Fig 1.
Growth of caudal neuroepithelium of the chicken embryo from 2 to 4 days of development.
A. Diagram of a chicken embryo at 48h of incubation (stage HH13-, 18 somites). B, transversal cryosections at the level of the forelimb at 48h, 72 and 96h of incubation. C-D, Confocal images of the intermediate region of the neuroepithelium at the level of the forelimb at 48, 72 and 96h of incubation, nuclei are stained with DAPI (grey) and actin with Phalloidin (green). Magenta line, apical domain; cyan line, basal domain. Note that the basal region progressively becomes devoid of nuclei. E, net dorsoventral length of the neural tube. F, net apicobasal length of the neural tube. G, number of pseudolayers of nuclei along the apicobasal axis. H, distribution of DAPI intensity along the apicobasal axis, normalized to the peak intensity in each dataset and to the apicobasal size so that the various stages can be compared. I, ratio of apicobasal and dorsoventral length of nuclei. J, straightness of apical domain (net distance between dorsal-most and ventral-most points of the apical surface divided by the actual length of the apical surface between these two points). K, positions of mitotic cells (either phospho-histone H3-positive cells or cells with condensed chromosomes from DAPI staining), at scale with the actual apicobasal size of the neural tube. L, positions of mitoses, raw data. Descriptive data collected from 20 embryos. Dots represent mean values. Error bars show the standard deviation. Box and whiskers plot: the box extends from the 25th to the 75th percentile; the whiskers show the extent of the whole dataset. The median is plotted as a line inside the box. AB, apicobasal; DV, dorsoventral; HH, Hamburger-Hamilton stages of chicken development; ss, somites.
Fig 2.
Agent-based model of the pseudostratified epithelium.
A. 3D confocal image of a single neuroepithelial cells expressing membrane-GFP, in a 2-day old chick neuroepithelium (see S1 Movie). B. Cells in the model are abstracted to a nucleus attached to a set of springs. C, implementation of cell cycle and INM during the simulation. Cells in the model constantly proliferate by going through a simplified cell cycle corresponding to three phases: a G1/S/passive G2 phase during which springs connected to the nucleus adjust to local constraints, a PRAM/active G2 phase during which apical-nucleus springs shrink while nucleus-basal springs elongate to recapitulate INM movements and the M phase during which springs behave as in G2. In addition, the hard core of nuclei enlarges in M phase to account for cell swelling and stiffening. Finally, at the end of the M phase each cell gives two daughter cells. One is systematically kept within the 2D-plane, the other daughter cell can be excluded. This parameter allows to control the rate of growth of the tissue independently of the pace of the cell cycle. See S1 Information for a detailed description of the model.
Fig 3.
INM opposes apicobasal elongation, generates apical nuclear crowding and enlarges the apical domain.
Simulations with passive apical-apical springs with INM (black and brown curves) or without INM (red curves) with constant cell number (see S2 Movie). A, Apicobasal length of the PSE (AB) and mean nuclear position along the AB axis, expressed in micrometers. Note that INM reduce apicobasal length (black curve with open circles (INM) is below the red curve (no INM)) and generate apical nuclear crowding (black curve with closed circles (INM) is above the red curve (no INM)). B, Number of layers of nuclei along the AB axis. C, straightness of apical domain (net distance between the first and last apical point divided by the actual distance between these two points). D, Position of mitoses along AB axis, 1 being apical. E-F, net width of apical (magenta), nuclear (black) and basal (cyan) domains of the PSE over time. For each domain, the distance the between the first and last point along the lateral axis is computed and its evolution plotted over time. Note that INM promote enlargement of the apical domain (panel 3G, arrow). Each simulation was performed over 480 iterations (48h of biological time) for 10 repetitions. Each curve represents the mean value of each dataset for the parameter plotted. Box and whiskers plot: the box extends from the 25th to the 75th percentile; the whiskers show the extent of the whole dataset. The median is plotted as a line inside the box.
Fig 4.
INM opposes apical contractility.
Same simulations as in Fig 3 but with contractile apical-apical springs with INM (black and brown curves) or without INM (red curve), with constant cell number (see S2 Movie). A, Apicobasal length of the PSE (AB) and mean nuclear position along the AB axis, expressed in micrometers. B, Number of layers of nuclei along the AB axis. C, straightness of apical domain (net distance between the first and last apical point divided by the actual distance between these two points). D, Position of mitoses along AB axis, 1 being apical. E-G, net width of apical (magenta), nuclear (black) and basal (cyan) domains of the PSE over time. For each domain, the distance between the first and last point along the lateral axis is computed and its evolution plotted over time. Note that apical contractility reduces the width of the apical domain (panel 4E, arrow), whereas introducing INM opposes apical contractility (panel 4G, arrow; red curve in G is higher than in E). Each simulation was performed over 480 iterations (48h of biological time) for 10 repetitions. Each curve represents the mean value of each dataset for the parameter plotted. Box and whiskers plot: the box extends from the 25th to the 75th percentile; the whiskers show the extent of the whole dataset. The median is plotted as a line inside the box.
Fig 5.
Increase in cell number drives pseudostratification and apical contractility feeds back into basal rearrangements.
Simulations with passive or contractile apical springs, normal INM and various rates of exclusion of daughter cells (see S3 Movie). A, apicobasal length of the PSE (AB) and mean nuclear position along the AB axis (N) over time expressed in micrometers. B, Number of pseudolayers of nuclei along the AB axis. C, straightness of apical domain (net distance between the first and last apical point divided by the actual distance between these two points). D-F, net width of apical (magenta), nuclear (black) and basal (cyan) domains of the PSE with passive apical-apical springs over time with 50 (D), 40 (E) and 30% (F) of daughter cells being excluded from the 2D plane. G-I, net width of apical (magenta), nuclear (black) and basal (cyan) domains of the PSE with contractile apical-apical springs over time with 50 (G), 40 (H) and 30% (I) of daughter cells being excluded from the 2D plane. For each domain, the distance between the first and last point along the lateral axis is computed and its evolution plotted over time. Note that apical contractility leads to basal rearrangements (compare cyan curves in H-I grow faster than in E-F). Each simulation was performed over 480 iterations (48h of biological time) for 10 repetitions. Each curve represents the mean value of each dataset for the parameter plotted.
Fig 6.
Mean volume and mean apical surface of chick neuroepithelial cells decreases between 2 and 3 days of development.
A, Diagram depicting the regions used for the preparation of cell suspensions and open book histology from chicken embryos at 2, 3 and 4 days of development (see Methods for details of the experimental procedures). Region monitored (from somite 15 to 20, forelimb region) is indicated by a dotted line. B, representative images of neuroepithelial cells in suspension after neural tube dissection and enzymatic dissociation. C, mean volume of neuroepithelial cells over time (n48h = 82; n72h = 568; n96h = 1168). Cells get significantly smaller from 2 to 3 days of development and remain stable. Box and whiskers plot: the box extends from the 25th to the 75th percentile; the whiskers show the extent of the whole dataset. The median is plotted as a line inside the box. One-way ANOVA (Kruskal-Wallis) followed by Dunn’s multiple comparisons. ****, p<0.0001; ns, p>0.9999. D, en face view of the apical domain of the intermediate region of the neuroepithelium (actin is stained by Phalloidin). E, mean area of the individual apical surfaces over time (n48h = 67; n72h = 66; n96h = 81). Apical surfaces shrink from 2 to 3 days of development and remain stable. Dots represent mean of the dataset, error bars represent S.D. One-way ANOVA followed by multiple comparisons. ****, p<0.0001; ns, p = 0.1586.
Fig 7.
Apicobasal elongation requires a specific elongation force.
Simulations with contractile apical springs, normal INM and high noise. A, apicobasal length of the PSE (AB) and mean nuclear position along the AB axis (N) over time expressed in micrometers (see S5 Movie). B, Number of pseudolayers of nuclei along the AB axis. C, straightness of apical domain (net distance between the first and last apical point divided by the actual distance between these two points). D-G, net width of apical (magenta), nuclear (black) and basal (cyan) domains of the PSE with dynamic apical-apical springs over time with 50 (D), 40 (E) and 30% (F-G) of daughter cells being excluded from the 2D plane. For each domain, the distance between the first and last point along the lateral axis is computed and its evolution plotted over time. D-F, slow update of basal point positions (as in all previous Figs). G, fast update of basal point positions (twice faster). Each simulation was performed over 480 iterations (48h of biological time) for 10 repetitions. Each curve represents the mean value of each dataset for the parameter plotted.
Fig 8.
Interplay between INM, proliferation and cell adhesion in the context of a specific apicobasal elongation program are needed to recapitulate normal PSE dynamics.
Green arrows indicate positive action, red arrows indicate negative/inhibitory action. Dotted line indicate weaker effect than plain lines. PRAM, pre-mitotic rapid apical migration; PRAR, post-mitotic rapid apical removal; INM, interkinetic nuclear movement.