Elongation of the nascent avian foregut requires coordination of intrinsic and extrinsic cell behaviors

Abstract

The foregut tube gives rise to the lungs and upper gastrointestinal tract, enabling vital functions of respiration and digestion. How the foregut tube forms during embryonic development has historically received considerable attention, but over the past few decades this question has primarily been addressed indirectly through studies on morphogenesis of the primitive heart tube, a closely related process. As a result, many aspects of foregut development remain unresolved. Here, we exploit the accessibility of the chick embryo to study the initial formation of the foregut tube, combining embryology with fate mapping, live imaging, and biomechanical analyses. The present study reveals that the foregut forms and elongates over a narrower time window than previously thought, and displays marked dorso-ventral and left-right asymmetries early in its development. Through tissue-specific ablation of endoderm along the anterior intestinal portal, we confirm its central role in driving foregut morphogenesis, despite not directly contributing cells to the elongating tube. We further confirm the important role of this cell population in formation of the heart tube, with evidence that this role extends to later stages of cardiac looping as well. Together, these data reveal the need for an intricate balance between intrinsic cell behaviors and extrinsic cues for normal foregut elongation, and set the stage for future studies aimed at understanding the underlying molecular cues that coordinate this balance.

Keywords: Biomechanics; Chick; Endoderm; Heart tube; Live imaging; Morphogenesis.

Figure 1:. AIP displacement is poorly correlated with foregut elongation.

(A, left) Representative depth-coded maximum intensity projections of DAPI-stained chick embryos from the initiation of foregut formation with the emergence of the head fold at HH stage 7 to onset of c-looping of the heart tube at HH stage 11. *indicates the position of the medial Anterior Intestinal Portal (AIP); scale = 200 μm. (A, right) Schematic sagittal section of the anterior chick embryo during foregut elongation; D = dorsal, V = ventral, A = anterior, P = posterior. The forming foregut lumen is highlighted in blue (light and dark blue indicating the foregut roof and floor, respectively), the extraembryonic endoderm is highlighted in yellow, the AIP endoderm in red, and the presumptive foregut roof endoderm in green. (B) Velocity of the AIP relative to somite 2 between HH stages 8–9, 9–10, and 10–11 (n = 10 per group); *p<0.05, ***p<0.001 as determined by one-way ANOVA with Tukey’s multiple comparisons test. (C, left) HH stage 8 embryos following bisection (0 hours) to mechanically separate AIP from regressing node region and following 10 hours incubation; scale = 200 μm. (C, right) Foregut length compared between bisected (n = 7) and intact (n = 8) control embryos; n.s. = not significant (p = 0.211 by Student’s t-test). (D, E) Foregut tube length normalized to antero-posterior embryo length (D) and foregut aspect ratio (E) across developmental stages HH stage 8- to 12 (n = 4–7 per stage). (F) Comparison of foregut elongation versus AIP displacement of individual embryos (n = 10 per stage) between HH stages 8–9 (left), 9–10 (middle), and 10–11 (right). Linear regression is indicated by a solid line, with shaded regions representing the 95% confidence interval calculated for the regression estimate, and dotted lines indicate y = x. Pearson correlation coefficient was used to measure the linear relationship between the two measures.

Figure 2:. 3D volumetric imaging of the nascent foregut lumen.

(A) At each HH stage between 8 and 12, 3D reconstructions of the foregut lumen volume are shown at left, with transverse optical sections along the antero-posterior axis shown at right (lumen pseudo colored blue); scale bar = 100 μm. (B-D) Time course quantification of lumen volume (B), surface area (C), and surface area-to-volume ratio (D) between HH stages 8 and 12; *p<0.05, **p<0.01, ***p<0.001 as determined by one-way ANOVA with Tukey’s multiple comparisons test (n = 3 per stage).

Figure 3:. Fate maps reveal limited involution of the midline endoderm across the AIP.

(A) DiI labels injected onto AIP endoderm between HH stages 7–10 (left, 0 hours) and following 4 hours of incubation (right); 4 hours spans the approximate time between HH stages in this range (HH7, n = 4; HH8, n = 7; HH9, n = 5; HH10, n = 3); scale = 200 μm. (B) DiI (red) and DiO (green) labels injected into midline presumptive roof endoderm at HH stage 8-move anteriorly toward the prechordal plate, but remain on the roof at HH stage 10+ (n = 4/4); scale = 200 μm. (C) Transverse sections through the foregut at HH stage 10 immunostained (red) for laminin (left) and E-cadherin (middle), and stained with phalloidin to visualize F-actin (right); counterstained with DAPI to visualize nuclei (cyan); R = foregut roof; F = foregut floor; scale = 50 μm (n = 3/3).

Figure 4:. Role of AIP endoderm in AIP movement and FG formation.

(A) mRNA electroporation of 3nls-GFP and DeAct-GFP into AIP endoderm (as indicated by the red square on the schematic) of HH stage 10 embryos followed by phalloidin staining to visualize effects on F-actin integrity. Phalloidin signal isolated in grayscale at right. Note that identical scales are used for DeAct-GFP and 3nls-GFP, and the appearance of smaller cells in 3nls-GFP embryos is because fluorescence is restricted to the nucleus; scale = 20 μm. (B) Fast Green injection into the foregut lumen to visualize foregut length (quantified in C) immediately following electroporation with 3nls-GFP or DeAct-GFP mRNA at HH stage 8 (targeted region indicated in green on schematic), and again after 10 hours incubation. Dashed white line indicates AIP lip; scale = 200 μm. (C) Percent elongation of foregut following mRNA electroporation of AIP endoderm with 3nls-GFP (n=9) or DeAct-GFP (n=10); ****p < 0.0001 by Student’s t-test. (D) Percentage of embryos displaying heart tube defects with 10 hours of incubation following mRNA electroporation with 3nls-GFP (left) or DeAct-GFP mRNA (right). (E) Representative images of embryos electroporated at HH stage 8+ with 3nls-GFP, yielding normal control embryos (left), compared to DeAct-GFP electroporated embryos that displayed either heart tube malformations (middle) no heart tube at all (right). White dashed lines indicate heart tube boundary; scale = 200 μm.

Figure 5:. Endodermal cell movements and morphology at the AIP during foregut elongation.

(A) Schematic ventral view of embryo during foregut morphogenesis, with regions imaged in each subpanel as indicated by red dashed boxes. (B) Snapshots of a time-lapse showing dynamic changes in cell morphology of CAAX-GFP electroporated endoderm cells at the AIP lip (dashed line) at HH stage 8-; white arrows indicate rapid polarization of cells; scale = 50 μm. (C) Progressive time-projected cell tracks from time-lapse (Movie 2) of a CAAX-GFP electroporated embryo spanning HH stages 8–11; white arrows indicate anterior cell flow in the presumptive foregut roof endoderm. Scale = 200 μm. (D, E) Whole-mount fluorescence imaging of tight junctions of an HH8+ embryo (ZO-1, green) and F-actin (Phalloidin, magenta), counterstained for nuclei (DAPI, blue), in the AIP endoderm (D) and presumptive roof endoderm (E). White arrow indicates multicellular F-actin cable; scale = 10 μm.

Figure 6:. Endodermal tissue deformations at the AIP during foregut elongation.

(A) Velocity fields quantified from time lapse of cell movements between HH stages 8+ and 10+. Orientation and magnitude of arrows indicate orientation and speed of cell movements, respectively; underlying heat map indicates speed (n = 3); scale = 200 μm. (B) Principal strain rates computed from velocity fields in (A), where major and minor cruciform axes represent maximum and minimum principal strain rates D1 and D2, respectively, and their orientations represent the associated principal directions. Negative/compressive principal strain rates are indicated in blue, and positive/extensional strain rates are indicated in red; strain rate scale = 0.01%/s; spatial scale is as in (A). (C) Maximum and minimum principal strain rates calculated from velocity fields at HH stage 8, following electroporation with 3nls-GFP RNA control (left) and DeAct-GFP RNA (right) (n=3); grey dashed line indicates AIP lip, principal strain rate scale is as indicated in (B) and spatial scale is as in (A).