XB-ART-40030Mech Dev October 1, 2009; 126 (10): 913-23.
Retinoic acid regulates anterior-posterior patterning within the lateral plate mesoderm of Xenopus.
The lateral plate mesoderm (LPM) lines the body cavities, gives rise to the heart and circulatory system and is responsible for patterning the underlying endoderm. We describe gene expression domains within the lateral plate mesoderm of the neurula stage Xenopus embryo that demonstrate a marked anterior posterior pattern in that tissue. FoxF1 and Nkx-2.5 are expressed in the anterior LPM, Hand1 in the middle and Xsal-1 in the posterior LPM. Since retinoic acid is known to pattern many tissues during development, and RALDH2, the enzyme primarily responsible for retinoic acid synthesis, is expressed in the anterior and dorsal LPM, we hypothesized that retinoic acid is necessary for correct patterning of the LPM. Exposure to exogenous retinoic acid during neurulation led to an expansion of the anterior and middle expression domains and a reduction of the posterior domain whereas exposure to a retinoic acid antagonist resulted in smaller anterior and middle expression domains. Furthermore, inhibition of RALDH2, which should decrease endogenous RA levels, caused a reduction of anterior domains indicating that endogenous RA is necessary for regulating their size. After altering retinoic acid signaling in a temporally restricted window, the displaced anterior-posterior pattern is maintained until gut looping, as demonstrated by permanently altered Hand1, FoxF1, xHoxC-10, and Pitx2 expression domains. We conclude that the broad expression domains of key transcription factors demonstrate a novel anterior-posterior pattern within the LPM and that retinoic acid can regulate the size of these domains in a coordinated manner.
PubMed ID: 19595764
Article link: Mech Dev
Genes referenced: aldh1a2 cyp26a1 foxf1 hand1 nkx2-5 pitx2 sall3 hoxc10
Article Images: [+] show captions
|Fig. 1. The expression pattern of Hand1 and Xsal-1 in the early Xenopus embryo. The white numbers indicate the developmental stage of the embryo. Expression of Hand1 (A) is first detectable immediately following gastrulation and continues into the swimming tadpole stage in the middle LPM. At stage 20, it is clear that the Hand1 staining does not extend to the posterior end of the LPM. Xsal-1 (B) is expressed during late gastrulation in the presumptive neural plate, and by stage 14 in the posterior LPM (arrows). Expression in the posterior LPM disappears by the end of neurulation, while expression in the neural tube persists into the late tailbud stage. The embryos are viewed from the side with dorsal up and the anterior end to the left, with the exception of the stage 12 Xsal-1 staining where the embryo is viewed from the posterior end looking at the dorsal surface.|
|Fig. 2. Model of gene expression as seen in the left LPM at the neurula stage. (A) Expression pattern of RALDH2 at the neurula stage Xenopus embryo showing expression in the anterior LPM (a) as well as expression in the dorsal LPM and somites (b). (B) Expression patterns of the LPM markers showing the predicted overlap of expression domains. Graphical representation shows the left LPM, with dorsal toward the top, and anterior toward the left. In grey: RALDH2, purple: Nkx-2.5, red: FoxF1, yellow: Hand1, light blue: Xsal-1.|
|Fig. 3. The expression domains of Nkx-2.5, FoxF1, Hand 1 and Xsal-1 are altered by RA signaling. Embryos were treated with RAA (left panel), RA (right panel) or DMSO (center panel) and were allowed to develop until stage 20/22. Whole-mount in situ hybridization for Nkx-2.5 (A–C) demonstrates little change under the RA treatment (C), as compared to the DMSO control (B). In contrast, the expression domain of the anterior-dorsal marker FoxF1 (D–F) extends further to the posterior end of the embryo under exogenous RA conditions (F) and reduced under treatment with the RAA (D) as compared to the DMSO control (E). Similarly, the middle marker Hand1 (G–L) demonstrates a reduced expression domain under the RAA treatment (G), while the domain is expanded when treated with RA (I) as compared to the DMSO control (H). Note that when viewed from the side (G–I), the greatest expansion in the Hand1 domain is on the dorsal edge, where there is also increased staining intensity (arrowhead), when treated with RA. In contrast, when viewed from the side (M–O), the lateral Xsal-1 domain is virtually absent under treatment with RA (O) when compared to either RAA-treated (M) or control embryos (N). When viewed from the anterior end (Q–S), other areas of Xsal-1 expression are also affected by RA. The banded expression in the developing brain is reduced and there is increased expression in staining in presumptive ganglia when compared to RAA-treated (Q) or control embryos (R).|
|Fig. 4. Altering endogenous RA levels altered the expression of Hand1 and FoxF1 but had little effect on the expression of Nkx-2.5 or Xsal-1. Embryos were treated with ketoconazole (left column) to increase endogenous RA levels by inhibiting Cyp26 and DEAB or citral was used to lower endogenous RA levels by inhibiting RALDH2. The expression domain of Nkx-2.5 (A–D) was not noticeably altered by any of the treatments when compared to DMSO controls (B). In contrast, the expression of FoxF1 (E–H) was not obviously altered by treatment with ketoconazole (E) but both DEAB and citral (G and H) treatments reduced the expression domain of FoxF1 as compared to the DMSO control embryos (F). Similarly, the size of the Hand1 (I–L) domain was not clearly altered by ketoconazole (I) when compared to controls (J), but both DEAB and citral (K and L) diminished the expression domain. The expression domain of Xsal-1 was not altered by any of the treatments (M–P). Panels A–D show the anterior pole of the embryo, with dorsal toward the top; panels E–P show the left lateral side of the embryo, with dorsal toward the top, anterior at left.|
|Fig. 5. Altered RA signaling alters the expression of Hand1 and Xsal-1 in the absence of dorsal tissue. As in the embryo, the Hand1 expression domain area was found to be enlarged under RA treatment (C) and reduced with RAA (A) when compared to controls (B). Measurements and comparison of the total staining area confirmed that this change was statistically significant (D; p < 0.05; n > 10 embryos/treatment). Ventral explants also show the same lack of Xsal-1 expression in the LPM (arrows) when treated with RA (F) when compared to control explants (E) suggesting that the RA is acting directly on the LPM to alter the expression domains of Hand1 and Xsal-1. Explants are oriented with the anterior pole toward the top.|
|Fig. 6. Changes in Xsal-1 but not Hand1, due to altered RA signaling, are blocked by the addition of cycloheximide. Embryos treated with cycloheximide had no change in the expression of either Hand1 (B) or Xsal-1 (F) when compared to control embryos (A and E). Expansion of the Hand1 domain, when treated with RA (C), was also observed in embryos treated with both RA and cycloheximide (D). When the embryos were treated with RA for one hour, there was a reduction in the Xsal-1 expression domain although not the severe loss seen in the 4–5 h treatments (G). In the presence of cycloheximide, the reduction in the Xsal-1 domain was not observed (H). In contrast, the expression domain of Xsal-1 under treatment with RA and cycloheximide is not reduced to the extent seen in RA alone (H). In each panel the left lateral side is shown with the anterior end to the left, dorsal to the top. Cycloheximide (Cx).|
|Fig. 7. The changes to the LPM patterning persist after temporally limited alterations in RA signaling. Experimental changes in RA signaling were limited to between stages 14–20 and the embryos were fixed at stage 34–36 for in situ hybridization for FoxF1 (A–C), Pitx2 (D–F), Hand1 (G–I), and HoxC-10 (J–L). The expression domain is reduced under treatment with RAA (A, D, and G), and expanded under RA (C, F, and I) as compared to the DMSO controls (B, E, and H) for all of FoxF1, Pitx2, and Hand1. While the domain of the posterior marker HoxC-10 was anteriorly displaced under RAA (J) as compared to DMSO controls (K), the domain is unaffected under treatment with RA. In each case, the lengths of the domains were measured as a ratio of the length from the cement gland to the most posterior point of LPM staining (x) to the length from the cement gland to the back of the body (y) for the anterior FoxF1, Pitx2, and Hand1 (M). The differences in the length of the staining domain were found to be significant (O, P and Q; p < 0.05; n > 15) between treatments (where *, **, and *** represent statistically significant groups). HoxC-10 was similarly measured (N) using the length of the body axis from the cement gland to the back of the body axis (y), however the length of the domain was measured from the back of the body axis to the anterior edge of the domain (z). The length was found to be significantly different between the DMSO controls and the RAA treatment (p < 0.05; n > 15) however no significant difference was found between RA treatments and controls (R). In all embryos the left lateral side is shown, with the anterior pole to the left, dorsal to the top.|
|sall3 (spalt-like transcription factor 3) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage12, posterior/blastoporal view, dorsal up.|
|sall3 (spalt-like transcription factor 3) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 20, lateral view, anterior left, dorsal up.|
|sall3 (spalt-like transcription factor 3) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 32, lateral view, anterior left, dorsal up.|
|sall3 (spalt-like transcription factor 3) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 35 & 36, lateral view, anterior left, dorsal up.|
|sall3 (spalt-like transcription factor 3) gene expression in Xenopus laevis embryo, assayed via in situ hybridization, NF stage 14, lateral view, anterior left, dorsal up.|