June 1, 2014;
Symmetry breakage in the frog Xenopus: role of Rab11 and the ventral-right blastomere.
Vertebrates display asymmetric arrangements of inner organs such as heart
. The Nodal
signaling cascade in the left lateral plate mesoderm
in all cases directs asymmetric morphogenesis and placement during organogenesis. Mechanisms that lead up to left
induction seem to differ between the vertebrates. Cilia
produce a leftward extracellular fluid flow in zebrafish, medaka, mouse, rabbit, and Xenopus embryos during neurulation. In Xenopus, earlier asymmetric cues were described. Some, such as Rab11
, apparently act in the zygote
. Others were efficiently manipulated in ventral
cells at the four-cell stage, a lineage presumably independent of the ciliated gastrocoel roof plate
(GRP) during neurulation. Here, we show that one- and four-cell manipulations of Rab11
showed equal low efficiencies of left
disturbances. We also reevaluated the lineage of the GRP. By tracing back future ciliated cells from the gastrula
to the four-cell stage, we show that ventral
cells contribute to ciliated sensory cells at the border of the GRP. Knockdown of the Nodal
in the ventral right
lineage resulted in embryos with ectopic right
expression. Together, these experiments support a cilia
-based mechanism of symmetry breakage in the frog Xenopus.
[+] show captions
Figure 1. Manipulation of Rab11 function at the one- or four-cell stage leads to indistinguishable and weak effects on Pitx2c expression in the lateral plate mesoderm. (a-f) Effectiveness of constructs. Both Xenopus and human dnRab11 constructs interfere with integration of multiciliated cells (MCC) into the tadpole epidermis. Four-cell embryos were injected with mRNA encoding nuclear β-galactosidase (nuclear β-Gal) alone (a, d) or in combination with Xenopus (b, c) or human (e, f) wildtype (b, e) or dominant-negative (c, f) Rab11 mRNAs. Embryos were cultured to tadpoles stages and processed for scanning electron microscopy. Note that wildtype Rab11 did not impact on MCC migration while MCCs were virtually absent from dnRab11-injected specimens, as described (Kim et al., ). Scale bar represents 10 µm. (g-l) Low-frequency laterality defects in Rab11-manipulated Xenopus tadpoles. Embryos were injected at the one- or four-cell stage into the marginal zone with 2.4 (low) or 3.2 (high) ng of Xenopus or 9.6 ng of human dnRab11, cultured to Stage 30 and processed for Pitx2c ISH. Specimens were scored for wildtype (g), bilateral (h), inverted (i), or absent (j) Pitx2c expression in the LPM. (k, l) Deviations from normal left-sided patterns were pooled as LR-defects in bar graphs representing statistical evaluation of Xenopus (k) and human (l) dnRab11 injections. Note that differences between one- and four-cell injections were not significant (n.s.) in most cases, and that in the two cases where significances were recorded, the four-cell injections were more efficient. Note also that effects were overall low except for cases when very high doses of human dnRab11 were injected (l). n, number of embryos analyzed.
Figure 2. Dynamic expression pattern of Foxj1 in the blastula and gastrula embryo. Staged embryos were fixed and processed for Foxj1 WMISH. (a, a′) Stage 9 blastula embryo in vegetal view, showing dorsal (a) and ventral (a′) expression. (b) Stage 10 embryo in vegetal view, showing dorsal (b) and ventral (b′) Foxj1 expression. (c) Staining in the superficial mesoderm of the early gastrula embryo at Stage 10.5 (vegetal view). (d, e) Movement of staining toward the dorsal lip of the blastopore in Stage 11 (d) and Stage 11.5 (e) embryos (vegetal view). (f, g) Late gastrula embryos at Stage 12 (f) and 12.5 (g) in dorsal view. (h) Onset of epidermal expression in Stage 13 embryo with parallel involution of last Foxj1-positive cells over the dorsal lip (dorsal view). fp, floor plate.
Figure 3. Back-tracking of Foxj1-positive superficial mesoderm (SM) cells to dorsal and ventral blastomeres at the four-cell stage. Four-cell embryos were oriented to be viewed from the dorsal (a-j) or lateral (k-t) side and time-lapse videos were recorded at about 1 frame per every 2 min. Embryos were fixed at Stage 10.5 and processed for Foxj1 WMISH (a, b, k, l). Movies were edited to mark Foxj1-positive regions or individual cells and back-tracked to the four-cell stage: Stage 10.5 (c, m), Stage 9.5 (d, n), Stage 9�9.5 (e, o), Stage 7�8 (f, p), Stage 6�7 (g, q), Stage 5 (h, r), 32-cell stage (i, s), four-cell stage (j, t). (a�j) Dorsal view. Central and lateral regions of the SM containing Foxj1-positive cells were outlined (a, b) and followed back to the four-cell stage (j). Note that lateral signals moved out of sight in dorsal perspective, starting at Stage 8 (f). (k-t) Right-lateral view. Six cells were marked with white (ventral) or black (dorsal) asterisks, according to their origin at the four-cell stage. One cell which originated from the ventral right blastomere and was characterized by Foxj1 expression at Stage 10.5 is marked by a white asterisk outlined in red.
Figure 4. Ventral blastomeres contribute to somitic GRP cells. Embryos were injected at the four-cell stage into defined regions of the dorsal or ventral right blastomere with mRNA encoding β-Gal as lineage tracer as indicated, cultured until Stage 18, and processed for Xnr1 WMISH. Dorsal explants (a-c) and histological sections (a′-c′and a′-c′; levels indicated by dashed lines in a-c) revealed Rose-Gal staining overlapping with Xnr1 mRNA expression in the somitic GRP cells (arrowheads).
Coco knockdown in the ventral-right blastomere results in right-sided Xnr1 induction in the LPM. Embryos were coinjected with β-Gal lineage tracer and Coco-MO1/2 into defined regions of the dorsal or ventral right blastomere at the four-cell stage, cultured until Stage 22, and processed for Xnr1 WMISH. (a-c) Representative specimens of dorsal-lateral (a), ventral-lateral (b), and ventral-central (c) injections, which resulted in ectopic induction of Xnr1 in the right LPM. Histological sections (a′-c′; planes indicated by dashed lines in a-c) and higher magnifications (a′-c′; region indicated by dashed rectangle in a′-c′) demonstrate Rose-Gal positive cells in the somites in all cases (outlined and marked with 's'). Notochord (N), somites (S), and lateral plate mesoderm (LPM) outlined by dashed lines in (a′-c′).
Figure 6. Ectopic Nodal cascade induction in Coco morphants targeted to the ventral-right blastomere depend on GRP Nodal. Embryos were coinjected with β-Gal lineage tracer and Coco-MO1/2 into defined regions of the dorsal or ventral right blastomere at the four-cell stage, cultured until Stage 22 (a) or 28-32 (b) and processed for Xnr1 (a) or Pitx2c (b) WMISH. (a) Very highly significant ectopic Xnr1 induction in the right LPM in all cases. (b) Very highly significant ectopic Pitx2c induction in the right LPM depended on the presence of Nodal in the targeted cells, as coinjection of Xnr1-MO into dorsal-right lateral and Coco-MO into ventral-right central regions rescued wildtype left-asymmetric Pitx2c expression. n, number of embryos analyzed.