These results indicate that, unlike in cranial NCCs, glucose is mobilized in trunk NCCs to feed the mitochondrial OXPHOS pathway rather than for lactate production, and that a better yield in energy production is obtained for glycolysis coupled to OXPHOS than for OXPHOS fueled by pyruvate alone.

Fig. 2.

Trunk NCC display a typical OXPHOS signature during dispersion. (A) OCR and ECAR profiles in mitochondrial stress assays of individual trunk NT explants cultured for 5 h in medium with glucose, pyruvate or both, using a Seahorse analyzer and analyzed using GraphPad Prism. (B) Energetic maps of 5-h trunk and cranial explants in indicated nutrient conditions (left) and of 5-h and 24-h trunk explants in medium with glucose and pyruvate (right). (C) OCR and ECAR profiles of 5-h trunk explants in glucose and pyruvate medium before and after addition of metabolic inhibitors. (D) Energetic maps of 5-h explants in indicated nutrient conditions and in glucose and pyruvate medium with inhibitors. (E–G) Scatter plots of the OCR/ECAR ratio (E) and intracellular ATP level (F), and diagram of extracellular lactate produced (G) in 5-h trunk explants in indicated nutrient conditions and in glucose and pyruvate medium with inhibitors. Scatter plots with mean±s.e.m. were analyzed using one-way ANOVA followed by Dunnett's multiple comparison test relative to the condition with glucose and pyruvate. Data in A–F and in G were collected from at least three and two independent experiments, respectively. (n) indicates the number of explants analyzed. Error bars in A–D and G are s.d. *P<0.05, **P<0.01, ****P<0.0001; ns, not significantly different, P>0.05.

Fig. 2.

Trunk NCC display a typical OXPHOS signature during dispersion. (A) OCR and ECAR profiles in mitochondrial stress assays of individual trunk NT explants cultured for 5 h in medium with glucose, pyruvate or both, using a Seahorse analyzer and analyzed using GraphPad Prism. (B) Energetic maps of 5-h trunk and cranial explants in indicated nutrient conditions (left) and of 5-h and 24-h trunk explants in medium with glucose and pyruvate (right). (C) OCR and ECAR profiles of 5-h trunk explants in glucose and pyruvate medium before and after addition of metabolic inhibitors. (D) Energetic maps of 5-h explants in indicated nutrient conditions and in glucose and pyruvate medium with inhibitors. (E–G) Scatter plots of the OCR/ECAR ratio (E) and intracellular ATP level (F), and diagram of extracellular lactate produced (G) in 5-h trunk explants in indicated nutrient conditions and in glucose and pyruvate medium with inhibitors. Scatter plots with mean±s.e.m. were analyzed using one-way ANOVA followed by Dunnett's multiple comparison test relative to the condition with glucose and pyruvate. Data in A–F and in G were collected from at least three and two independent experiments, respectively. (n) indicates the number of explants analyzed. Error bars in A–D and G are s.d. *P<0.05, **P<0.01, ****P<0.0001; ns, not significantly different, P>0.05.

Then, we evaluated the impact of metabolic inhibitors (Fig. S1A) on the bioenergetic activity of trunk NT explants. 2-DG caused a sharp drop in OCR with no alteration in ECAR, shifting the metabolic profile toward metabolic quiescence (Fig. 2C–E). A similar result was obtained with oligomycin, an inhibitor of mitochondrial ATP synthase (Fig. 2C–E). Moreover, both 2-DG and oligomycin decreased OCR levels (Fig. 2D,E) and sharply reduced ATP production (Fig. 2F) but did not affect extracellular lactate production (Fig. 2G). Of interest, inhibitor effects were similar in all nutrient conditions, but in medium containing glucose and pyruvate, 2-DG reduced ATP production to the same levels as in glucose alone (Fig. 2F), suggesting that pyruvate cannot efficiently compensate for glycolysis inhibition.

These results demonstrate that, in trunk NT explants in culture, most of the ATP production relies primarily on glucose oxidation through OXPHOS, and reveal that OXPHOS inhibition does not cause metabolic rewiring to aerobic glycolysis, as observed previously in cranial NCCs (Bhattacharya et al., 2020), pointing to possible different metabolic regulatory processes in distinct NCC subpopulations.

OXPHOS plays a critical role in trunk NCC development

We next tested whether the OXPHOS pathway is required for trunk NCC development. Similar to what was seen with 2-DG, in ovo injection of oligomycin at 1 µM caused a significant decrease in Snail-2, Foxd-3 and Sox-10 expression (Fig. 3A) and a severe reduction in Foxd-3 expression in the dorsal NT at delamination after 5 h, then a massive decrease in the number of Sox-10+ cells during migration after 24 h, and finally a complete disruption of the sensory and sympathetic ganglia after 48 h (Fig. 3B,C; Fig. S4A–C). Moreover, at 5 µM oligomycin, patterning of the NT and paraxial mesoderm was strongly disrupted after 24 h, resulting in complete absence of the lower trunk (Fig. S4A–C).

Fig. 3.

OXPHOS plays a critical role in trunk NCC development. (A) Normalized levels of Snail-2, Foxd-3 and Sox-10 mRNAs in the trunk of embryos 5 h after oligomycin or vehicle (no-drug) injection measured by qRT-PCR. Measurements were performed in triplicates per embryo and gene, analyzed by two-way ANOVA Tukey post test relative to the no-drug condition. (B) Whole-mount views of the trunk of embryos 5 h, 24 h and 48 h after oligomycin or vehicle injection, hybridized for Foxd-3 and Sox-10 or immunolabeled for HNK-1. Vertical bars delineate the axial levels where NCC development is defective. Arrowheads, arrows and double arrows point at delaminating, migrating and differentiating NCCs, respectively. The corresponding global views of the embryos are shown in Fig. S4A. (C) Cross-sections of whole-mount ISH for Sox-10 through the caudal (left), mid (center), and anterior (right) trunk of embryos 24 h after oligomycin or vehicle injection. Arrows point at NCCs. (D) Overall aspect of NT explants in 5 h and 24 h cultures in glucose and pyruvate medium with metabolic inhibitors. (E) Scatter plot with mean±s.e.m. of NCC outgrowth area at 5 h and 24 h in conditions as in D, analyzed using one-way ANOVA followed by Dunnett's multiple comparison test relative to the condition with glucose and pyruvate. (F) Distance covered over time by the NCC migratory front in NT explants cultured as in D and subjected to video microscopy. t0 corresponds to onset of recording 2–4 h after initiation of culture. Inhibitors were applied immediately before t0. Front progression was analyzed using a custom GUI written in Matlab. Values in A and F are mean±s.e.m. Data are from at least three independent experiments. Images in B, C, and D are from representative experiments. (n) indicate the number of explants analyzed. *P<0.05, ***P<0.001, ****P<0.0001; ns, not significantly different, P>0.05. Scale bars: 200 µm (B); 100 µm (C,D). ao, aorta; dmt, dermamyotome; ec, ectoderm; nt, neural tube; pm, unsegmented paraxial mesoderm; sc, sclerotome; so, somite.

Fig. 3.

OXPHOS plays a critical role in trunk NCC development. (A) Normalized levels of Snail-2, Foxd-3 and Sox-10 mRNAs in the trunk of embryos 5 h after oligomycin or vehicle (no-drug) injection measured by qRT-PCR. Measurements were performed in triplicates per embryo and gene, analyzed by two-way ANOVA Tukey post test relative to the no-drug condition. (B) Whole-mount views of the trunk of embryos 5 h, 24 h and 48 h after oligomycin or vehicle injection, hybridized for Foxd-3 and Sox-10 or immunolabeled for HNK-1. Vertical bars delineate the axial levels where NCC development is defective. Arrowheads, arrows and double arrows point at delaminating, migrating and differentiating NCCs, respectively. The corresponding global views of the embryos are shown in Fig. S4A. (C) Cross-sections of whole-mount ISH for Sox-10 through the caudal (left), mid (center), and anterior (right) trunk of embryos 24 h after oligomycin or vehicle injection. Arrows point at NCCs. (D) Overall aspect of NT explants in 5 h and 24 h cultures in glucose and pyruvate medium with metabolic inhibitors. (E) Scatter plot with mean±s.e.m. of NCC outgrowth area at 5 h and 24 h in conditions as in D, analyzed using one-way ANOVA followed by Dunnett's multiple comparison test relative to the condition with glucose and pyruvate. (F) Distance covered over time by the NCC migratory front in NT explants cultured as in D and subjected to video microscopy. t0 corresponds to onset of recording 2–4 h after initiation of culture. Inhibitors were applied immediately before t0. Front progression was analyzed using a custom GUI written in Matlab. Values in A and F are mean±s.e.m. Data are from at least three independent experiments. Images in B, C, and D are from representative experiments. (n) indicate the number of explants analyzed. *P<0.05, ***P<0.001, ****P<0.0001; ns, not significantly different, P>0.05. Scale bars: 200 µm (B); 100 µm (C,D). ao, aorta; dmt, dermamyotome; ec, ectoderm; nt, neural tube; pm, unsegmented paraxial mesoderm; sc, sclerotome; so, somite.

When applied onto NT explants in culture, oligomycin affected initial NCC progression, with cohesive cells and reduced outgrowth area (Fig. 3D,E; Fig. S5A). Then, NCC expansion decreased strongly, producing a small outgrowth of sparse, poorly spread cells at 24 h (Fig. 3D,E; Fig. S5A,B). We also tested the effect of other OXPHOS inhibitors, such as rotenone combined with antimycin-A (Rot.-AA), as well as an inhibitor of the mitochondrial pyruvate carrier UK-5099 (Fig. S1A). Both Rot.-AA and UK5099 had a similar effect to oligomycin except that reduction in dispersion was more immediate and massive with Rot.-AA (Fig. 3D–F; Fig. S5A,B). Oligomycin also exerted a much stronger effect on NCC cultured in pyruvate than in glucose, with abrogated cell dispersion (Fig. S5C). Likewise, UK-5099 affected severely NCC dispersion and spreading in pyruvate and more moderately in glucose (Fig. S5C). Finally, we verified that, consistent with their aerobic glycolytic profile, cranial NCC dispersion in culture is strongly affected by 2-DG and only weakly by oligomycin (Fig. S5D), as also shown in the chick (Bhattacharya et al., 2020). These results establish that active OXPHOS is required for trunk NCC development and more dispensable in cranial NCCs, further illustrating the metabolic specificities of each population.

PPP is mobilized during trunk NCC dispersion

Rapid and sustained expansion of trunk NCCs most likely requires an intense biosynthetic activity beside high bioenergetic needs. We therefore investigated whether NCC development also relies on PPP using 6-aminonicotinamide (6-AN), a blocker of PPP. In culture, 6-AN showed no apparent effect on the initial dispersion of NCCs, but the progression of cells slowed down after 5–8 h, and a strongly reduced outgrowth with many round cells was observed (Fig. 3D–F; Fig. S5A,B). 6-AN greatly affected NCC dispersion in glucose but had little effect in pyruvate (Fig. S5C). Moreover, in Seahorse analyses, 6-AN had no effect on OCR and ATP levels (Fig. 2C–F), suggesting a minor contribution of PPP to energy production. When applied in vivo, 6-AN did not show much effect on NCC delamination, as judged by assessing the Foxd-3 expression pattern, but it inhibited markedly NCC migration after 24 h, similar to what was seen with 2-DG and oligomycin (Fig. S4B,D). These results thus show that beside glucose oxidation, PPP contributes to trunk NCC development, thereby suggesting the involvement of multiple metabolic pathways, possibly acting in a cooperative manner to support coordination between multiple cellular processes: EMT during delamination, changes of substratum and cell adhesion, activation of the locomotion machinery, and proliferation.

Glucose oxidation is required for EMT during trunk NCC delamination

For delamination, we evaluated first the NT capacity to produce NCCs over time in 20 h-culture under different nutrients and in the presence of metabolic inhibitors. We found that the duration of NCC production was significantly shorter with pyruvate than glucose and considerably reduced by all inhibitors (Fig. 4A). Next, we developed an assay suitable for discriminating delaminated from migrating NCCs on 5 h explants (Fig. 4B). We observed that the number of delaminated NCCs was much higher in glucose and glucose supplemented with pyruvate than in pyruvate alone and was halved by 2-DG, oligomycin or Rot.-AA, but not by 6-AN (Fig. 4C). Immunolabeling showed that Snail-2, a major EMT player (Duband, 2010; Nieto, 2002), was present at high levels in a large proportion of the delaminated NCCs and at lower levels in migrating cells in glucose-containing medium, but was strongly diminished in both NCC compartments in explants cultured in pyruvate or with 2-DG (Fig. 4D,E). In contrast, and consistent with an unchanged delamination rate in response to 6-AN (Fig. 4C), NCCs displayed similar Snail-2 staining under PPP inhibition to untreated cells in glucose-containing medium (Fig. 4D,E). Intriguingly, with OXPHOS inhibitors, and particularly with Rot.-AA, Snail-2 completely disappeared in delaminated NCCs but remained high in migrating cells (Fig. 4D,E). Expression of Snail-2 mRNA was strongly diminished in pyruvate or with 2-DG and halved by oligomycin (Fig. 4F,G), consistent with the reduction of Snail-2 protein. Conversely, Snail-2 expression was not much altered by Rot.-AA (Fig. 4F,G), probably as a result of its increase in the migrating pool. Together, these data indicate that glucose oxidation is required for NCC delamination and Snail-2 expression. In addition, PPP is dispensable for delamination but is necessary for long-term production of NCCs, as indicated by the experiments using pyruvate alone or 6-AN.

Fig. 4.

Glucose oxidation is required for Snail-2 expression during trunk NCC delamination. (A) Scatter plot of the NCC production duration from NT explants in indicated nutrient conditions and in glucose and pyruvate medium with metabolic inhibitors. (B) Schematic description of the delamination assay. (C) Scatter plot of the number of delaminated NCCs per explant in indicated nutrient conditions and in glucose and pyruvate medium with metabolic inhibitors. (D) Immunofluorescence staining for Snail-2 in delaminated (delam.) and migrating (migr.) NCCs in conditions as in C. (E) Scatter plot of the proportion of Snail-2+ NCCs in delaminated cells per explant in conditions as in C. The scatter plots with mean±s.e.m. were analyzed using one-way ANOVA followed by Dunnett's multiple comparison test and relative to the medium in both nutrients without inhibitor. (n) indicate the number of explants analyzed. (F) Normalized levels of Snail-2 mRNAs measured by qRT-PCR in explants in conditions as in C except that the NT was not removed before mRNA extraction. Measurements were performed in triplicates per condition and gene, and analyzed using an unpaired two-tailed t-test, and expressed as mean±s.e.m. (n) indicate the number of experiments. (G) ISH for Snail-2 mRNAs in trunk NT explants in conditions as in F. Arrowheads point at pre-migratory NCCs in the dorsal NT and arrows point at delaminated NCC in both sides of the NT. Data were collected from at least three independent experiments. Images in D and G are from representative experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; ns, not significantly different, P>0.05. Scale bars: 50 µm (D); 100 µm (G).

Fig. 4.

Glucose oxidation is required for Snail-2 expression during trunk NCC delamination. (A) Scatter plot of the NCC production duration from NT explants in indicated nutrient conditions and in glucose and pyruvate medium with metabolic inhibitors. (B) Schematic description of the delamination assay. (C) Scatter plot of the number of delaminated NCCs per explant in indicated nutrient conditions and in glucose and pyruvate medium with metabolic inhibitors. (D) Immunofluorescence staining for Snail-2 in delaminated (delam.) and migrating (migr.) NCCs in conditions as in C. (E) Scatter plot of the proportion of Snail-2+ NCCs in delaminated cells per explant in conditions as in C. The scatter plots with mean±s.e.m. were analyzed using one-way ANOVA followed by Dunnett's multiple comparison test and relative to the medium in both nutrients without inhibitor. (n) indicate the number of explants analyzed. (F) Normalized levels of Snail-2 mRNAs measured by qRT-PCR in explants in conditions as in C except that the NT was not removed before mRNA extraction. Measurements were performed in triplicates per condition and gene, and analyzed using an unpaired two-tailed t-test, and expressed as mean±s.e.m. (n) indicate the number of experiments. (G) ISH for Snail-2 mRNAs in trunk NT explants in conditions as in F. Arrowheads point at pre-migratory NCCs in the dorsal NT and arrows point at delaminated NCC in both sides of the NT. Data were collected from at least three independent experiments. Images in D and G are from representative experiments. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; ns, not significantly different, P>0.05. Scale bars: 50 µm (D); 100 µm (G).

Multiple metabolic pathways cooperate for trunk NCC locomotion

We next investigated the contribution of glucose metabolism to trunk NCC locomotion in culture using video-microscopy analyses. In the glucose condition, NCCs generally exhibited high velocity and persistence over 18 h, leading to linear trajectories oriented perpendicular to the NT (Fig. 5A; Fig. S6A,B), whereas in pyruvate, velocity and persistence decreased gradually, leading to shorter trajectories (Fig. 5A; Fig. S6A,B). In absence of both glucose and pyruvate, NCC migration tracks were very short, resulting from low velocity associated with fair persistence (Fig. 5A; Fig. S6A,B), whereas in both nutrients, NCCs displayed high and sustained velocity and persistence over time and long trajectories as in glucose alone. With 2-DG, cell trajectories were at first relatively linear but became more random with time (Fig. S6C), the most striking effect being a rapid and continuous decrease of velocity over time resulting in migration arrest (Fig. 5B). Cell trajectories were generally not affected very much by oligomycin (Fig. S6C), whereas velocity was decreased during the first 6 h. Subsequently NCC migration became extremely random with decreased persistence and higher velocity (Fig. 5B). Interestingly, Rot.-AA showed a much stronger effect than oligomycin. Trajectories were very short because velocity was almost knocked out immediately after addition and remained weak during the next 8 h (Fig. 5B, Fig. S6C), but as with oligomycin, velocity increased progressively with time and persistence decreased. In UK-5099, cell trajectories were more random and associated with low persistence (Fig. 5B; Fig. S6C). Finally, 6-AN affected velocity strongly and gradually at least during the first 8 h (Fig. 5B, Fig. S6C). These results therefore illustrate the critical contribution of glucose oxidation to NCC locomotion and also reveal that this process relies on additional metabolic processes, notably PPP.

Fig. 5.

Glucose metabolism is crucial for trunk NCC locomotion, adhesion and mechanical properties. (A–D) Effect of nutrients and metabolic inhibitors on NCC locomotion and adhesion. (A,B) Scatter plots of the velocity throughout the duration of the experiment (left) and graphs of the mean velocity over time (right) of individual NCCs cultured in indicated nutrient conditions (A) and in glucose and pyruvate medium with inhibitors (B), as measured using video-microscopy using Metamorph software. t0 corresponds to onset of recording 2–4 h after initiation of culture. Inhibitors were applied immediately before recording. (C) Immunofluorescence staining for paxillin (green) with DAPI visualization of nuclei (blue) of NCCs at 5 h in indicated nutrient conditions (left) and in glucose and pyruvate medium with metabolic inhibitors (right). (D) Scatter plots of the number of substrate adhesions normalized to NCC count at 5 h and 24 h in conditions as in C. (E–G) Effect of nutrients on NCC adhesion and migration on gels of different rigidities. (E,F) Scatter plots of the number of substrate adhesions normalized to NCC count at 5 h (E) and of velocity and persistence of individual NCCs over 16 h (F) in culture on stiff (6.3 kPa) or soft (1.2 kPa) PAA gels in indicated nutrient conditions. (G) Distance covered over time by the NCC migratory front in explants as in F, subjected to video-microscopy and measured using MATLAB. (H–J) Effect of nutrients on trunk NCC mechanical properties on stiff and soft gels. (H) Scatter plot of NCC stiffness in 5-h cultures in the indicated nutrient and gel conditions. The green bars indicate the stiffness of the gels. (I) Merged images of phase contrast with fluorescent for red-stained nuclei and pseudo-color traction map images of NCCs in indicated conditions. Higher traction levels are in orange–yellow. (J) Scatter plot of NCC traction activity expressed as stress per individual cell. Each dot represents the average stress obtained from an independent TFM assessment on a NCC outgrowth, divided by the number of cell nuclei. Scatter plots with mean±s.e.m. were analyzed using one-way ANOVA followed by Dunnett's multiple comparison test, relative to glucose and pyruvate medium without inhibitor (A–D), or relative to condition on stiff gel (E–J). *P<0.05, **P<0.01, ****P<0.0001; ns, not significantly different, P>0.05. In A, B and F, (n) indicates the number of NCCs analyzed in several explants, with for each explant 20 cells analyzed at the periphery of the outgrowth. In C–E and G, (n) indicates the number of explants, and in H, the number of NCCs analyzed. Data presented were from at least three independent experiments. Error bars are s.e.m. Images representative from at least three and seven experiments for C,D, and I, respectively. Scale bars: 20 µm (C); 50 µm (I).

Fig. 5.

Glucose metabolism is crucial for trunk NCC locomotion, adhesion and mechanical properties. (A–D) Effect of nutrients and metabolic inhibitors on NCC locomotion and adhesion. (A,B) Scatter plots of the velocity throughout the duration of the experiment (left) and graphs of the mean velocity over time (right) of individual NCCs cultured in indicated nutrient conditions (A) and in glucose and pyruvate medium with inhibitors (B), as measured using video-microscopy using Metamorph software. t0 corresponds to onset of recording 2–4 h after initiation of culture. Inhibitors were applied immediately before recording. (C) Immunofluorescence staining for paxillin (green) with DAPI visualization of nuclei (blue) of NCCs at 5 h in indicated nutrient conditions (left) and in glucose and pyruvate medium with metabolic inhibitors (right). (D) Scatter plots of the number of substrate adhesions normalized to NCC count at 5 h and 24 h in conditions as in C. (E–G) Effect of nutrients on NCC adhesion and migration on gels of different rigidities. (E,F) Scatter plots of the number of substrate adhesions normalized to NCC count at 5 h (E) and of velocity and persistence of individual NCCs over 16 h (F) in culture on stiff (6.3 kPa) or soft (1.2 kPa) PAA gels in indicated nutrient conditions. (G) Distance covered over time by the NCC migratory front in explants as in F, subjected to video-microscopy and measured using MATLAB. (H–J) Effect of nutrients on trunk NCC mechanical properties on stiff and soft gels. (H) Scatter plot of NCC stiffness in 5-h cultures in the indicated nutrient and gel conditions. The green bars indicate the stiffness of the gels. (I) Merged images of phase contrast with fluorescent for red-stained nuclei and pseudo-color traction map images of NCCs in indicated conditions. Higher traction levels are in orange–yellow. (J) Scatter plot of NCC traction activity expressed as stress per individual cell. Each dot represents the average stress obtained from an independent TFM assessment on a NCC outgrowth, divided by the number of cell nuclei. Scatter plots with mean±s.e.m. were analyzed using one-way ANOVA followed by Dunnett's multiple comparison test, relative to glucose and pyruvate medium without inhibitor (A–D), or relative to condition on stiff gel (E–J). *P<0.05, **P<0.01, ****P<0.0001; ns, not significantly different, P>0.05. In A, B and F, (n) indicates the number of NCCs analyzed in several explants, with for each explant 20 cells analyzed at the periphery of the outgrowth. In C–E and G, (n) indicates the number of explants, and in H, the number of NCCs analyzed. Data presented were from at least three independent experiments. Error bars are s.e.m. Images representative from at least three and seven experiments for C,D, and I, respectively. Scale bars: 20 µm (C); 50 µm (I).

Glucose metabolism coordinates trunk NCC substrate and cell adhesions

To gain insight into the role of glucose metabolism on trunk NCC adhesion in culture, we examined by immunolabeling the cellular localization of paxillin and β-catenin, two cytoplasmic partners of the integrin and cadherin adhesion complexes, respectively. NCCs developed substrate adhesions in all conditions at 5 h but they were reduced in the absence of glucose and pyruvate (Fig. 5C,D). At 24 h, their number increased with glucose whereas it remained stable in pyruvate or in the absence of nutrients (Fig. 5D). In addition, the number of substrate adhesions was significantly reduced with 2-DG and increased with oligomycin, consistent with changes in cell spreading (Fig. 5C,D). Substrate adhesions differed also in their shape and area with the nutrient conditions – they were larger and more elongated in pyruvate compared to in the other conditions (Fig. S6E). Finally, NCCs were always engaged in cell–cell contacts with their neighbors in medium containing glucose, whereas in pyruvate, they remained mostly as individuals (Fig. S6F). The number of cell–cell adhesions at 5 h was increased substantially with oligomycin and Rot-AA (Fig. S6G,H). These data indicate that the balance between substrate and cell–cell adhesion in trunk NCCs is influenced by glucose metabolism.

Trunk NCC mechanics and their response to external stiffness are modulated by nutrient inputs

NCCs are able to sense the extracellular matrix rigidity in their environment, which in turn modulates their adhesion and migration capacity (Barriga et al., 2018; Chevalier et al., 2016; Marchant et al., 2022; Shellard and Mayor, 2021). To determine whether mechanotransduction activities in trunk NCCs are under metabolic control, we investigated the impact of nutrients on their adhesive and migratory response in culture to fibronectin-coated polyacrylamide gels with different stiffness in the range of avian tissue rigidity at early embryonic stages, from soft (1.2 kPa) to stiff (6.3 kPa) (Chevalier et al., 2016). As on glass (Fig. 5C), NCCs on stiff and soft gels developed substrate adhesions irrespective of the nutrient supplied (Fig. S6I). Moreover, we found important discrepancies in the NCC response to gel rigidity with the nutrients (Fig. 5E–G). In glucose, the number of substrate adhesions (Fig. 5E) and the progression of the population were not significantly different between stiff and soft gels, whereas velocity and persistence were only slightly lower on soft gels (Fig. 5F,G). In contrast, on soft gels, in medium containing pyruvate or both nutrients, NCCs displayed fewer substrate adhesions, and their velocity and the progression of the population were significantly lower (Fig. 5E-G).

We then studied NCC stiffness and traction activity on the gel using atomic force microscopy (AFM) and traction force microscopy (TFM) (Fig. 5H–J). Measurement of cellular stiffness revealed that with glucose, NCCs adapted their stiffness to that of the substrate in both soft and stiff gels, whereas in pyruvate they failed to increase their stiffness to the level of the stiff gel (Fig. 5H). NCC traction activity in soft and stiff gels (expressed as stress/cell) was constant in glucose, whereas in the presence of pyruvate, it was higher in stiff gels (Fig. 5I,J). By calculating the product of the average distance covered per hour and the average force (stress/cell/area) we estimated the energy produced when NCCs moved on the gels under the different nutrient conditions (Table 1). Energy produced was similar in all nutrients on stiff gels, whereas it was lower on soft gels in pyruvate compared to glucose in which it was almost unchanged.

Table 1.

Energy produced by trunk NCC on soft and stiff gels in different nutrient conditions

Energy produced by trunk NCC on soft and stiff gels in different nutrient conditions
Energy produced by trunk NCC on soft and stiff gels in different nutrient conditions

These data indicate that trunk NCCs are mechanoresponsive as described previously for cranial and enteric NCCs (Barriga et al., 2018; Chevalier et al., 2016; Marchant et al., 2022; Shellard and Mayor, 2021), but also that their response is determined by specific metabolic activities. In glucose, owing to their capacity to accommodate their stiffness and energy production to their environment, NCCs are endowed with the ability to migrate under diverse biophysical constraints. In pyruvate, NCCs are less prone to modulating their stiffness and negatively respond to modifications in the rigidity of their environment by decreasing their adhesive and migratory response.

Multiple metabolic pathways are required for cell cycle progression and proliferation of trunk NCCs

In contrast to what is seen for many motile cells, NCCs maintain active cell division throughout migration, and this process contributes strongly to expansion of the population (Ridenour et al., 2014). On the other hand, cell cycle progression from G1-S phase is required for delamination (Burstyn-Cohen and Kalcheim, 2002). We therefore next analyzed the influence of glucose metabolism on trunk NCC cycling and proliferating activity in culture using EdU incorporation. During delamination, numerous EdU+ NCCs could be identified along the NT apical side (Fig. S7A), and addition of glycolysis or OXPHOS inhibitors dramatically reduced their numbers (Fig. S7A). During migration, whereas the EdU-staining intensity was weak and the proportion of EdU+ NCCs small at 5 h and minimal after 24 h in absence of glucose and pyruvate (Fig. S7B), these two parameters remained very high over time in glucose alone and in glucose and pyruvate (Fig. S7B). In pyruvate alone, in contrast, the proportion of EdU+ NCCs and EdU incorporation were high at 5 h but the proportion of EdU+ NCCs was dramatically reduced after 24 h (Fig. S7B). All metabolic inhibitors decreased both the EdU-staining intensity and the proportion of EdU+ NCCs at both 5 h and 24 h (Fig. S7C). These data indicate that multiple metabolic pathways are implicated in a cooperative manner in the control of cell cycling and proliferation of trunk NCCs.

Glucose and pyruvate differently influence trunk NCC pluripotency and differentiation potential

To investigate whether metabolic activity affects NCC stem cell capacity and influences their differentiation potential, we analyzed in culture the expression patterns of the transcription factors Sox-10 and Foxd-3, which have been shown to be essential regulators of NCC pluripotency in addition to their role in delamination and migration (Lukoseviciute et al., 2018; Schock and LaBonne, 2020; Simões-Costa et al., 2012). At 5 h (Fig. 6A,B,E), levels of Sox-10 protein and Sox-10 and Foxd-3 mRNAs were much reduced in delaminating and migrating NCCs cultured with pyruvate compared with levels in those cultured with glucose, in which the vast majority of migrating NCCs exhibited Sox-10+ nuclei. 2-DG strongly diminished the number of Sox-10+ cells, whereas oligomycin and Rot.-AA caused an overall reduction of Sox-10 intensity in all NCCs (Fig. 6C). Likewise, Sox-10 and Foxd-3 mRNA levels dropped strongly with 2-DG (Fig. 6D), as also observed in vivo (Fig. 1C–E), whereas those for Sox-10 showed a trend toward a reduced expression that failed statistical testing with oligomycin and Rot-AA. In addition, Foxd-3 expression was selectively repressed by 2-DG, oligomycin and Rot.-AA in pre-migratory NCCs but less so in delaminating and migrating cells (Fig. 6E). At 24 h, repression of Sox-10 and Foxd-3 expression was even more pronounced in absence of glucose and in the presence of metabolic inhibitors, including 6-AN (Fig. S8A,B).

Fig. 6.

Glucose and pyruvate differently influence trunk NCC pluripotency and differentiation potential. (A,C) Immunofluorescence staining for Sox-10 in migrating NCCs (A) and in delaminated (delam.) versus migrating (migr.) NCCs (green, with DAPI visualization of nuclei in blue, C), after 5-h cultures in indicated nutrient conditions and in glucose and pyruvate medium with inhibitors. (B,D) Normalized levels of Foxd-3 and Sox-10 mRNAs in explants cultured for 5 h as in A and C, measured by qRT-PCR. Measurements were performed in triplicates per embryo and gene, analyzed by using an unpaired two-tailed t-test, and presented as mean±s.e.m. (n) indicates the number of experiments. (E) ISH for Foxd-3 mRNAs in NT explants cultured for 5 h in indicated nutrient conditions and in glucose and pyruvate medium with inhibitors. Arrowheads point at pre-migratory NCCs in the dorsal NT and arrows point at delaminated NCC in both sides of the NT. (F) Scatter plots with mean±s.e.m. of the proportion of Sox-10+, HNK-1+, Tuj-1+, MITF+, and α-SMA+ NCC in explants cultured for 1–3 days (d) under the indicated nutrient conditions, and analyzed using an unpaired two-tailed t-test with the comparisons for each marker in one nutrient made relative to the condition with glucose and pyruvate at the same time. (n) indicates the number of explants analyzed. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; ns, not significantly different, P>0.05. Data were collected from at least three independent experiments, and the images in A and C are representative images from these independent experiments. Scale bars: 50 µm (A,C); 100 µm (E).

Fig. 6.

Glucose and pyruvate differently influence trunk NCC pluripotency and differentiation potential. (A,C) Immunofluorescence staining for Sox-10 in migrating NCCs (A) and in delaminated (delam.) versus migrating (migr.) NCCs (green, with DAPI visualization of nuclei in blue, C), after 5-h cultures in indicated nutrient conditions and in glucose and pyruvate medium with inhibitors. (B,D) Normalized levels of Foxd-3 and Sox-10 mRNAs in explants cultured for 5 h as in A and C, measured by qRT-PCR. Measurements were performed in triplicates per embryo and gene, analyzed by using an unpaired two-tailed t-test, and presented as mean±s.e.m. (n) indicates the number of experiments. (E) ISH for Foxd-3 mRNAs in NT explants cultured for 5 h in indicated nutrient conditions and in glucose and pyruvate medium with inhibitors. Arrowheads point at pre-migratory NCCs in the dorsal NT and arrows point at delaminated NCC in both sides of the NT. (F) Scatter plots with mean±s.e.m. of the proportion of Sox-10+, HNK-1+, Tuj-1+, MITF+, and α-SMA+ NCC in explants cultured for 1–3 days (d) under the indicated nutrient conditions, and analyzed using an unpaired two-tailed t-test with the comparisons for each marker in one nutrient made relative to the condition with glucose and pyruvate at the same time. (n) indicates the number of explants analyzed. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001; ns, not significantly different, P>0.05. Data were collected from at least three independent experiments, and the images in A and C are representative images from these independent experiments. Scale bars: 50 µm (A,C); 100 µm (E).

We next evaluated the influence of glucose metabolism on the capacity of trunk NCCs to differentiate in culture. After 3 days, several populations of cells were identified: Sox-10+ and HNK-1+ NCC progenitors, characterized by their stellate shape, Tuj-1+ neurons displaying long cell processes, as well as other cells presenting a great diversity of morphologies, including MITF+ melanoblasts and large flattened α-SMA+ myofibroblasts (Fig. S8C). In glucose-containing medium, the proportion of NCC progenitors remained high over time, whereas the number of neurons and melanoblasts increased up to 20% and 30–40%, respectively (Fig. 6F). In contrast, in pyruvate medium, the proportion of NCC progenitors dropped to less than 50% with time at the benefit of myofibroblasts, reaching 40% of the population (Fig. 6F). Of note, the number of melanoblasts did not increase much in pyruvate medium, whereas the number of HNK1+ cells was constantly very low (Fig. 6F). A likely explanation for this persistent absence of HNK-1 staining is that the HNK-1 epitope is a glycoconjugate moiety needing glucose for its biosynthesis (Tucker et al., 1984).

These results indicate that glucose metabolism is required for maintenance of high levels of pluripotency markers during NCC migration and that NCC differentiation programs are influenced by nutrient inputs, with glucose favoring neural and melanoblast differentiation, whereas pyruvate promotes a limited range of phenotypes where myofibroblasts are predominant.

Our study illustrates the critical role of glucose metabolism in quail NCC development. We report that glucose is required both in vivo and in in vitro culture for quail trunk NCC dispersion. In addition, trunk NCCs display a typical OXPHOS signature, and glucose oxidation through glycolysis coupled to OXPHOS plays a critical role in their dispersion. Blockers of glycolysis and OXPHOS affected every aspect of NCC development associated with a sharp drop in mitochondrial respiration and ATP production. Additionally, our data establish that OXPHOS is not sufficient to support all the cellular events involved in this process, as systematic inhibition of glycolysis compromised NCC dispersion, even in the presence of pyruvate to support OXPHOS. Likewise, a PPP inhibitor strongly affected NCC dispersion despite a minimal effect on mitochondrial respiration and ATP production. Finally, pyruvate alone failed to support complete dispersion, division and differentiation of NCCs, whereas glucose could elicit the complete cellular response of NCCs (i.e. delamination, adhesion, migration, proliferation, maintenance of stemness and widespread differentiation). Glucose also potentiates NCC stiffness adaptation to that of their microenvironment and supports optimal dispersion, illustrating the prominent role of glucose metabolism in NCC adaptation to physical changes in the environment.

Thus, in contrast to the general assertion that OXPHOS has a bioenergetic role in differentiated cells and that aerobic glycolysis has a biosynthetic role in proliferating non-differentiated stem cells (Ito and Suda, 2014; Shyh-Chang et al., 2013), trunk NCC development relies on the concerted and sustained mobilization of all metabolic pathways dependent on glucose metabolism (i.e. glycolysis, OXPHOS and PPP, cooperating together; see model in Fig. 7). These findings illustrate the necessity for the NCC population to make the most of the potential of the carbon metabolism route, so that they are endowed with the capacity to meet the metabolic demands required for the coordinated execution of diverse cellular events. How NCCs maintain balanced activities of the different metabolic pathways is currently unknown. This could be achieved through regulated levels of metabolic enzymes and products of each pathway, under the control of intrinsic genetic programs as well as extrinsic factors, including nutrient availability and demand (Lempradl et al., 2015). This view is supported by our observations as well as a recent report (Oginuma et al., 2017) that glycolytic enzymes are not uniformly expressed in the embryo at the time of NCC dispersion.

Fig. 7.

Schematic representation of the integration of carbon metabolism in NCC development. During development, trunk NCCs maximize the potential of the glucose pathway, so that they are endowed with the capacity to meet the high metabolic demands suitable for the coordinated execution of diverse cellular programs in a rapidly evolving environment (EMT, migration, proliferation, survival, adaptation to spatial and mechanical constraints, fate decision and differentiation). This is achieved through the concerted and sustained mobilization of all metabolic pathways downstream of glucose uptake, including glycolysis, OXPHOS, and PPP, cooperating together to produce ATP and nucleotides as bioenergetic and biosynthetic supplies and possibly other metabolites, such as NADP, NADPH and reactive oxygen species (ROS). All together, these metabolites contribute to regulate gene networks, promote cytoskeletal network dynamics, and stimulate cell cycling and proliferation.

Fig. 7.

Schematic representation of the integration of carbon metabolism in NCC development. During development, trunk NCCs maximize the potential of the glucose pathway, so that they are endowed with the capacity to meet the high metabolic demands suitable for the coordinated execution of diverse cellular programs in a rapidly evolving environment (EMT, migration, proliferation, survival, adaptation to spatial and mechanical constraints, fate decision and differentiation). This is achieved through the concerted and sustained mobilization of all metabolic pathways downstream of glucose uptake, including glycolysis, OXPHOS, and PPP, cooperating together to produce ATP and nucleotides as bioenergetic and biosynthetic supplies and possibly other metabolites, such as NADP, NADPH and reactive oxygen species (ROS). All together, these metabolites contribute to regulate gene networks, promote cytoskeletal network dynamics, and stimulate cell cycling and proliferation.

Our data reveal that trunk NCCs display metabolic requirements and features toward glucose distinct from their cranial counterparts. Indeed, quail cranial NCCs display a Warburg effect similar to that seen in chick cranial NCCs (Bhattacharya et al., 2020). Thus, although trunk and cranial NCCs possess in common a number of cellular and molecular features, such as signaling pathways, EMT and migration properties, they do not share the same metabolic status. This reflects the strong adaptability of NCCs to a great variety of environmental conditions throughout development, at the origin of the diversity in their migratory behaviors and fates. Indeed, difference in metabolic activity between cranial and trunk NCCs can also reside in the local O2 concentration in their environment. As for cancer cells in which hypoxia is known to promote metastatic spread (Semenza, 2012), cranial NCCs reside in a hypoxic milieu, and exposure to high O2 levels or knockdown of the hypoxia-inducible factor HIF-1 cause attenuation of their production associated with strong reduction of Snail-2 expression (Scully et al., 2016). Trunk NCC dispersion in contrast is intimately associated with blood vasculature (Schwarz et al., 2009; Thiery et al., 1982), and it is not affected by high O2 levels or by deletion of the HIF-1-encoding gene (Iyer et al., 1998; Morriss and New, 1979).

We could not attribute a specific role to any metabolic pathway in the basic cellular processes of adhesion, locomotion and division. For example, despite PPP having a well-recognized role in nucleotide synthesis and maintenance of cellular redox balance (Patra and Hay, 2014), this pathway was not recruited solely for NCC proliferation but significantly contributed to cell velocity and persistence of movement as well. Likewise, OXPHOS and ATP production in mitochondria were not exclusively aimed at supporting active migration, but also played a key role in EMT and proliferation. A likely explanation is that these processes are extremely demanding in both bioenergetic and biosynthetic supplies (Salazar-Roa and Malumbres, 2017; Zanotelli et al., 2021). Another reason is that enzymes, metabolites and byproducts of metabolic pathways are known to influence cellular programs independently of their canonical bioenergetics and biosynthetic roles (Miyazawa and Aulehla, 2018). For example, NADH produced through the tricarboxylic acid (TCA) cycle protects cells from oxidative stress (Bigarella et al., 2014; Khacho et al., 2016). Likewise, similar to what has been found for neurons (Herrero-Mendez et al., 2009), the use of the PPP for metabolizing glucose can ensure NADPH production for the maintenance of a cellular redox balance. TCA cycle intermediates, in particular acetyl-CoA and α-ketoglutarate, are involved in epigenetic regulation through histone acetylation and methylation (Kaelin and McKnight, 2013). Regulation of gene expression during NCC formation is under tight epigenetic control, involving DNA methylation, histone modifications and ATP-dependent chromatin remodelers (Hu et al., 2014). Our finding of alterations of developmental programs, such as delamination and maintenance of stemness, upon manipulating nutrient availability and metabolic pathways in NCCs is therefore in favor of a direct role of cellular metabolism in regulation of nuclear transcription programs in NCCs through epigenetic regulation (Boon et al., 2020; Traxler et al., 2021; Tsogtbaatar et al., 2020).

In conclusion, our data show that NCCs in the embryonic trunk region rely primarily on glucose oxidation through glycolysis and mitochondrial respiration for energy production, and mobilize a large range of metabolic pathways downstream to glucose uptake to meet their bioenergetics and biosynthetic needs as well as to instruct their gene circuits driving their behavior and fate. These findings therefore reveal the intricate integration of cellular metabolism to basic cellular processes underlying cell behavior. How these metabolic pathways are regulated to accommodate the different steps of NCC formation both temporally and spatially remains to be investigated.

Reagents

Bovine plasma fibronectin (1 mg/ml stock) was from Sigma (cat. no. F1141). Dispase II at 5 U/ml stock solution in Hanks' balanced saline was from Stemcell Technologies (cat. no. 07913). The following chemicals were prepared and used according to the manufacturers' guidelines: 2-deoxy-D-glucose (2-DG, Sigma, cat. no. D8375), oligomycin-A (Sigma, cat. no. 75351), rotenone (Sigma, cat. no. R8875), antimycin-A (Sigma, cat. no. A8674), FCCP (Sigma, cat. no. C2920), 6-aminonicotinamide (6-AN, Cayman Chemicals, cat. no. 10009315) and UK-5099 (Tocris Biosciences, cat. no. 4186). 2-DG was prepared as a 1 M stock solution in glucose-, pyruvate- and glutamine-free DMEM (Gibco, cat. no. A14430-01) and used at 5–10 mM for in vitro experiments and at 10–25 mM for in ovo injections; oligomycin, rotenone-antimycin-A (Rot.-AA), and FCCP were prepared in the vehicle DMSO at 10 mM, diluted as stock solution in culture medium at 100 µM, and used at 1 µM for in vitro experiments and at 1-5 µM for in ovo injections; UK-5099 was prepared as 100 mM solution in DMSO and used at 100 µM; and 6-AN was prepared as 100 mM solution in DMSO and used at 100 µM for in vitro and in vivo experiments, and at 500 µM in Seahorse analyses. As we previously showed that a 1:1000 dilution of DMSO has no effect on NCC behavior (Monier-Gavelle and Duband, 1995), in experiments using this dilution, the corresponding controls were performed without addition of DMSO. When DMSO dilution was below 1:1000, the same dilution of DMSO but without the drug was used for controls.

Embryos

Quail (Coturnix coturnix Japonica) embryos were used throughout the study. Fertilized quail eggs were purchased from a local farm (La Caille de Chanteloup, Corps Nuds, France). Eggs were incubated at 37–38°C until embryos reached the desired developmental stages. Embryos were staged using the Hamburger and Hamilton (HH) chart (Hamburger and Hamilton, 1951) based on the number of somite pairs. All animal experiments were performed according to approved guidelines.

In ovo injection of drugs

In vivo loss-of-function experiments were performed on embryos at stage HH 12–13 (15–16 somite pairs). After removing 1 ml of albumin from the egg and opening of the shell with curved dissecting scissors to expose the embryo to the experimenter, the drugs were injected at the tip of the tail bud of the embryo into the lumen of the NT and under the vitelline membrane in the vicinity of the NT over the unsegmented region and the last 3–5 somites (10 μl delivered per injection). For controls, embryos were injected with the vehicle only. The eggs were sealed with tape and were further incubated at 38°C in a moist atmosphere for 5–48 h. Monitoring of NCC development was performed by following expression of marker genes by whole mount ISH at 5, 24 and 48 h followed by sectioning and qRT-PCR at 5 h (see below).

Cell culture

Generation of NCC primary cultures

Trunk and cranial NCC cultures were produced from NT explants obtained from quail embryos at stages HH 13–15 (19–25 somite pairs) and at stages HH 8–9 (5–8 somite pairs), respectively (Duband et al., 2020). After opening of the eggshell, the yolk was transferred into phosphate-buffered saline (PBS). The embryo was cut off from the yolk and transferred into PBS in an elastomer-containing dish. An embryo portion of ∼750-µm long was excised at the level of the last five somites for trunk NCCs and of the whole cranial region up to the first somite for cranial NCCs with a scalpel under a stereomicroscope and subjected to mild enzymatic digestion by treatment with dispase II at 2.5 U/ml for 5–10 min at room temperature. The NT was dissected out manually using fine dissection pins under a stereomicroscope, freed from the surrounding tissues, and transferred for 30–60 min in DMEM medium (without glucose and pyruvate, Gibco, cat. no. 11966025) supplemented with 0.5% fetal bovine serum for recovery from enzyme treatment. The NT was explanted onto a culture dish or glass coverslip previously coated with fibronectin at 10 µg/ml in PBS (i.e. ∼5 µg/cm2) for a minimum of 1 h at 37°C. To ensure rapid initiation of NCC migration, the NT was positioned with its dorsal side oriented down toward the substratum. Explants were cultured at 37°C under normoxic conditions in a humidified 5% CO2 incubator in DMEM containing 1% serum, 100 U/ml penicillin, 100 µg/ml streptomycin, 2 mM glutamine, and supplemented with 5 mM glucose and 1 mM pyruvate. In some specified cases, glucose was used at 0.1–25 mM and pyruvate at 0.1–5 mM. The choices of the culture dish, culture medium and duration of the culture were determined according to the purpose of the experiment and the method of analysis used (see below). Throughout each experiment, the morphology of the NT explant, area and progression of the NCC outgrowth were evaluated, imaged and assessed regularly under an inverted phase contrast Nikon microscope equipped with 6.3×, 10× and 20× objectives.

Quantification of outgrowth area and density

To measure the NCC outgrowth area and NCC density, phase contrast images of explants encompassing half of the NCC outgrowth with the NT situated along the long axis of the image were taken with the 10× objective. The size of the image corresponded to a rectangle of ∼400 µm wide and 600 µm long. The area occupied by NCCs in the whole image and defined as surface occupancy in pixel2 was measured using ImageJ whereas NCC density defined as number of cells per field was calculated by counting the total number of cells in a square of 200 µm on each side.

Delamination assay

NT explants were cultured for 5 h, a time sufficient to allow NCC segregating from the dorsal NT to adhere to the substrate. The NT was then removed manually from the dish using fine dissection pins under a stereomicroscope or by gentle flushing of culture medium with a pipet tip, to uncover delaminated cells. Delaminating cells were discriminated from migratory cells by several criteria: their strong Snail-2 content, their central position in the outgrowth often separated from the migration zone by a cell-free gap, their reduced size and their compact shape. The numbers of delaminated and migrating cells were measured using images taken with a 10× objective and encompassing the mid-part of the explant along its long axis. Cells were counted in a rectangle 100-µm wide and 250-µm long covering the areas of delaminated cells or of migrating cells using ImageJ.

Cellular bioenergetic analyses

OCR and ECAR measurements

Bioenergetic profiles of NCC primary cultures were determined using a Seahorse Bioscience XF24 Analyzer. A single NT explant was deposited precisely at the center of each well of 24-well Seahorse plate previously coated with fibronectin in culture medium at 37°C in a humidified 5% CO2 incubator, and NCCs were allowed to undergo migration for 1.5 h or overnight depending to the purpose of the experiment. Before preparation of the plate for Seahorse analysis, the areas of the NT explants were measured for normalization. Cultures were then rinsed in Seahorse XF medium (Agilent, cat. no. 103575-100) without serum and incubated for 1 h at 37°C in a normal atmosphere. Then, the Seahorse assay was run according to the manufacturer's instructions. Oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) values, as readouts of basal mitochondrial respiration and glycolysis, respectively, were assessed regularly every cycle (mix, wait and measurement during 3 min), the basal measurement consisting of four cycles and drug injections, and three cycles during 30 min. The key parameters of mitochondrial respiration (ATP-linked respiration, maximal respiratory capacity and proton leak) were measured by means of a MitoStress test after sequential additions of oligomycin (1 µM), FCCP (0.75 µM) and Rot.-AA (1 µM) through three cycles of measurements in 30 min. When appropriate, NCC primary cultures were processed after Seahorse analysis for intracellular ATP level and lactate measurement.

Seahorse Agilent assay defined energetic maps for cell metabolic phenotypes comprised of four quadrants: quiescent, aerobic (OXPHOS), glycolytic and energetic. We designed an adaptation of Agilent energetic map to our system by assessing OCR and ECAR in NT explants under starvation medium, i.e. in DMEM with 2 mM glutamine without glucose and pyruvate, and in conditions with 5 mM glucose and 1 mM pyruvate. Basal respiration measured was in the range of 120–200 pmole/min under the starvation condition and 300–400 pmole/min in the presence of nutrients, respectively. This difference between the levels of OCR in these two conditions allowed us to fix the OCR threshold between the quiescent and OXPHOS quadrant at 200 pmole/min. The ECAR threshold between the quiescent and glycolytic quadrants was set at 30 mpH/min to achieve an OCR/ECAR ratio of 4, using as references previous studies which established that glycolysis was the major contributor to ECAR for baseline cellular OCR/ECAR ratio <4 (Konrad et al., 2017; Mercier-Letondal et al., 2021). Data shown in our energetic maps correspond to the mean±s.d. OCR and ECAR values of the fourth measurement of basal respiration and of the third measurement after inhibitor treatment.

ATP and lactate measurements

Intracellular ATP and extracellular lactate levels were measured for each NCC primary cultures after Seahorse analysis using an ATPlite Bioluminescence assay kit from PerkinElmer (cat. no. 6016943) and the Lactate Fluorometric Assay kit from BioVision (cat. no. K607), respectively, according to the manufacturers' instructions.

mRNA quantification by quantitative RT-PCR

For cell cultures, total RNA of cells from 5–8 NT explants were extracted using the Ambion PureLink RNA Mini Kit (Invitrogen, cat. no. 12183018A) following the manufacturer's guidelines. For injected embryos, the NT situated in the caudal part encompassing the last five somites up to the tail bud was dissected manually with fine dissection pins, and the total RNA was extracted using Trizol (Invitrogen) according to manufacturer instructions. After quantification of RNA concentration, 500 ng RNA was used for cDNA synthesis using SuperScript IV reverse transcriptase (Invitrogen, cat. no. 18090050). Quantitative real time PCR (qRT-PCR) were performed using the Power Syber Green Master Mix (Applied Biosystems, cat. no. 4368708) in a StepOne Plus RT-PCR apparatus (Applied Biosystems). Gene expression was assessed by the comparative CT (ΔΔCt) method with β-actin as the reference gene. The following primers were used: Snail2fwd, 5′-GATGCGCTCGCAGTGATAGT-3′; Snail2rev, 5′-AGCTTTCATACAGGTATGGGGATA-3′; Sox10fwd, 5′-CGGAGCACTCTTCAGGTCAG-3′; Sox10rev, 5′-CCCTTCTCGCTTGGAGTCAG-3′; Foxd3fwd, 5′-TCTGCGAGTTCATCAGCAAC-3′; Foxd3rev, 5′-TTCACGAAGCAGTCGTTGAG-3′; Glut1fwd, 5′-AAGATGACAGCTCGCCTGATG-3′; Glut1rev, 5′-AGTCTTCAATCACCTTCTGCGG-3′; PFKPfwd, 5′-TTGGAATTGTCAGCTGCCCG-3′; PFKPrev, 5′-TGCAGACAACTTTCATAGGCATCAG-3′; βActinfwd, 5′-CTGTGCCCATCTATGAAGGCTA-3′; and βActinrev, 5′-ATTTCTCTCTCGGCTGTGGTG-3′.

Immunofluorescence labeling of cultures

For immunolabeling, the following primary antibodies were used: rabbit monoclonal antibody (mAb) to Snail-2 (clone C19G7, cat. no. 9585, Cell Signaling, 1:300), mouse mAb to Sox-10 (clone A2, cat. no. sc-365692, Santa Cruz Biotechnology, 1:200), mouse mAb to β-catenin (clone 14, cat. no. 610154, BD-Transduction Laboratories, 1:200), mouse mAb to paxillin (clone 165, cat. no. 610620, BD-Transduction Laboratories, 1:100), mouse mAb to α-SMA conjugated to Cy3 (clone 1A4, cat. no. C6198, Sigma, 1:300), mouse mAb to βIII-tubulin conjugated to Alexa Fluor 488 (clone Tuj-1, R&D Systems, 1:500), mouse mAb to HNK-1, as described previously (Tucker et al., 1984; undiluted culture supernatant), mouse mAb to MITF (clone C5, cat. no. ab12039, Abcam, 1:200), and mouse mAb to Glut-1 (cat. no. ab14683, Abcam, 1:50). Filamentous actin was detected using Texas Red™-X Phalloidin (Molecular Probes, cat. no. T7471). NCC primary cultures were performed in four-well plates or in eight-well chambered glass coverslips (Nunc, cat. no. 154511) coated with fibronectin, and fixed in 4% paraformaldehyde (PFA) in PBS for 15 min at room temperature for detection of all antigens, except for Snail-2 (5 min in 4% PFA), β-catenin (45 min in 1.5% PFA), and Glut-1 (see below). After permeabilization with 0.5% Triton X-100 in PBS for 5 min, cultures were blocked in PBS with 3% BSA and subjected to immunofluorescence labeling using primary antibodies followed by incubation with appropriate secondary antibody conjugated to Alexa Fluor 488, Cy-3 or Cy-5 (Jackson Immunoresearch Laboratories), and processed for DAPI or Hoechst 33358 staining to visualize the nuclei of cells before mounting in ImmuMount medium (Shandon). For Glut-1 labeling, unfixed NCC cultures were incubated with the Glut-1 antibody diluted in DMEM for 15 min at 15°C to avoid antibody internalization and immediately fixed in 100% ethanol at −20°C before rinsing in PBS and secondary antibody treatment. Preparations were observed with a Zeiss AxioImager M2 epifluorescence microscope equipped with 10×–63× fluorescence objectives (Acroplan 10×/0.25, 20×/0.45, Plan-Neofluar 40×/0.75 and 63×/1.25 oil) or with a Zeiss LSM 900 confocal microscope equipped with 10×–40× fluorescence objectives (Plan-Apochrome 40×/1.30 oil). Data were collected using the Zen or Airyscan2 software and processed using ImageJ software. For each series of experiments, images were acquired using equal exposure times and settings.

2-NDBG uptake

For 2-NDBG uptake in life embryos, a portion of stage HH13-14 quail embryos taken from the last 10 formed somites and a similar portion of the unsegmented mesoderm were cut and incubated at 37°C for 2 h into 100 µl of 1.5 mM 2-NDBG (Life technologies) solution in glucose and pyruvate-free medium placed in a suspension culture dish. After culture, embryos were fixed in 4% PFA for 20 min, and then washed with PBS for 5 min at least three times, positioned on microscope slide into an Immumount (Shandon) droplet, covered with a coverslip. For 2-NDBG uptake in cell cultures, NT explants were incubated at 37°C for 2 h in 100 µl of 1.5 mM 2-NDBG solution in DMEM without glucose and pyruvate, fixed in 4% PFA for 20 min, and mounted as for immunolabelings in ImmuMount. Embryos and cultures were imaged using a Zeiss AxioImager M2 epifluorescence microscope.

In situ hybridization for mRNA detection on NCC primary cultures and whole-mount embryos

The following plasmids for mRNA probe synthesis were used: Snail-2 from Angela Nieto [Instituto de Neurociencias (CSIC-UMH), Sant Joan d'Alacant, Spain], Foxd-3 from Carol A. Erickson (Department of Molecular and Cellular Biology, UC Davis, Davis, CA, USA), Sox-10 from Paul Scotting (University of Nottingham, Nottingham, UK), Glut-1, PFK, PGK-1 and PKM from Olivier Pourquié (Department of Genetics, Harvard Medical School, Boston, MA, USA). Linearized plasmid DNA was used to synthesize digoxigenin-UTP (Roche)-labeled antisense probes with RNA polymerases from Promega and RNA probes were purified with Illustra ProbeQuant G-50 microcolumns (GE Healthcare). ISH was performed either on NCC primary cultures produced in fibronectin-coated plastic dishes or on whole-mount intact embryos collected at the appropriate developmental stages, using essentially the same procedure. Samples of embryos and cultures were fixed in 4% PFA in PBS for 2 h at room temperature or overnight at 4°C. They were hybridized overnight at 65°C with the digoxygenin-UTP-labeled RNA probes in 50% formamide, 10% dextran sulfate and Denhart's buffer (0.5 µg probe/ml hybridization buffer) and washed twice in 50% formamide, 1× SSC and 0.1% Tween-20 at 65°C, then four times at room temperature in 100 mM maleic acid, 150 mM NaCl pH 7.5 and 0.1% Tween-20 (MABT buffer). After a 1-h pre-incubation in MABT buffer containing 10% blocking reagent (Roche) and 10% heat-inactivated lamb serum, samples were incubated overnight at room temperature with the anti-digoxygenin antibody (Roche). After extensive rinsing with MABT buffer, they were preincubated in 100 mM NaCl, 50 mM MgCl2, 1% Tween-20 and 25 mM Tris-HCl, pH 9.5, and stained with NBT-BCIP (Roche) following manufacturer's guidelines. Preparations were observed and imaged with a Nikon stereomicroscope and data analyzed with ImageJ software.

Cryosectioning

To obtain embryo sections following in situ hybridization, embryos were washed in PBS for 1 h at room temperature and in 15% sucrose solution overnight at 4°C. Next, they were incubated in 7.5% porcine gelatin (dissolved in 15% sucrose solution) for 2–3 h at 40°C, embedded in gelatin-sucrose in cup, snap-frozen in chilled isopentane at −70°C and stored at −80°C. 20-µm sections were obtained using the Leica cryostat and collected on Superfrost/Plus slides (Thermo Fisher Scientific). For imaging, the slides were immersed in PBS at 40°C for 2 h for gelatin removal, washed in PBS and mounted in Aquatex mounting medium (Merck). Preparations were observed and imaged with a Hamamatsu NanoZoomer Digital Slide Scanner and treated with NDP.View2 software.

Quantification of substrate and cell–cell adhesions

Several fields of paxillin or β-catenin immunofluorescence staining were selected randomly at the periphery of the NCC outgrowth over at least three different NT explants per condition in at least two independent experiments. Quantifications of number, area and aspect ratio of substrate adhesions were performed as follows. For each experiment, images were segmented by manual thresholding using ImageJ. For each image, the area and aspect ratio of each single substrate adhesion were measured using ImageJ Analyze Particle plugin for particle size (micron): 0.2–infinity. The number of substrate adhesion per NCC corresponds to the mean of all the ratios, obtained for the images of each NT explant, between the total number of substrate adhesions quantified in the image and the number of NCCs present on the same image. Quantification of the number of substrate adhesion per NCC at 24 h was performed by manual counting on each image. Each dot represents the mean of substrate adhesion number counted per NCC per neural tube explant. Quantification of cell–cell contact numbers was performed as follows. For each experiment the number of contact each NCC engaged with its neighbors was counted manually on each image. The manual counting was done by the experimenter and checked independently by another researcher. The number of cell–cell adhesions engaged per NCC corresponds to the mean of the values obtained for the images of each NT explant. The data were expressed as percentage of NCCs with zero to three contacts with their neighbors.

Cell proliferation assay

Cell proliferation in culture was monitored using the Plus EdU Cell proliferation kit for imaging from Life Technologies (cat. no. C10638). Briefly, NCC primary cultures were generated on fibronectin-coated glass coverslips as for immunolabelings and were incubated with EdU at 20 µM in culture medium for 1 h at various times of NCC development. Immediately after EdU incorporation, cultures were fixed in 4% PFA in PBS for 15 min at room temperature, permeabilized in 0.5% Triton X-100 for 15 min and treated for EdU detection using Alexa Fluor™ 555 Azide in accordance with the manufacturer's guidelines. After DNA staining with Hoechst 33342, cultures were optionally processed for immunostaining and analyzed as described for the immunolabelings.

Cell locomotion assays

NCC primary cultures were performed in eight-well chambered glass coverslips coated with fibronectin. Up to four NT explants were distributed separately into each well, and were maintained at 37°C in a humidified 5%-CO2 incubator for about 2 h until the NT adhered firmly to the dish and NCCs initiated migration on the substratum; then the cultures were transferred into a heated chamber (Ibidi) with a humid atmosphere containing 5% CO2/95% air placed on the motorized stage of a Leica DMIRE2 microscope equipped with a CoolSNAP HQ camera (Roper Scientific). Time-lapse video-microscopy was performed with a 10× objective and phase contrast images were captured every 5 min during 16-24 h using the Micromanager software. The progression of the migratory front of the NCC population every hour was measured using a custom GUI (available upon request) written in MATLAB® (MathWorks®, Natick, MA, USA). Trajectories and positions of individual NCC in several explants recorded in parallel were tracked using Metamorph 7 software. The velocity of each NCC was calculated as the ratio between the total length of its trajectory and the duration of the acquisition time. The persistence of movement was calculated as the ratio between the linear distance from the initial to final positions of the NCCs and the total distance covered by the cell. Evolution of speed and persistence every hour were calculated as the length of NCC trajectory per 1 h and the persistence of movement as the ratio between the linear distance from the initial to final positions of the cells at each hour and the total distance covered by the cell during the corresponding period.

Preparation of polyacrylamide gels

Polyacrylamide (PAA) gels were prepared on 32-mm glass coverslips and 8-well chambered glass coverslips for AFM and TFM experiments, respectively. The surface on which the PAA gels were polymerized was activated by immersion into a solution of 3-methacryloxypropyltrimethoxysilane (0.3%, Bind-Silane, Sigma, cat. no. 440159), 10% acetic acid aqueous solution (3%) in absolute ethanol during 3 min to enhance gel adhesion, washed three times with ethanol, and air-dried for 45 min. Thin sheets of PAA gel of different elastic properties were prepared from concentrations of acrylamide/bisacrylamide at 5%/0.225% and 3%/0.08%, supplemented with 0.5% ammonium persulfate and 0.05% tetramethylethylenediamine (Sigma) in H2O. Practically, for AFM experiments, 81 µl of these different mixtures were placed onto the surface of a 32 mm-diameter coverslip and covered with 14 mm-diameter coverslip. For immunostaining and video-microscopy experiments, 2.5 µl of the mixtures were placed in the wells of 8-well chambered glass coverslips and covered with a 6 mm-diameter coverslip. Top coverslips were treated previously with Repel-Silane ES (Merck, cat. no. GE17-1332-01) during 5 min to prevent PAA gel adhesion. After 30 min of polymerization, the top coverslips were removed and the PAA gels were rinsed three times with PBS, treated twice for 5 min with sulfo-SANPAH, UV-photoactivated with Bio-Link Crosslinker BLX-E254 (Biotech), and rinsed three times with 50 mM HEPES at pH 8.5. For traction force microscopy experiments, the gel mixtures contain FluoSpheres from Sigma (TM carboxylate modified, 0.2 µm, red (580/605), cat. no. F8810), at 1/50 dilution. A solution of fibronectin was layered onto the PAA gels at 30 µg/cm² and incubated overnight at 4°C under agitation.

AFM assessments on polyacrylamide gels and NCC

The elastic modulus for both PAA gels and NCCs cultured for 4-6 h was assessed by AFM JPK NanoWizard Sence+ (Brucker, Billerica, Massachusetts, USA) coupled to a Zeiss Axio-Observer z1 inverted microscope, as described previously (Ben Bouali et al., 2020) using µMash CSC38/NO AL probes (MikroMasch®, Sofia, Bulgaria). PAA gels and NCCs were set in culture medium containing 25 mM HEPES and maintained at 37°C. For PAA gels, measurements were achieved considering a measuring point grid with 0.4- µm meshes over a 4.8×4.8 µm² surface set in gel center. For NCC, measurements were achieved considering a measuring point grid with 1- µm mesh over a 12×12 µm² surface set on cells using visual inspection with a phase contrast microscope. Setting Poisson's ratio ν=0.5 for both PAA gels and NCCs, AFM spectroscopy curves were analyzed according to (Bilodeau, 1992), for quadrilateral pyramid probe using a home-written program in Matlab®. The elastic modulus of the PAA gels prepared with proportion of acrylamide/bisacrylamide 5%/0.225% and 3%/0.08% were of 6320±25 Pa and 1250±32 Pa (±s.e.m.), respectively, and were referred in the text as stiff and soft gels. The impact of PAA gels stiffness on Young's moduli values of NCCs was measured by AFM and the obtained values were corrected as previously described (Rheinlaender et al., 2020).

TFM assessments

NT explants were cultured onto bead-embedded PAA gels in various nutrients for 4–6 h and transferred into a heated chamber with a humid atmosphere containing 5% CO2/95% air placed on the motorized stage of a Zeiss LSM 900 confocal microscope. A specific injector connected to eight-well Chambered glass coverslips was designed for trypsin delivery and installed on the chambered glass coverslips. Images of bead positions were acquired using Plan-Apochrome 40×/1.30 oil objective as follows. Images with force were acquired with 2-min interval leading to a 12-min observation time before trypsinization (stress images) and one image was acquired without force (reference image) taken after the NCCs were detached by trypsinization and no longer exerting forces on the substrate. Confocal stress and reference images were aligned using a MATLAB® script. The gel deformation was measured by the Particle Image Velocimetry (PIV) technique using the ImageJ plug-in PIV to get the bead displacement field. Finally, the ImageJ plug-in Fourier Transform Traction Cytometry (FTTC; PIV and FTTC ImageJ plugins are described at Martiel et al., 2015) was used to measure the traction force, exerted by NCC and responsible of PAA gel deformation and bead displacement (Martiel et al., 2015). The number of NCCs in the traction force field was measured by counting NCC nuclei stained with NucSpot® Live 650 and a confocal image was generated before starting TFM measurements. The stress taken into account is the mean stress calculated from six pairs of reference–stress images corresponding to 2, 4, 6, 8, 10 and 12 min TFM assessments. It was observed that between 2 and 12 min calculated stress was almost constant. The stress per cell is the stress divided by the number of nuclei. The force per cell (nN/cell) was calculated by multiplying the total stress/cell (Pa/cell) by the area of the interrogation window (µm²) used in the PIV.

Energy estimation

We estimated produced energy during NCC locomotion on soft and stiff gels by calculating the product of the average force F obtained by TFM and the average distance D covered per hour obtained by video-microscopy tracking during the time frame of TFM studies. The s.d. of the product F by D was calculated based on the following formula (Goodman, 1960): σ (F×D)=√ (〖E(F)〗^2 〖σ(D)〗^2+〖E(D)〗^2 〖σ(F)〗^2+〖σ(F)〗^2 〖σ(D)〗^2), with σ the s.d. and E(F) and E(D) mean value of F and D respectively, assuming F and D with normal distribution. To evaluate the statistical significance of two energy values, i.e., m1±σ1 and m2±σ2, we calculated Student's t and degrees of freedom, n1+n2−2, assuming n1=n2=5 for size of m1 and m2, to obtain P-values. For this, we considered , and σ common s.d, given by .

Statistical methods

Statistical analyses were performed using Prism 7 (GraphPad). For statistical analysis of data, we used a one-way ANOVA parametric test after the validation of normality and equality of variances using Shapiro–Wilk and Brown–Forsythe methods, respectively. Otherwise, non-parametric Krustal–Wallis test was used. For comparison between two conditions we use an unpaired t-test two-tailed test after the validation of normality and equality of using Shapiro–Wilk and equality of variances. Otherwise, the non-parametric Mann–Whitney test was used. Unless specified, at least three independent experiments were carried out for each procedure. Each NT explant was considered as an individual sample. The data obtained for each drug or nutrient condition were compared to that of the medium without drug or with both nutrients, respectively. The n of samples (explants or embryos) analyzed is indicated in each graph in brackets. For statistical analysis of qRT-PCR results of in vitro experiments, mRNA expression from a pool of 5–8 NT explants are presented as mean±s.e.m. of triplicates per experiment and gene. In in vivo experiments, mRNA expression was measured per embryo after analyzing of three replicates measurements acquired per embryo and genes. The data obtained from embryos injected with drugs were analyzed using two-way ANOVA relative to the data of embryos injected with vehicle. Data are expressed as mean±s.d. or s.e.m. values and the P-values are given as *P<0.05, **P<0.01; ***P<0.001; ****P<0.0001. Results are considered statistically significantly different when P<0.05.

We deeply thank Chantal Thibert and Sakina Torch for providing advice in cellular metabolism and critical reading of the manuscript. Special thanks to Olivier Pourquié and Masayuki Oginuma for providing cDNA probes for chicken glycolytic enzymes. Many thanks to Xavier Decrouy from the IMRB imaging platform (INSERM and Université Paris-Est Créteil) and Xavier Laffray from the histology and imaging platform of the Laboratoire Gly-CRRET (Université Paris-Est Créteil) for advice.

Author contributions

Conceptualization: J.-L.D., S.D.; Methodology: N.N.M., R. Fodil, S.F., A.D., M.D., J.-L.D., S.D.; Software: R. Fodil; Validation: N.N.M., R. Fodil, S.F., R.M., R. Foresti, J.-L.D., S.D.; Formal analysis: N.N.M., R. Fodil, J.-L.D., S.D.; Investigation: N.N.M., R. Fodil, J.-L.D., S.D.; Resources: N.N.M., R. Fodil, S.F., A.D., R.M., R. Foresti, J.-L.D., S.D.; Data curation: N.N.M., J.-L.D., S.D.; Writing - original draft: J.-L.D., S.D.; Visualization: N.N.M., R. Fodil, J.-L.D., S.D.; Supervision: J.-L.D., S.D.; Project administration: J.-L.D., S.D.; Funding acquisition: F.R., J.-L.D., S.D.

Funding

This work was supported by funding from Institut National de la Santé et de la Recherche Médicale, Université Paris-Est Créteil, AAP IMRB cross-teams project, and Fondation ARC pour la Recherche sur le Cancer (no. PJA 20181207844). N.N.-M. was funded by doctoral fellowship of Université Paris-Est Créteil and by the Labex REVIVE during the last year of her PhD. M.D. was funded by CDD contract of Agence Nationale de la Recherche (no. ANR-12-BSV2-0019).

Data availability

All relevant data can be found within the article and its supplementary information.

Barriga
,
E. H.
and
Mayor
,
R.
(
2019
).
Adjustable viscoelasticity allows for efficient collective cell migration
.
Semin. Cell Dev. Biol.
93
,
55
-
68
.
Barriga
,
E. H.
,
Franze
,
K.
,
Charras
,
G.
and
Mayor
,
R.
(
2018
).
Tissue stiffening coordinates morphogenesis by triggering collective cell migration in vivo
.
Nature
554
,
523
-
527
.
Ben Bouali
,
A.
,
Montembault
,
A.
,
David
,
L.
,
Von Boxberg
,
Y.
,
Viallon
,
M.
,
Hamdi
,
B.
,
Nothias
,
F.
,
Fodil
,
R.
and
Féréol
,
S.
(
2020
).
Nanoscale mechanical properties of chitosan hydrogels as revealed by AFM
.
Prog. Biomater.
9
,
187
-
201
.
Bhattacharya
,
D.
,
Azambuja
,
A. P.
and
Simoes-Costa
,
M.
(
2020
).
Metabolic reprogramming promotes neural crest migration via Yap/Tead signaling
.
Dev. Cell
53
,
199
-
211.e6
.
Bhattacharya
,
D.
,
Khan
,
B.
and
Simoes-Costa
,
M.
(
2021
).
Neural crest metabolism: At the crossroads of development and disease
.
Dev. Biol.
475
,
245
-
255
.
Bigarella
,
C. L.
,
Liang
,
R.
and
Ghaffari
,
S.
(
2014
).
Stem cells and the impact of ROS signaling
.
Development
141
,
4206
-
4218
.
Bilodeau
,
G. G.
(
1992
).
Regular pyramid punch problem
.
J. Appl. Mech.
59
,
519
-
523
.
Boon
,
R.
,
Silveira
,
G. G.
and
Mostoslavsky
,
R.
(
2020
).
Nuclear metabolism and the regulation of the epigenome
.
Nat. Metab.
2
,
1190
-
1203
.
Bronner
,
M.
(
2018
).
Riding the crest for 150 years!
.
Dev. Biol.
444
Suppl 1
,
S1
-
S2
.
Burstyn-Cohen
,
T.
and
Kalcheim
,
C.
(
2002
).
Association between the cell cycle and neural crest delamination through specific regulation of G1/S transition
.
Dev. Cell
3
,
383
-
395
.
Chevalier
,
N. R.
,
Gazguez
,
E.
,
Bidault
,
L.
,
Guilbert
,
T.
,
Vias
,
C.
,
Vian
,
E.
,
Watanabe
,
Y.
,
Muller
,
L.
,
Germain
,
S.
,
Bondurand
,
N.
et al. 
(
2016
).
How tissue mechanical properties affect enteric neural crest cell Mmigration
.
Sci. Rep.
6
,
20927
.
Duband
,
J. L.
(
2010
).
Diversity in the molecular and cellular strategies of epithelium-to-mesenchyme transitions: Insights from the neural crest
.
Cell Adh. Migr.
4
,
458
-
482
.
Duband
,
J. L.
,
Dady
,
A.
and
Fleury
,
V.
(
2015
).
Resolving time and space constraints during neural crest formation and delamination
.
Curr. Top. Dev. Biol.
111
,
27
-
67
.
Duband
,
J. L.
,
Nekooie-Marnany
,
N.
and
Dufour
,
S.
(
2020
).
Establishing primary cultures of trunk neural crest cells
.
Curr. Protoc. Cell Biol.
88
,
e109
.
Dupin
,
E.
,
Calloni
,
G. W.
,
Coelho-Aguiar
,
J. M.
and
Le Douarin
,
N. M.
(
2018
).
The issue of the multipotency of the neural crest cells
.
Dev. Biol.
444
(Suppl 1)
,
S47
-
S59
.
Erickson
,
A. G.
,
Kameneva
,
P.
and
Adameyko
,
I.
(
2023
).
The transcriptional portraits of the neural crest at the individual cell level
.
Semin. Cell Dev. Biol.
138
,
68
-
80
.
Goodman
,
L. A.
(
1960
).
On the exact variance of products
.
J. Am. Statist. Ass.
55
,
708
-
713
.
Gu
,
W.
,
Gaeta
,
X.
,
Sahakyan
,
A.
,
Chan
,
A. B.
,
Hong
,
C. S.
,
Kim
,
R.
,
Braas
,
D.
,
Plath
,
K.
,
Lowry
,
W. E.
and
Christofk
,
H. R.
(
2016
).
Glycolytic metabolism plays a functional role in regulating human pluripotent stem cell state
.
Cell Stem Cell
19
,
476
-
490
.
Hamburger
,
V.
and
Hamilton
,
H. L.
(
1951
).
A series of normal stages in the development of the chick embryo
.
J. Morphol.
88
,
49
-
92
.
Herrero-Mendez
,
A.
,
Almeida
,
A.
,
Fernández
,
E.
,
Maestre
,
C.
,
Moncada
,
S.
and
Bolaños
,
J. P.
(
2009
).
The bioenergetic and antioxidant status of neurons is controlled by continuous degradation of a key glycolytic enzyme by APC/C-Cdh1
.
Nat. Cell Biol.
11
,
747
-
752
.
Hu
,
N.
,
Strobl-Mazzulla
,
P. H.
and
Bronner
,
M. E.
(
2014
).
Epigenetic regulation in neural crest development
.
Dev. Biol.
396
,
159
-
168
.
Ito
,
K.
and
Suda
,
T.
(
2014
).
Metabolic requirements for the maintenance of self-renewing stem cells
.
Nat. Rev. Mol. Cell Biol.
15
,
243
-
256
.
Iyer
,
N. V.
,
Kotch
,
L. E.
,
Agani
,
F.
,
Leung
,
S. W.
,
Laughner
,
E.
,
Wenger
,
R. H.
,
Gassmann
,
M.
,
Gearhart
,
J. D.
,
Lawler
,
A. M.
,
Yu
,
A. Y.
et al. 
(
1998
).
Cellular and developmental control of O2 homeostasis by hypoxia-inducible factor 1 alpha
.
Genes Dev.
12
,
149
-
162
.
Johnson
,
M. T.
,
Mahmood
,
S.
and
Patel
,
M. S.
(
2003
).
Intermediary metabolism and energetics during murine early embryogenesis
.
J. Biol. Chem.
278
,
31457
-
31460
.
Kaelin
,
W. G.
, Jr
and
McKnight
,
S. L.
(
2013
).
Influence of metabolism on epigenetics and disease
.
Cell
153
,
56
-
69
.
Khacho
,
M.
,
Clark
,
A.
,
Svoboda
,
D. S.
,
Azzi
,
J.
,
MacLaurin
,
J. G.
,
Meghaizel
,
C.
,
Sesaki
,
H.
,
Lagace
,
D. C.
,
Germain
,
M.
,
Harper
,
M. E.
et al. 
(
2016
).
Mitochondrial dynamics impacts stem cell identity and fate decisions by regulating a nuclear transcriptional program
.
Cell Stem Cell
19
,
232
-
247
.
Konrad
,
C.
,
Kawamata
,
H.
,
Bredvik
,
K. G.
,
Arreguin
,
A. J.
,
Cajamarca
,
S. A.
,
Hupf
,
J. C.
,
Ravits
,
J. M.
,
Miller
,
T. M.
,
Maragakis
,
N. J.
,
Hales
,
C. M.
et al. 
(
2017
).
Fibroblast bioenergetics to classify amyotrophic lateral sclerosis patients
.
Mol. Neurodegener.
12
,
76
.
Lempradl
,
A.
,
Pospisilik
,
J. A.
and
Penninger
,
J. M.
(
2015
).
Exploring the emerging complexity in transcriptional regulation of energy homeostasis
.
Nat. Rev. Genet.
16
,
665
-
681
.
Li
,
Y.
,
Vieceli
,
F. M.
,
Gonzalez
,
W. G.
,
Li
,
A.
,
Tang
,
W.
,
Lois
,
C.
and
Bronner
,
M. E.
(
2019
).
In vivo quantitative imaging provides insights into trunk neural crest migration
.
Cell Rep
26
,
1489
-
1500.e3
.
Lukoseviciute
,
M.
,
Gavriouchkina
,
D.
,
Williams
,
R. M.
,
Hochgreb-Hagele
,
T.
,
Senanayake
,
U.
,
Chong-Morrison
,
V.
,
Thongjuea
,
S.
,
Repapi
,
E.
,
Mead
,
A.
and
Sauka-Spengler
,
T.
(
2018
).
From pioneer to repressor: bimodal Foxd3 activity dynamically remodels neural crest regulatory landscape in vivo
.
Dev. Cell
47
,
608
-
628.e6
.
Marchant
,
C. L.
,
Malmi-Kakkada
,
A. N.
,
Espina
,
J. A.
and
Barriga
,
E. H.
(
2022
).
Cell clusters softening triggers collective cell migration in vivo
.
Nat. Mater.
21
,
1314
-
1323
.
Martiel
,
J.-L.
,
Leal
,
A.
,
Kurzawa
,
L.
,
Balland
,
M.
,
Wang
,
I.
,
Vignaud
,
T.
,
Tseng
,
Q.
and
Théry
,
M.
(
2015
).
Measurement of cell traction forces with ImageJ
.
Methods Cell Biol.
125
,
269
-
287
.
Martik
,
M. L.
and
Bronner
,
M. E.
(
2017
).
Regulatory logic underlying diversification of the neural crest
.
Trends Genet.
33
,
715
-
727
.
Mercier-Letondal
,
P.
,
Marton
,
C.
,
Godet
,
Y.
and
Galaine
,
J.
(
2021
).
Validation of a method evaluating T cell metabolic potential in compliance with ICH Q2 (R1)
.
J. Transl. Med.
19
,
21
.
Miyazawa
,
H.
and
Aulehla
,
A.
(
2018
).
Revisiting the role of metabolism during development
.
Development
145
,
dev131110
.
Monier-Gavelle
,
F.
and
Duband
,
J.-L.
(
1995
).
Control of N-cadherin-mediated intercellular adhesion in migrating neural crest cells in vitro
.
J. Cell Sci.
108
,
3839
-
3853
.
Morriss
,
G. M.
and
New
,
D. A.
(
1979
).
Effect of oxygen concentration on morphogenesis of cranial neural folds and neural crest in cultured rat embryos
.
J. Embryol. Exp. Morphol.
54
,
17
-
35
.
Nieto
,
M. A.
(
2002
).
The Snail superfamily of zinc-finger transcription factors
.
Nat. Rev. Mol. Cell Biol.
3
,
155
-
166
.
Oginuma
,
M.
,
Moncuquet
,
P.
,
Xiong
,
F.
,
Karoly
,
E.
,
Chal
,
J.
,
Guevorkian
,
K.
and
Pourquie
,
O.
(
2017
).
A gradient of glycolytic activity coordinates FGF and Wnt signaling during elongation of the body axis in amniote embryos
.
Dev. Cell
40
,
342
-
353.e10
.
Patra
,
K. C.
and
Hay
,
N.
(
2014
).
The pentose phosphate pathway and cancer
.
Trends Biochem. Sci.
39
,
347
-
354
.
Pavlova
,
N. N.
and
Thompson
,
C. B.
(
2016
).
The emerging hallmarks of cancer metabolism
.
Cell Metab.
23
,
27
-
47
.
Perestrelo
,
T.
,
Correia
,
M.
,
Ramalho-Santos
,
J.
and
Wirtz
,
D.
(
2018
).
Metabolic and mechanical cues regulating pluripotent stem cell fate
.
Trends Cell Biol.
28
,
1014
-
1029
.
Rheinlaender
,
J.
,
Dimitracopoulos
,
A.
,
Wallmeyer
,
B.
,
Kronenberg
,
N. M.
,
Chalut
,
K. J.
,
Gather
,
M. C.
,
Betz
,
T.
,
Charras
,
G.
and
Franze
,
K.
(
2020
).
Cortical cell stiffness is independent of substrate mechanics
.
Nat. Mater.
19
,
1019
-
1025
.
Ridenour
,
D. A.
,
McLennan
,
R.
,
Teddy
,
J. M.
,
Semerad
,
C. L.
,
Haug
,
J. S.
and
Kulesa
,
P. M.
(
2014
).
The neural crest cell cycle is related to phases of migration in the head
.
Development
141
,
1095
-
1103
.
Rothstein
,
M.
and
Simoes-Costa
,
M.
(
2023
).
On the evolutionary origins and regionalization of the neural crest
.
Semin. Cell Dev. Biol.
138
,
28
-
35
.
Rovasio
,
R. A.
,
Delouvee
,
A.
,
Yamada
,
K. M.
,
Timpl
,
R.
and
Thiery
,
J. P.
(
1983
).
Neural crest cell migration: requirements for exogenous fibronectin and high cell density
.
J. Cell Biol.
96
,
462
-
473
.
Salazar-Roa
,
M.
and
Malumbres
,
M.
(
2017
).
Fueling the cell division cycle
.
Trends Cell Biol.
27
,
69
-
81
.
Santiago
,
A.
and
Erickson
,
C. A.
(
2002
).
Ephrin-B ligands play a dual role in the control of neural crest cell migration
.
Development
129
,
3621
-
3632
.
Schell
,
J. C.
,
Wisidagama
,
D. R.
,
Bensard
,
C.
,
Zhao
,
H.
,
Wei
,
P.
,
Tanner
,
J.
,
Flores
,
A.
,
Mohlman
,
J.
,
Sorensen
,
L. K.
,
Earl
,
C. S.
et al. 
(
2017
).
Control of intestinal stem cell function and proliferation by mitochondrial pyruvate metabolism
.
Nat. Cell Biol.
19
,
1027
-
1036
.
Schock
,
E. N.
and
LaBonne
,
C.
(
2020
).
Sorting Sox: Diverse roles for Sox transcription factors during neural crest and craniofacial development
.
Front. Physiol.
11
,
606889
.
Schwarz
,
Q.
,
Maden
,
C. H.
,
Vieira
,
J. M.
and
Ruhrberg
,
C.
(
2009
).
Neuropilin 1 signaling guides neural crest cells to coordinate pathway choice with cell specification
.
Proc. Natl. Acad. Sci. USA
106
,
6164
-
6169
.
Scully
,
D.
,
Keane
,
E.
,
Batt
,
E.
,
Karunakaran
,
P.
,
Higgins
,
D. F.
and
Itasaki
,
N.
(
2016
).
Hypoxia promotes production of neural crest cells in the embryonic head
.
Development
143
,
1742
-
1752
.
Semenza
,
G. L.
(
2012
).
Molecular mechanisms mediating metastasis of hypoxic breast cancer cells
.
Trends Mol. Med.
18
,
534
-
543
.
Shellard
,
A.
and
Mayor
,
R.
(
2021
).
Collective durotaxis along a self-generated stiffness gradient in vivo
.
Nature
600
,
690
-
694
.
Shyh-Chang
,
N.
,
Daley
,
G. Q.
and
Cantley
,
L. C.
(
2013
).
Stem cell metabolism in tissue development and aging
.
Development
140
,
2535
-
2547
.
Simões-Costa
,
M.
and
Bronner
,
M. E.
(
2015
).
Establishing neural crest identity: a gene regulatory recipe
.
Development
142
,
242
-
257
.
Simões-Costa
,
M. S.
,
McKeown
,
S. J.
,
Tan-Cabugao
,
J.
,
Sauka-Spengler
,
T.
and
Bronner
,
M. E.
(
2012
).
Dynamic and differential regulation of stem cell factor FoxD3 in the neural crest is Encrypted in the genome
.
PLoS Genet.
8
,
e1003142
.
Soldatov
,
R.
,
Kaucka
,
M.
,
Kastriti
,
M. E.
,
Petersen
,
J.
,
Chontorotzea
,
T.
,
Englmaier
,
L.
,
Akkuratova
,
N.
,
Yang
,
Y.
,
Haring
,
M.
,
Dyachuk
,
V.
et al. 
(
2019
).
Spatiotemporal structure of cell fate decisions in murine neural crest
.
Science
364
,
eaas9536
.
Théveneau
,
E.
,
Duband
,
J.-L.
and
Altabef
,
M.
(
2007
).
Ets-1 confers cranial features on neural crest delamination
.
PLoS ONE
2
,
e1142
.
Thiery
,
J. P.
,
Duband
,
J.-L.
and
Delouvée
,
A.
(
1982
).
Pathways and mechanism of avian trunk neural crest cell migration and localization
.
Dev. Biol
93
,
324
-
343
.
Trainor
,
P
. (
2013
).
Neural Crest Cells: Evolution, Development and Disease
:
Elsevier
.
Traxler
,
L.
,
Lagerwall
,
J.
,
Eichhorner
,
S.
,
Stefanoni
,
D.
,
D'Alessandro
,
A.
and
Mertens
,
J.
(
2021
).
Metabolism navigates neural cell fate in development, aging and neurodegeneration
.
Dis. Model. Mech.
14
,
dmm048993
.
Tsogtbaatar
,
E.
,
Landin
,
C.
,
Minter-Dykhouse
,
K.
and
Folmes
,
C. D. L.
(
2020
).
Energy Metabolism Regulates Stem Cell Pluripotency
.
Front. Cell Dev. Biol.
8
,
87
.
Tucker
,
G. C.
,
Aoyama
,
H.
,
Lipinski
,
M.
,
Tursz
,
T.
and
Thiery
,
J. P.
(
1984
).
Identical reactivity of monoclonal antibodies HNK-1 and NC-1: conservation in vertebrates on cells derived from the neural primordium and on some leukocytes
.
Cell Diff.
14
,
223
-
230
.
Vander Heiden
,
M. G.
,
Cantley
,
L. C.
and
Thompson
,
C. B.
(
2009
).
Understanding the Warburg effect: the metabolic requirements of cell proliferation
.
Science
324
,
1029
-
1033
.
Vincent
,
M.
and
Thiery
,
J. P.
(
1984
).
A cell surface marker for neural crest and placodal cells: Further evolution in peripheral and central nervous system
.
Dev. Biol.
103
,
468
-
481
.
Zanotelli
,
M. R.
,
Zhang
,
J.
and
Reinhart-King
,
C. A.
(
2021
).
Mechanoresponsive metabolism in cancer cell migration and metastasis
.
Cell Metab.
33
,
1307
-
1321
.
Zhang
,
J.
,
Zhao
,
J.
,
Dahan
,
P.
,
Lu
,
V.
,
Zhang
,
C.
,
Li
,
H.
and
Teitell
,
M. A.
(
2018
).
Metabolism in pluripotent stem cells and early mammalian development
.
Cell Metab.
27
,
332
-
338
.
Zhu
,
J.
and
Thompson
,
C. B.
(
2019
).
Metabolic regulation of cell growth and proliferation
.
Nat. Rev. Mol. Cell Biol.
20
,
436
-
450
.

Competing interests

The authors declare no competing or financial interests.

Supplementary information