A Gradient of Glycolytic Activity Coordinates FGF and Wnt Signaling during Elongation of the Body Axis in Amniote Embryos

Abstract

Mammalian embryos transiently exhibit aerobic glycolysis (Warburg effect), a metabolic adaptation also observed in cancer cells. The role of this particular type of metabolism during vertebrate organogenesis is currently unknown. Here, we provide evidence for spatiotemporal regulation of glycolysis in the posterior region of mouse and chicken embryos. We show that a posterior glycolytic gradient is established in response to graded transcription of glycolytic enzymes downstream of fibroblast growth factor (FGF) signaling. We demonstrate that glycolysis controls posterior elongation of the embryonic axis by regulating cell motility in the presomitic mesoderm and by controlling specification of the paraxial mesoderm fate in the tail bud. Our results suggest that glycolysis in the tail bud coordinates Wnt and FGF signaling to promote elongation of the embryonic axis.

Keywords: FGF; axial elongation; embryo; glycolysis; lactate; metabolism; paraxial mesoderm; somitogenesis.

Figure 1. A posterior glycolytic gradient in the mouse tail bud

(A) Lateral view of the posterior region of E 9.5 day mouse embryo showing the 3 domains used for metabolomics analysis. TB: Tail bud, S: somite, A-PSM and P-PSM: Anterior and Posterior Presomitic Mesoderm respectively. Each domain contains approximately 5000 cells. Scale bar :100 μm. (B) Clustering analysis of the 39 metabolites showing the most significant differential expression along the posterior region of the embryo. Color bar shows fold change from mean. Red box highlights metabolites downregulated and blue box metabolites upregulated during differentiation. (C) Schematic representation of the mouse glycolytic cascade. Red labeling indicates metabolites detected in the metabolomics analysis. Yellow labeling indicates the glycolytic enzymes transcripts downregulated in the mouse embryo tail bud and PSMs. (D) Expression profiles of the metabolites associated to glycolysis detected in the metabolomics analysis (* Significant at p<0.05 with t-test). The color bar shows fold change from mean of all triplicate samples. (E–G) Enzymatic detection of relative lactate levels, Cytochrome C oxidase activity, and relative ATP concentrations during tail bud differentiation. Graphs show triplicate experiments. Values were normalized by P-PSM, Error bars are ±SD. Statistical significance was assessed with one way ANOVA followed by Tukey’s test, * p<0.05, ** p<0.01. (H) Expression profiles of transcripts coding for glycolytic enzymes downregulated in the mouse PSM during differentiation. Each value is normalized to the mean of MAS values of all triplicate mouse microarray series. The color bar shows fold change from mean. (I) Lateral view of the tail bud region of 9.5-day mouse embryo stained with an antibody against Glut3 (n=5). Maximum projection of confocal sections. Fluorescence intensity is shown by pseudo-color image (16 color) using image J. Higher levels of Glut3 proteins are indicated in red-yellow. Scale bar : 100 μm. See also Figure S1, Table S1, Table S2

Figure 2. Conservation of the posterior glycolytic gradient in the chicken embryo

(A). Schematic representation of the posterior portion of a two-day old chicken embryo. The fragments micro-dissected to generate the microarray series are indicated. Dorsal view, anterior to the right. TB: Tail bud, PSM: Presomitic Mesoderm. (B) Expression profiles of transcripts coding for glycolytic enzymes down-regulated in chicken PSM during differentiation as detected in the microarray series. Each value is normalized to the mean of MAS values of all duplicate chicken microarray series. Color bar shows fold change from mean. (C) Schematic representation of the chicken glycolytic cascade. Yellow labeling indicates the glycolytic enzymes whose transcripts are downregulated in the chicken embryo tail bud and PSM. (D) Left, posterior region of 2-day chicken embryos hybridized with probes for the glucose transporter GLUT1, and for the glycolytic enzymes PFKP, PGK1, PKM and LDHB. Right, fluorescent glucose (2-NDBG) uptake in the posterior region of a 2-day chicken embryo (n=11). Maximum projection of confocal sections. Fluorescence intensity is shown by pseudocolor image (16 color) using image J. Higher levels of fluorescent 2-NDBG are indicated in red-yellow. Ventral view, anterior to the top. Scale bar :100 μm. (E–G) Enzymatic detection of relative lactate level, Cytochrome C oxidase activity, and relative ATP concentration during tail bud differentiation. Graphs represent triplicate experiments. Values were normalized by the P-PSM, Error bars ±SD. Statistical significance was assessed with one way ANOVA followed by Tukey’s test, * p<0.05, ** p<0.01, *** p<0.001). See also Figure S2, Table S3

Figure 3. FGF signaling regulates glycolysis in the posterior part of the embryo

(A)Fluorescent glucose (2NDBG) uptake in a 2-day old chicken embryo. Maximum projection of confocal sections. Fluorescence intensity is shown by pseudo-color image (16 color) using image J. Higher levels of uptake indicated in red-yellow. Dorsal view, anterior to the top. Scale bar :100 μm. (B–C) Whole mount in situ hybridization of 2-day chicken embryos with FGF8 (B) and SPRY2(C) probes. Dorsal view, anterior to the top. Scale bar :100 μm. (D–F) Enzymatic detection of relative lactate level (D), of Cytochrome C oxidase activity (E), and of relative ATP concentration (F) in the posterior part of 2-day chicken embryos treated with inhibitors of FGF (SU5402), MAPK (PD0325901), Notch (DAPT) and Retinoic Acid (BMS204493) (n=6 for each condition). Graphs represent triplicate experiments. Values are normalized by untreated control embryos. Error bars ±SD. Statistical significance was assessed with one way ANOVA followed by Tukey’s test, ***p<0.001, ns p>0.05) (G) qPCR analysis of glycolytic enzymes expression levels in control (blue) and in embryos treated with the MAPK-inhibitor PD0325901 (green). Graphs represent triplicate experiments. Values are normalized by untreated control embryos, Error bars are ±SD. Statistical significance was assessed with unpaired two-tailed student t-test., **p<0.01, ***p<0.001 (H–I) Whole mount in situ hybridization with probes for the rate limiting glycolytic enzymes LDHB (H: n=8, I: n=7) and PKM (N: n=8, O: n=8), in control (H, J) and PD0325901-treated (I, K) 2-day old chicken embryos. Ventral view, anterior to the top.

Figure 4. Glycolysis controls posterior elongation of the embryonic axis

(A–C) Enzymatic detection of relative lactate level (A), of cytochrome C oxidase activity (B) and of relative ATP concentration (C) in the posterior region of control 2-day old chicken embryos and in embryos treated with 2DG or NaN₃ (n=6 embryos for each condition). Graphs represent triplicate experiments. Values are normalized by untreated control embryos. Error bars are ±SD. Statistical significance was assessed with one way ANOVA followed by Tukey’s test, **p<0.01, ***p<0.001, ns p>0.5) (D) Increase in axis length (elongation) measured over time using time lapse microscopy (mean ±SD). Blue, control embryos (n=8); green, 2DG-treated embryos (n=7); yellow, NaN3-treated embryos (n= 5). (E–G) Elongation time course in a control (E), in a 2DG-treated (F) and in a NaN3-treated (G) 2-day chicken embryo. Bright field micrographs of the posterior region of a chicken embryos taken at 1.5 hour intervals. Somites formed at the last time point are indicated by asterisks on the right. Ventral views, anterior to the top. See also Movie S1 (H–I) Elongation time course in chemically defined DMEM-based culture with 0.15% glucose (H), and without glucose (I). Bright field micrographs of the posterior region of 2-day chicken embryos taken at 1.5 hour intervals. Somites formed at the last time point are indicated by asterisks on the right. Ventral views, anterior to the top. (J) Increase in axis length (elongation) measured over time using time lapse microscopy (mean ±SD). Magenta, embryos in EC culture (n=5); green, embryos in defined 0.15% glucose DMEM-based cultures (n=4); blue, embryos in defined DMEM-based culture without glucose (n= 5). (K,L) Enzymatic detection of relative lactate level (K) and of relative ATP concentration (L) in the posterior region of control 2-day old chicken embryos and in embryos cultured in defined DMEM-based medium with or without 0.15% glucose and in 2DG treated embryos (n=4 embryos for each condition). Graphs represent quadruple experiments. Values are normalized to 0.15% glucose embryos. Error bars are ±SD. Statistical significance was assessed with one way ANOVA followed by Tukey’s test, **p<0.01, ***p<0.001, ns p>0.5) See also Figure S3

Figure 5. Glycolysis inhibition decreases cell motility and increases pH in the posterior PSM

(A–F) Effect of 2DG treatment on cell motility (diffusion) and PSM elongation in 2-day chicken embryos. (A) Electroporated PSM cells expressing H2B-Venus are shown in yellow. (B) and (C) PSM cell trajectories for control and 2DG treated embryos respectively. Only tracks of cells located in the posterior PSM inside the box are used in the analysis. Scale bars: 100 μm. (D) Elongation curves showing the posterior displacement of the tail bud (orange box) as a function of time in the wild-type and 2DG-treated embryos. (E) Elongation rates of the embryos shown in (D), error bars are ±SD of the linear adjustments to data in (D). (F) Cell diffusion for control and 2DG- treated embryos shown from 2h–6h and from 6h–10h (mean ±SD, *p=0.0004, **p< .0001, t-test). See also Movie S2 (G–K) Cell motility (diffusion) and PSM elongation in chemically controlled conditions. (G) and (H) PSM cell trajectories for 2-day chicken embryos cultured in DMEM and DMEM without glucose respectively. Only tracks of cells located in the posterior PSM inside the box are used in the analysis. Scale bars: 100 μm. (I) Elongation curves showing the posterior displacement of the tail bud as a function of time in the DMEM and DMEM without glucose conditions. (J) Elongation rates of the embryos shown in (I), error bars are ±SD of the linear adjustments to data in (I). (K) Cell diffusion for DMEM and DMEM without glucose conditions shown from 0h–3h and from 3h–6h (mean ±SD, *p < .00001, **p=.0069, t-test). See also Movie S3 (L) Extracellular pH in the posterior part of a 2-day old chicken embryo labeled with pHrodo Red showing the gradient of extracellular pH (pHe). Red fluorescence labels regions of lower pH. Electroporated PSM cells expressing Gap-GFP are shown in green (Okada et al., 1999). Maximum projection of confocal sections of a 2-day chicken embryo. Ventral views, anterior to the top. pHrodo intensity was measured specifically in the PSM in the region highlighted by a white dashed box. Scale bar: 100μm. (M) Quantification of intensity of pHrodo Red. Phrodo Intensity measured in 3 different embryos along the posterior-anterior axis using stripes in PSM as described in (L). Measurements were binned every 200μm and the average intensity (±SD) is shown. Absolute Phrodo intensity values may vary as a result of different embryo/imaging conditions. Embryo 1 is (L). (N) Effect of culturing embryos on 2DG-containing and on alkaline plates on the pHe gradient. Each dot represents average fluorescence intensity measured in a ~0.26mm² area of each embryo along the posterior to anterior axis. Lines represent average intensity (±SD) (Control: n=7, alkaline: n=8, 2DG; n=7). (*p=0.001, **p=0.001, ***p=0.003, ****p=0.006, (n.s.) p=0.15, t-tests). ******(red, Control) p=0.002, n.s.(blue, 2DG) p=0.14, n.s.(black, NaOH) p=0.35. Paired t-tests to compare the A–P ends of the pHe gradient. (O–Q) Effect of culture on alkaline plates on cell motility and PSM elongation in 2-day chicken embryos. (O) Elongation curves for control and embryos cultured on alkaline plates. (P) Elongation rate corresponding to the curves in (O). (Q) Cell diffusion for control and embryos cultured on alkaline plates (mean ±SD, p<.01 between control and NaOH cases and n.s. among NaOH using one way ANOVA followed by Tukey’s test). See also Movie S4, Figure S4

Figure 6. Inhibition of glycolysis phenocopies Wnt signaling inhibition in the tail bud

(A–C) Whole mount in situ hybridizations showing the posterior region of 2-day old control (A, n=24), 2DG-treated (B, n=16) and NaN3-treated (C, n=13) chicken embryos hybridized with LFNG probe. Scale bar :100 μm. Asterisk indicate newly formed somites. (D) qPCR analysis of the posterior region of 2-day old chicken embryos treated or not with 2DG and incubated for 10 h. Statistical significance was assessed with unpaired two-tailed student t-test. * p<0.05, **p<0.01, ***p<0.001. (E–N) Whole mount in situ hybridizations showing the posterior region of 2-day old control (E–I) or 2DG-treated (J–N) chicken embryos hybridized with the following probes: T (BRACHYURY)(E: n=4, J: n=6); CMESPO (F: n=5, K: n=5); SOX2 (G: n=7, L: n=4); CMESO1 (H: n=5, M: n=4); Axin2 (I: n=5, N: n=4). Asterisks indicate the last formed somite, and arrowheads indicate tail bud region. Ventral view, anterior to the top. Scale bar: 100 μm. (O–T) SOX2 (O, R) and T /BRACHYURY (P, S) protein expression in control (O–Q) and 2DG-treated (R–T) 2-day chicken embryos (n=4 for each condition). (Q, T) Higher magnification of the tail bud stained with SOX2 and T/BRACHYURY. Stippled line delineates the double positive cells in the chordo-neural hinge region. Left panel shows schematic representation of the posterior part of a 2-day chicken embryo indicating the position of the sections shown in O–T (axial red dotted line). (U–Z) Expression of CMESPO (U, X) and CTNNB1 (Beta-Catenin) (V–W, Y–Z) proteins in sagittal sections of a 2-day old control (U–Z, n=5), and a 2DG-treated (X–Z, n=4) chicken embryo. (W–Z) Higher magnification of the regions boxed in red in (V, Y) showing the nuclear localization of B-CATENIN. Top panel shows anterior PSM. Nuclei labeled with DAPI are shown in blue. Left panel shows a schematic representation of the posterior part of a 2-day chicken embryo indicating the position of the sections shown in U–Z (Paraxial blue dotted line). Scale bar : 100 μm (A–C, E–T, U–V, X–Y), 10 μm (W, Z) Asterisk indicates newly formed somites, and arrowheads indicate tail bud region. See also Figure S5, Movie S5

Figure 7. Phenotypic correlation between FGF and glycolysis inhibition

(A–B) Micrographs taken at 1.5 hour intervals of the posterior region of 2-day old chicken embryo control (A) and treated with PD0325901 (B). (C) Graph showing the increase in axis length (elongation) measured using time lapse microscopy. in 2-day chicken control embryos (blue) and PD0325901-treated embryos (green). Data represent the average of 5 embryos per conditions, error bars are mean ±SD. (D–I) Expression of CTNNB1 (Beta-Catenin) (green, D, F, H) and phosphorylated MAPK (magenta, E, G, I) in longitudinal sections of control (D, E: n=5), 2DG-treated (F, G: n=4) and PD0325901-treated (H, I: n=4) 2-day chicken embryos. Sections shown in (D, F, H) are counterstained with DAPI (in blue) to visualize the nuclei. (J–Q) Expression of AXIN2 (J, M, O), SPRY2 (K, N, P), CMESPO (L, Q) detected by whole mount in situ hybridization in the posterior part of 2-day old chicken control embryos (J: n=5, K: n=5, L: n=4), 2DG-treated embryos (M: n=4, N: n=3) and PD0325901-treated embryos (O: n=4, P: n=3, Q: n=4). Scale bar : 100 μm (D–Q). Ventral view, anterior to the top. Asterisk indicate newly formed somites, and arrowheads indicate tail bud region.