Coordination of cell proliferation and anterior-posterior axis establishment in the mouse embryo

Abstract

During development, the growth of the embryo must be coupled to its patterning to ensure correct and timely morphogenesis. In the mouse embryo, migration of the anterior visceral endoderm (AVE) to the prospective anterior establishes the anterior-posterior (A-P) axis. By analysing the distribution of cells in S phase, M phase and G2 from the time just prior to the migration of the AVE until 18 hours after its movement, we show that there is no evidence for differential proliferation along the A-P axis of the mouse embryo. Rather, we have identified that as AVE movements are being initiated, the epiblast proliferates at a much higher rate than the visceral endoderm. We show that these high levels of proliferation in the epiblast are dependent on Nodal signalling and are required for A-P establishment, as blocking cell division in the epiblast inhibits AVE migration. Interestingly, inhibition of migration by blocking proliferation can be rescued by Dkk1. This suggests that the high levels of epiblast proliferation function to move the prospective AVE away from signals that are inhibitory to its migration. The finding that initiation of AVE movements requires a certain level of proliferation in the epiblast provides a mechanism whereby A-P axis development is coordinated with embryonic growth.

Fig. 1.

Random distribution of S-phase cells in the VE at the time of AVE migration. (A) Stills taken from a time-lapse movie of a Hex-GFP mouse embryo showing how GFP-positive cells remain at the distal tip of the embryo (red arrow) when leading cells of the anterior visceral endoderm (AVE) domain are migrating. Time is shown in hours:minutes. (B) BrdU staining of a pre-distal Hex-GFP embryo. Nuc, TOPO-3 staining of DNA. (C) BrdU-incorporation index in pre-distal group embryos. Results are mean ± s.e.m.; n=5. (D) BrdU staining of a distal Hex-GFP embryo. Arrows indicate the embryonic/extra-embryonic boundary. (E) BrdU-incorporation index in distal group embryos for the distal visceral endoderm (DVE), embryonic VE (embVE) and extra-embryonic VE (exeVE) domains (see Fig. 2B). Results are mean ± s.e.m.; n=6. (F) BrdU staining of a tilted group Hex-GFP embryo. (G) BrdU-incorporation index of the anterior and posterior VE of tilted group embryos. Results are mean ± s.e.m.; n=11. (H) Relative fold difference in BrdU incorporation between the anterior and posterior VE regions of tilted group embryos. The red line indicates no relative difference between the two areas. The relative fold difference values 1-15 are 0.76, 1.24, 0.97, 0.57, 1.50, 1.29, 1.04, 0.89, 0.89, 0.98, 1.46, 2.60, 0.62, 1.08 and 1.18. (I) BrdU staining of a 6.25-dpc Hex-GFP embryo. (J) BrdU-incorporation index in the anterior and posterior VE of 6.25-dpc embryos. Results are mean ± s.e.m.; n=11. There were no significant differences between the anterior and posterior VE by Student's t-test for paired samples (P>0.05).

Fig. 2.

Random distribution of mitotic cells at the onset of AVE migration. (A) Phospho-histone-H3 (PH3) staining of pre-distal, distal and tilted group Hex-GFP mouse embryos. The A-P axis is indicated. (B) The VE of six distal embryos was divided into three regions (DVE, embVE and exeVE) as illustrated. The mean mitotic index was calculated for each of these regions. Results are mean ± s.e.m.; n=6. (C) The VE of five tilted group embryos was divided into anterior (A), posterior (P), lateral left (L) and lateral right (R) quadrants. The mean mitotic index was calculated for each of these regions. Results are mean ± s.e.m.; n=5. (D) 5.5-dpc Hex-GFP embryos were cultured in the presence of nocodazole (200 nM) to cause mitotic arrest. Embryos were stained for PH3 and nuclei visualised with DAPI. (E) Treatment with nocodazole caused a 3-fold accumulation of mitotic cells (control, n=16; treated, n=14). (F) The VE of these embryos was divided into A, P, L and R quadrants and the mean mitotic index of each quadrant calculated. There was a statistically significant difference between the L and R quadrants (**P<0.01, χ² test).

Fig. 3.

Cell cycle rates in the AVE and posterior VE. (A) PH3 staining of 6.25-dpc Hex-GFP mouse embryos. (B) The embVE of six 6.25-dpc embryos was divided into A, P, L and R quadrants (see Fig. 2C). The mean mitotic index for each of these regions was calculated. Results are mean ± s.e.m.; n=6. No significant differences were observed between quadrants as assessed by χ² test. (C) 6.25-dpc embryos were treated with nocodazole for up to 5 hours and stained for PH3 to analyse the accumulation of mitotic nuclei. (D) Accumulated mitotic nuclei in the AVE and posterior VE were quantified in each embryo and the mean mitotic indexes were plotted as a function of length of nocodazole exposure. Results are mean ± s.e.m.; 0 hours, n=11; 0.5 hours, n=6; 1 hour, n=11; 1.5 hours, n=8; 2 hours, n=9; 3 hours, n=10; 4 hours, n=6; 5 hours, n=4. No significant differences in mitotic index between anterior and posterior VE were observed as assessed by χ² test (B) and a Student's t-test for paired samples (P>0.05; D).

Fig. 4.

Distribution of mitotic cells between 5.25 and 6.25 dpc. (A) Differences in the mean mitotic index between the epiblast and VE of PH3-stained pre-distal (n=5), distal (n=5) and tilted (n=8) groups and 6.25-dpc (n=11) mouse embryos. (B) The epiblast of five PH3-stained Hex-GFP tilted group embryos was divided into anterior (A) and posterior (P) halves, proximal (Prox) and distal (Dist) halves or left (L) and right (R) halves and the mean mitotic index calculated. (C) The epiblast of seven PH3-stained Hex-GFP embryos was divided as in B and the mean mitotic index calculated. All results are mean ± s.e.m. *P<0.05, **P<0.01, ***P<0.001 by Student's t-test for paired samples.

Fig. 5.

Nodal signalling is required for epiblast proliferation and AVE migration. (A,A′) BrdU staining of control cultured Hex-GFP 5.5-dpc mouse embryos (A) and Hex-GFP embryos cultured overnight in the presence of the Nodal receptor inhibitor SB-431542 (10 μM) (A′). (B) BrdU-incorporation index of the epiblast (Epi), AVE and posterior VE (PVE) of control (n=11) and SB-431542-treated (n=14) embryos. (C,C′) PH3 staining of control cultured Hex-GFP 5.5-dpc embryos (C) and Hex-GFP embryos cultured overnight in 10 μM SB-431542 (C′). (D) Mean mitotic index in the epiblast of control (n=16) and SB-431542-treated (n=17) embryos. (E,E′) PH3 staining of 5.5-dpc control (E) and Cripto mutant (E′) embryos. (F) Mean mitotic index for the epiblast and VE of control (n=21) and Cripto-null (n=8) embryos. All results are mean ± s.e.m. *P<0.05, **P<0.01 by Student's t-test.

Fig. 6.

Proliferation in the epiblast is required for the migration of the AVE. (A,A′) PH3 staining of control (A) and hydroxyurea-treated (A′) Hex-GFP mouse embryos. (B) Mean mitotic index in the epiblast and VE of control (n=6) and hydroxyurea-treated (n=7) embryos. (C,C′) PH3 staining of control (C) and genistein-treated (C′) embryos (n=8 for both). (D) Mean mitotic index in the epiblast and VE of control (n=8) and non-migrated genistein-treated (n=8) embryos. All results are mean ± s.e.m.; *P<0.05, **P<0.01, Student's t-test.

Fig. 7.

Dkk1 rescues the block in AVE migration caused by inhibition of proliferation. (A,B) 5.5-dpc mouse embryos cultured overnight in the presence of nocodazole (A; n=31) or nocodazole and Dkk1 (B; n=33). (C) The percentage of embryos with a migrated, partially migrated or non-migrated AVE for those treated with nocodazole versus nocodazole plus Dkk1.