XB-ART-5992Curr Biol December 23, 2002; 12 (24): 2136-41.
Lefty-dependent inhibition of Nodal- and Wnt-responsive organizer gene expression is essential for normal gastrulation.
During gastrulation, diffusible "organizer" signals, including members of the TGFbeta Nodal subfamily, pattern dorsal mesoderm and the embryonic axes. Simultaneously, negative regulators of these signals, including the Nodal inhibitor Lefty, an atypical TGFbeta factor, are induced by Nodal. This suggests that Lefty-dependent modulation of organizer signaling might regulate dorsal mesoderm patterning and axial morphogenesis. Here, Xenopus Lefty (Xlefty) function was blocked by injection of anti-Xlefty morpholino oligonucleotides (MO). Xlefty-deficient embryos underwent exogastrulation, an aberrant morphogenetic process not predicted from deregulation of the Nodal pathway alone. In the absence of Xlefty, both Nodal- (Xnr2, gsc, cer, Xbra) and Wnt-responsive (gsc, Xnr3) organizer gene expression expanded away from the dorsal blastopore lip. Conversely, coexpression of Xlefty with Nodal or Wnt reduced the ectopic expression of Nodal- (Xbra) and Wnt-responsive (Xnr3) genes in a dose-dependent manner. Furthermore, Xlefty expression in the ectodermal animal pole inhibited endogenous Nodal- and Wnt-responsive gene expression in distant mesoderm cells, indicating that Xlefty inhibition can spread from its source. We hypothesize that Xlefty negatively regulates the spatial extent of Nodal- and Wnt-responsive gene expression in the organizer and that this Xlefty-dependent inhibition is essential for normal organizer patterning and gastrulation.
PubMed ID: 12498689
Article link: Curr Biol
Species referenced: Xenopus
Genes referenced: a2m cdx2 cer1 gsc lefty myod1 nodal nodal1 nodal2 nodal3.1 nodal3.2 shh tbxt tgfb1
Morpholinos: lefty MO1 lefty MO2
Article Images: [+] show captions
|Figure 1. Depletion of Xlefty Results in Exogastrulation(A) Xnr2 and (B) Xlefty are dorsally expressed in complementary adjacent domains at stage 10.5. Vegetal views, with dorsal at the top. Arrowheads indicate the dorsal lip. (C) Model of Nodal signaling at the dorsal lip in the presence (left) or absence (right) of Lefty function. Arrows represent the extent of Nodal signaling response away from the dorsal lip; vegetal view. (D–G) Xlefty depletion by MO injection results in exogastrulation that is rescued by coinjection of an Xlefty DNA construct that lacks the MO-targeted sequence. (D) Uninjected embryo (59/60 normal). (E) Exogastrulae from injection of 8 pmol anti-Xlefty MO (49/60 exogastrulated). Complete exogastrula on left, partial exogastrula on right. The arrowhead indicates the pigmented ectoderm at the posterior end (confirmed by in situ hybridization, Supplementary Figure S3) of the complete exogastrula. Complete exogastrulae (24/60) exhibited no morphologically obvious neural plate during neurulation and no dorsal closure after neurulation, whereas partial exogastrulae (25/60) had a morphologically recognizable neural plate and ≤50% dorsal closure over the yolk. (F) Injection of 100 pg CS2+/Xlefty DNA results in aberrant development but not exogastrulation (0/46 exogastrulated). (G) Coinjection of 8 pmol anti-Xlefty MO + 100 pg CS2+/Xlefty DNA rescues gastrulation (1/60 exogastrulated). Post-neurula embryos with neural plate closure and >50% dorsal closure over the yolk were scored as rescued. (D, F, and G) Left lateral view at stage 26, with anterior to the left. For exogastrulae (E), shown is a dorsal view at stage 26, with anterior at the top.|
|Figure 2. The Expansion of Nodal- and Wnt-Responsive Gene Expression in Xlefty-Deficient Embryos Indicates an Expansion of the Organizer and Dorsal Mesoderm(A and B) In Xlefty-deficient embryos (MO), expression of the organizer markers Xnr2 (Nodal responsive), gsc (Nodal/Wnt responsive), and Xnr3 (Wnt responsive) and the pan-mesodermal marker Xbra (Nodal responsive) expands into or beyond the normal Xlefty expression domain. The Nodal-responsive expression of Xlefty is also upregulated and expanded in these embryos (our unpublished data). A downregulation of the expression of the Nodal/Wnt/BMP inhibitor Cer cannot account for the expansion of Nodal- and Wnt-responsive gene expression in the Xlefty-deficient embryos. edd expression in Xlefty-deficient embryos is unchanged, suggesting that endodermal fate determination is an Xlefty-independent process. Top panels show uninjected embryos. Bottom panels show anti-Xlefty MO-injected embryos. Twenty embryos (ten pigmented and ten albino) from each treatment (uninjected versus MO injected) were examined for each marker by in situ hybridization. Albino embryos are shown for all markers except gsc and Xnr3. Markers were examined in stage 10.25–10.5 embryos, except for edd (stage 11). (A) Vegetal views. Arrowheads indicate the dorsal lip. (B) Sagittal section through the center of the dorsal lip of the embryo in (A). Dorsal is at the left, vegetal is at the bottom. Black arrowheads indicate the dorsal lip. Red arrowheads indicate the normal anterior expression boundary of Xnr2, gsc, and Xnr3, which approximately coincides with the normal posterior expression boundary of Xlefty and Xbra. Blue arrowheads indicate the normal anterior expression boundary of Xlefty and Xbra. Green arrowheads indicate the anterior boundary of expanded expression in Xlefty-deficient embryos.|
|Figure 3. Xlefty Inhibits the Induction of Nodal- and Wnt-Responsive Gene Expression. (A–C) The induction of ectopic Xbra expression in animal cap ectoderm by nodal (Xnr1) injection is inhibited by Xlefty coinjection. (D–F) Similarly, induction of ectopic Xnr3 expression by Wnt (Xwnt8) is inhibited by Xlefty coinjection. (G–I) The weak induction of Xnr3 expression by Xnr1 suggests that Xlefty does not indirectly inhibit the Xwnt8-dependent induction of Xnr3 by negatively regulating Nodal signaling. RNAs (indicated at the top of each panel) were injected into an animal cap cell of an albino 8- to 16-cell embryo. Embryos were fixed at stage 10.25–10.5 for in situ hybridization. Probes are indicated at the bottom of each panel. Animal pole views, with dorsal at the top. (A, D, and G) Uninjected embryos did not express Xbra or Xnr3 in the animal cap (n = 127). (B) Embryos injected with 800 pg Xnr1 expressed ectopic Xbra (n = 37/37). (C) Coinjection of 800 pg Xnr1 + 800 pg Xlefty reduced ectopic Xbra expression (n = 37/37). (E) Injection of 60 pg Xwnt8 induced ectopic Xnr3 expression (n = 49/49), but not Xbra expression (n = 10/10; not shown). (F) Coinjection of 60 pg Xwnt8 + 600 pg Xlefty reduced ectopic Xnr3 expression (n = 18/20). (H) Injection of 800 pg Xnr1 weakly induced Xnr3 (shown; n = 2/21). Most embryos exhibited even weaker (n = 15/21) or no Xnr3 (n = 4/21) induction. (I) Coinjection of 800 pg Xnr1 + 800 pg Xlefty eliminated (shown; n = 11/20) or greatly reduced (n = 7/20) ectopic Xnr3 expression.|
|Figure 4. Xlefty Acts at a Distance to Inhibit Endogenous Nodal- and Wnt-Responsive Gene Expression. (A) Xlefty RNA (100 pg) was injected into a dorsal left cell of the top tier of a pigmented 32-cell embryo. Coinjection of a lineage tracer confirmed targeting (our unpublished data). A is animal pole, Vg is vegetal pole, D is dorsal, and V is ventral. (B–G) Injected embryos were fixed at stage 10.25–10.5 and examined by in situ hybridization so that the endogenous expression of (B and E) Xbra, (C and F) Xnr3, and (D and G) gsc could be assessed. Vegetal views, with dorsal at the top. Arrowheads indicate the dorsal lip. (B–D) Uninjected embryos displayed normal expression of Xbra ([B], n = 20/20), Xnr3 ([C], n = 19/20), and gsc ([D], n = 10/10). In contrast, injection of Xlefty into the animal cap severely reduced or eliminated the endogenous expression of Xbra ([E], n = 20/20 reduced), Xnr3 ([F], n = 11/20 absent; 9/20 reduced), and gsc ([G], n = 9/10 absent; 1/10 reduced). These results indicate that ectopic Xlefty expression in an ectoderm lineage (green, [A]) can inhibit endogenous Nodal- and Wnt-responsive gene expression in distant mesoderm lineages (red and yellow, [A]).|
|Figure 5. Xlefty Knockdown Sequentially Results in Organizer Expansion, a Spatial Shift of Dorsal Mesoderm, Convergent Extension of Noninvoluted Dorsal Mesoderm, and Exogastrulation (A) In the absence of Xlefty function, dorsal Nodal- and Wnt-responsive gene expression domains, indicative of the dorsal organizer (gsc, Xnr2, Xnr3) and the presumptive dorsal mesoderm (Xbra), expand toward the animal cap. Vegetal view, with dorsal to the top. Arrows represent the extent of the Nodal/Wnt signaling responses that specify the organizer. (B) (top) Normally by mid-gastrulation, the Xbra-expressing cells of the presumptive dorsal mesoderm have involuted and begun to undergo convergent extension. By the end of neurulation, the convergent extension of dorsal embryonic tissues has resulted in the elongation of the notochord and anterior (A)/posterior (P) axis. (bottom) In the absence of Xlefty function, the dorsal organizer (gsc) and presumptive dorsal mesoderm (Xbra) expand. Furthermore, the expansion of the organizer shifts the presumptive dorsal mesoderm further from the dorsal lip. We propose that this shift prevents or delays the dorsal mesoderm in Xlefty-deficient embryos from involuting by mid-gastrulation, yet the cell-autonomous convergent extension movements of the dorsal mesoderm still initiate normally. Subsequently, an external extension of the noninvoluted dorsal mesoderm results in exogastrulation and an inversion of the A/P axis. The gastrulae are depictions of a sagittal section through the dorsal lip. The neurulae are lateral views. Blue arrows indicate the direction of convergent extension.|
|Figure S1. Xlefty-Deficient Exogastrulation Is a Mid-Gastrulation Event In Xlefty-deficient embryos (MO), gastrulation initiates normally, as indicated by the presence of the dorsal blastopore groove (black arrowheads) at stage 10.25. By mid-gastrulation (stage 11), blastopore closure in MO embryos as compared to controls is clearly delayed. By stage 12, yolk eversion from the blastopore of the MO embryo indicates exogastrulation has begun, and by stage 12.5, exogastrulation appears to be relatively complete. Uninjected (left, n 6) and MO-injected (right, n 6) gastrulae incubated at 15C were photographed every 2 hrs over a 12 hr period. Vegetal views, with dorsal at the top, except for stage 12 and 12.5 exogastrulae (lateral views, with dorsal at the top and anterior to the left). Note that in Xlefty-deficient embryos the presence of a dorsal blastopore groove and bottle cells (black arrowhead) at stage 10.25, a dorsal blastopore lip at stage 10.5 (Figure 2B in the main text), and a ventral blastopore lip (red arrowheads) at stage 11, in addition to some blastopore closure (MO stage 10.25 versus MO stage 11), suggests that some involution and epiboly have occurred. Additionally, evidence of involution was derived from dissections of developing exogastrulae that revealed cell migration along the blastocoel roof and a subsequent reduction of the blastocoel cavity (our unpublished data). Thus, exogastrulation is probably not due to loss of involution or epiboly. Inhibition of tissue separation is probably not the cause of Xlefty-deficient exogastrulation, either, because Brachet’s cleft forms (Figure 2B in the main text, gsc and Xnr3), the cells migrating on the blastocoel roof of the exogastrulae do not sink into the roof (our unpublished data), and exogastrulation did not result from other experimental approaches that block tissue separation [S7]. Furthermore, the presence of Brachet’s cleft and involution suggest that vegetal rotation occurs normally in Xleftydeficient embryos and is not the cause of exogastrulation [S8].|
|Figure S3. Post-Gastrula Xlefty-Deficient Exogastrulae Exhibit a Reversal of the A/P Axis and Normally Patterned Mesoderm and Endoderm The expression of gsc and Xbra, markers of the prechordal plate and notochord, respectively, and of Xcaudal2 (Xcad2) [S9], a marker of posterior development, indicates a reversed A/P axis and normal dorsal patterning in the Xlefty-deficient exogastrulae. In addition to Xbra, the expression of Xenopus sonic hedgehog (Xshh) [S10] and myoD [S11], markers of the axial and paraxial mesoderm, respectively, indicates a normal complement of relatively normally patterned mesoderm in the exogastrulae. The expression of edd indicates a normal complement of endoderm in the exogastrulae. Left and right panels show uninjected and anti-Xlefty morpholinoinjected (MO) albino embryos, respectively. gsc and Xbra were examined at stage 13/14 (dorsal views, with anterior to the left). All other markers were examined at stage 25/26 (left lateral views, with dorsal to the top and anterior to the left). Fifteen to twenty embryos were examined for each marker by whole-mount in situ hybridization. Previously, exogastrulae have been used for studying vertical versus planar neural induction, with the assumption that in exogas trulae the mesoderm and ectoderm are never vertically apposed because involution does not occur [S12]. However, the validity of this assumption is still controversial [S13, S14]. Because the Xlefty- deficient exogastrulae undergo some involution (Figure S1), transient vertical apposition of the mesoderm and ectoderm probably occurs in these embryos and renders them not useful for resolving this controversy regarding neural induction.|
Solnica-Krezel, Vertebrate development: taming the nodal waves. 2003, Pubmed