PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Nat Cell Biol. Author manuscript; available in PMC Jan 12, 2010.
Published in final edited form as:
Published online Feb 19, 2006. doi:  10.1038/ncb1374
PMCID: PMC2805411
NIHMSID: NIHMS165925
Retinoic-acid signalling in node ectoderm and posterior neural plate directs left–right patterning of somitic mesoderm
Ioan Ovidiu Sirbu1 and Gregg Duester1,2
1 OncoDevelopmental Biology Program, Burnham Institute for Medical Research, 10901 North Torrey Pines Road, La Jolla, CA 92037, USA
2 Correspondence should be addressed to G.D. (duester/at/burnham.org)
Somitogenesis requires bilateral rhythmic segmentation of paraxial mesoderm along the antero-posterior axis1. The location of somite segmentation depends on opposing signalling gradients of retinoic acid (generated by retinaldehyde dehydrogenase-2; Raldh2) anteriorly and fibroblast growth factor (FGF; generated by Fgf8) posteriorly2,3. Retinoic-acid-deficient embryos exhibit somite left–right asymmetry46, but it remains unclear how retinoic acid mediates left–right patterning. Here, we demonstrate that retinoic-acid signalling is uniform across the left–right axis and occurs in node ectoderm but not node mesoderm. In Raldh2−/− mouse embryos, ectodermal Fgf8 expression encroaches anteriorly into node ectoderm and neural plate, but its expression in presomitic mesoderm is initially unchanged. The late stages of somitogenesis were rescued in Raldh2−/− mouse embryos when the maternal diet was supplemented with retinoic acid until only the 6-somite stage, demonstrating that retinoic acid is only needed during node stages. A retinoic-acid-reporter transgene marking the action of maternal retinoic acid in rescued Raldh2−/− embryos revealed that the targets of retinoic-acid signalling during somitogenesis are the node ectoderm and the posterior neural plate, not the presomitic mesoderm. Our findings suggest that antagonism of Fgf8 expression by retinoic acid occurs in the ectoderm and that failure of this mechanism generates excessive FGF8 signalling to adjacent mesoderm, resulting initially in smaller somites and then left–right asymmetry.
In the ‘clock and wave front’ model of somitogenesis, a moving wavefront of Fgf8 gene expression in the tailbud regresses posteriorly as the body axis extends, and mesodermal segmentation occurs just anterior to the Fgf8-expression domain7,8. The anterior extent of Fgf8 expression is limited by retinoic acid that is generated in the presomitic mesoderm and antagonizes Fgf8 expression in the tailbud2,9. Retinoic acid synthesized in the presomitic mesoderm by RALDH2 is required to maintain bilateral symmetry between the left and right somite columns46. Presomitic mesoderm in mouse Raldh2−/− embryos displays left–right asymmetric expression of Hes7 and Lfng, genes involved in oscillator function4, suggesting that a loss of retinoic acid allows left–right asymmetry to occur in presomitic mesoderm where it normally does not occur. However, lateral-plate mesoderm in Raldh2−/− embryos still maintains asymmetric expression of Nodal and Pitx2, genes required for left-right asymmetry10. Retinoic acid acts as a buffer to prevent left–right asymmetry from occurring in presomitic mesoderm, but its mechanism of action is unclear.
Genetic studies in mice have demonstrated that oxidation of retinol to retinaldehyde is ubiquitously catalysed by several alcohol dehydrogenases11, whereas oxidation of retinaldehyde to retinoic acid occurs in a tissue specific manner, under the control of Raldh2 (Aldh1a2)12,13. All-trans-retinoic acid synthesized by RALDH2 is the endogenous retinoid ligand needed for embryogenesis and functions through nuclear retinoic-acid receptors (RARs), whereas 9-cis-retinoic acid (which binds the retinoid X receptor, RXR), is undetectable in mouse embryos and unnecessary for embryogenesis14. In early-somite-stage mouse embryos, Raldh2 expression was limited to somitic mesoderm and anterior presomitic mesoderm (Fig. 1a). Retinoic-acid activity detected in mouse embryos carrying the retinoic-acid response element (RARE)–lacZ-reporter transgene15 demonstrated that retinoic-acid signalling occurs in somitic and presomitic mesoderm, as well as in adjacent neural plate (Fig. 1b; n = 5 of 5). Transverse sections through the node revealed no left–right difference in retinoic-acid signalling activity in presomitic or lateral plate mesoderm at early somite stages (Fig. 1d; n = 3 of 3). Raldh2−/− embryos carrying RARE–lacZ lose all RARE–lacZ expression from presomite to 10-somite stages, demonstrating that RARE–lacZ is specific for detection of retinoic-acid activity and that RALDH2 is responsible for all retinoic-acid synthesis (Fig. 1c; n = 6 of 6). A sagittal section of a 4-somite wild-type embryo double-stained for RARE–lacZ and Fgf8 expression indicated retinoic-acid activity extending from the posterior neural plate into the node ectoderm where it meets the Fgf8 expression domain in the epiblast (Fig. 1e). Although retinoic-acid activity exists in presomitic mesoderm (Fig. 1b, d), it was absent in the node mesoderm (Fig. 1e). As our data shows that retinoic-acid signalling is absent in the node mesoderm and equivalent in left and right presomitic and lateral plate mesoderm, they do not support the hypothesis that cilia-driven node-vesicular parcels carry retinoic acid from right to left mesoderm as previously proposed16. Instead, our findings suggest that retinoic acid functions in the node ectoderm during left–right patterning of somites.
Figure 1
Figure 1
Retinoic-acid signalling is uniform across the left–right axis and present only in node ectoderm. (a) Raldh2 mRNA expression in wild-type 5-somite mouse embryo (dorsal view). (b, c) RARE–lacZ expression (detected with salmon–gal) (more ...)
Loss of retinoic-acid signalling results in left–right asymmetry of somites after the eight somite stage46. This was also observed in 8–14-somite Raldh2−/− embryos by analysis of Uncx4.1, a homeobox gene involved in somite patterning (Fig. 2a–b; n = 7 of 7) and Papc, endoding a paraxial protocadherin involved in convergent extension (Fig. 2c–d; n = 4 of 4). Mesp2 expressed in the presomitic mesoderm establishes segmental borders through suppression of Notch activity17. The Xenopus laevis homologue of Mesp2 was reported to require retinoic acid for expression3. Mesp2 expression in Raldh2−/− embryos at somite stages 4–5 shifted bilaterally and anteriorly, but no significant reduction in expression was observed (Fig. 2e–f; n = 3 of 3). Double-staining for RARE–lacZ and Mesp2 expression in a 5-somite wild-type embryo showed that retinoic-acid signalling activity is normally detected in the Mesp2 expression domain at this stage (Fig. 2g). Our results indicate that mouse Mesp2 does not require retinoic acid for expression, but that retinoic acid is required to set its position along the antero-posterior axis. Tbx18 encodes a T-box transcription factor that marks the anterior domain of each somite18. Tbx18 expression in Raldh2−/− embryos at somite stages 5–14 demonstrated that somites were smaller all along the antero-posterior axis, but it was also observed that Tbx18 expression was always reduced or lost in somites 3–14 on both the left and right sides (Fig. 2h–k; n = 7 of 7). This differs from Uncx4.1 expression which was differentially reduced on the right side after the 8-somite stage (Fig. 2b). These findings indicate that before the left–right somite patterning defect, a loss of retinoic acid results in reduced somite size that is associated with a bilateral anterior shift of Mesp2 expression and bilateral reduction of Tbx18 expression.
Figure 2
Figure 2
Bilateral somite-size defects precede left–right defects in the absence of retinoic-acid signalling. (a, d) Uncx4.1 mRNA27 expression (a, b) at 11 somites marks the posterior domain of each somite and Papc mRNA28 expression (c, d) at 14 somites (more ...)
Fgf8 expression in the posterior region of retinoic-acid-deficient embryos is upregulated and in some cases left–right asymmetric expression is observed2,46. As Fgf8 is normally expressed in both the presomitic mesoderm and the epiblast (primitive ectoderm), the deregulated Fgf8 expression observed in retinoic-acid-deficient embryos was examined in more detail. Fgf8 expression in Raldh2−/− embryos with 11–13 somites was found to be higher on the right side than on the left in both neuroectoderm and presomitic mesoderm (Fig. 3a, b; n = 3 of 3). At somite stages 3–5, left–right asymmetric expression of Fgf8 in Raldh2−/− embryos was not observed, nor was any effect on expression in presomitic mesoderm; however, a bilateral anterior shift of Fgf8 expression into node ectoderm and neural plate, where it is not normally expressed, was evident (Fig. 3c–f; n = 4 of 4). Raldh2−/− embryos at the one-somite stage already displayed the anterior shift of Fgf8 expression in the ectoderm (Fig. 3g,h); sagittal sections revealed that a loss of retinoic acid allows Fgf8 expression to expand from epiblast into the node ectoderm and the neural plate ectoderm (Fig. 3i–l; n = 2 of 2). Furthermore, a loss of retinoic acid did not result in an anterior advance of Fgf8 expression in the presomitic mesoderm (where expression remained unchanged in early somite embryos), and Fgf8 expression was still absent in the node mesoderm (Fig. 3i–l). Thus, we concluded that retinoic acid antagonizes Fgf8 expression in ectoderm (but not mesoderm) at early somite stages, and that left–right asymmetric expression of Fgf8 in presomitic mesoderm of later stage retinoic acid-deficient embryos is a secondary event.
Figure 3
Figure 3
Retinoic-acid signalling activity antagonizes Fgf8 expression only in ectoderm. (a, b) Fgf8 mRNA expression at 13 somites shows left–right asymmetry in an Raldh2−/− embryo (dorsal view). (c, d) Fgf8 mRNA expression at 3 somites (more ...)
To determine whether retinoic acid is required for patterning all somites during embryogenesis, we took advantage of the ability to rescue Raldh2−/− embryos by supplementing the maternal diet with low doses of retinoic acid10,14. It is also possible to detect embryonic sites of retinoic-acid signalling by analysing these rescued Raldh2−/− embryos carrying the RARE–lacZ transgene. When maternal-dietary retinoic-acid supplementation was ended at embryonic-day (E)8.25 (approximately six-somite stage) and embryos were examined at E9.5 (22–24 somites) for Uncx4.1 expression, all somites in rescued Raldh2−/− embryos were of normal size and exhibited normal left–right bilateral symmetry (Fig. 4a, b; n = 3 of 3). Double-staining of these embryos for RARE–lacZ and Uncx4.1 expression indicated that retinoic-acid signalling activity was completely absent in the somitic and presomitic mesoderm of rescued Raldh2−/− embryos (Fig. 4a, b). Similar experiments demonstrated that Mesp2 was expressed normally in the presomitic mesoderm of E9.5 Raldh2−/− embryos rescued to only E8.25 (Fig. 4c, d; n = 5 of 5). In 4–7-somite Raldh2−/− embryos rescued to E8.25, the antero-posterior border of Fgf8 expression was returned to normal, and retinoic-acid signalling activity was undetectable in the mesoderm but present in the neural plate where it was adjacent to Fgf8 expression in the epiblast (Fig. 3m, n; n = 3 of 3). These findings indicate that retinoic acid is not required for later-forming somites and that retinoic acid limits ectodermal Fgf8 expression to the epiblast during early somite stages.
Figure 4
Figure 4
Retinoic-acid-signalling requirement for somitogenesis occurs only during the early somite stages. (a–d) E9.5 wild-type embryos (WT) and Raldh2−/− embryos rescued with maternal dietary retinoic-acid supplementation to E8.25 (Res (more ...)
These observations prompted us to examine more closely the expression patterns of Raldh2 (the retinoic acid source) and Fgf8 (a target of retinoic acid action). Raldh2 is initially expressed at E7.5, before somitogenesis in paraxial mesoderm, where it functions as a source of the retinoic acid required for hindbrain antero-posterior patterning19. At the two-somite stage, Raldh2 mRNA is localized in somitic and presomitic paraxial mesoderm but not the node, whereas Fgf8 mRNA is found in presomitic mesoderm and epiblast, with an anterior expression boundary at the posterior lip of the node (Fig. 4e, f). Thus, Raldh2 and Fgf8 expression overlaps in the presomitic mesoderm at this early stage. Double-staining demonstrated that, as development proceeds, a gap between the Raldh2 and Fgf8 expression domains occurs in the presomitic mesoderm by the 6-somite stage (Fig. 4g), but that the close relationship between the Mesp2 and Fgf8 expression domains remains constant (Fig. 4h). Thus, after six somites, Raldh2 expression does not occur in presomitic mesoderm. Double-staining of wild-type embryos for RARE–lacZ and Mesp2 expression shows that retinoic-acid activity retracts anteriorly from somite stages 7–10 such that by 10 somites there is no longer retinoic-acid activity around the Mesp2 expression domain (Fig. 4i–k). These findings provide evidence that after the node regresses (10 somites) Raldh2 may no longer be needed to limit Fgf8 expression. The timing of this separation is consistent with our observation that maternal dietary retinoic-acid supplementation is not needed after the 6-somite stage to achieve correct somite patterning in rescued Raldh2−/− embryos.
As maternal dietary supplementation of Raldh2−/− embryos to E8.25 is sufficient for correct patterning of early and late somites, we examined these embryos carrying RARE–lacZ at E8.25 to detect embryonic sites of retinoic-acid signalling that were stimulated by maternally derived retinoic acid. RARE–lacZ detected in rescued embryos must be derived from maternal retinoic acid (the levels of which have been elevated by dietary retinoic-acid supplementation). Surprisingly, rescued Raldh2−/− embryos at somite stages 2–10 exhibited no RARE–lacZ expression in somitic or presomitic mesoderm (normally the source of retinoic acid), but RARE–lacZ was expressed in the posterior neural plate (Fig. 5a; n = 5 of 5). Transverse sections verified the absence of RARE–lacZ expression in Raldh2−/− mesoderm (Fig. 5f, g). Para-sagittal and mid-sagittal sections through a two-somite-stage rescued Raldh2−/− embryo indicated the presence of strong RARE–lacZ expression in the posterior neural plate and node ectoderm, but no expression in somitic mesoderm, presomitic mesoderm, or node mesoderm (Fig. 5h, i). However, RARE–lacZ expression was observed in the cranial mesoderm of rescued Raldh2−/− embryos, indicating that the absence of the RARE–lacZ signal in trunk-paraxial mesoderm is authentic and not caused by mutant mesodermal retinoic-acid activity being below the limit of detection (Fig. 5h). These observations indicate that retinoic-acid signalling does not need to function in the presomitic mesoderm where it is synthesized, but instead retinoic acid travels to the adjacent ectoderm to function in somitogenesis.
Figure 5
Figure 5
The targets of retinoic-acid signalling for somitogenesis are the node ectoderm and the posterior neural plate. (a–i) E8.25 wild-type embryos (WT) and Raldh2−/− embryos rescued with maternal dietary retinoic-acid supplementation (more ...)
Our findings suggest that a low dose of maternal dietary retinoic acid functions preferentially in the posterior neuroectoderm, rather than the somitic or presomitic mesoderm. One possibility is that retinoic acid may be preferentially degraded in certain tissues — as demonstrated in the hindbrain, which induces the retinoic-acid-degrading enzyme encoded by Cyp26c1 to create a retinoic-acid boundary19. However, expression of Cyp26a1, Cyp26b1 and Cyp26c1 in early somite-stage mouse embryos does not occur in the somitic mesoderm19,20. A second possibility is that incoming retinoic acid may be preferentially sequestered by cellular retinoic-acid-binding proteins (CRABPs) that facilitate retinoic-acid signalling21. Consistently, we found that Crabp2 was preferentially expressed in the posterior neural plate at somite stages 6–8 (Fig. 5b, c; n = 3 of 3). A final possibility is that although retinoic-acid receptors are widely expressed throughout the embryo at early stages22, they may not respond to retinoic acid in certain tissues because of the presence of factors that antagonize retinoic-acid-mediated transcription, such as the orphan nuclear receptor, COUP–TF, which binds the same DNA response element as retinoic-acid receptors and functions as a repressor23. At somite stages 3–5 we found that COUP–TFII was expressed in the somitic and anterior presomitic mesoderm but was absent in the neural plate (Fig. 5d, e; n = 2 of 2). Expression of Crabp2 and COUP–TFII at early somite stages is consistent with low-dose maternal retinoic-acid functioning preferentially in the posterior neuroectoderm.
Here, we provide significant insight into the role of retinoic acid during left–right patterning and antagonism of Fgf8. Our RARE–lacZ experiments indicate that retinoic-acid signalling does not occur in the node mesoderm (that is, ventral node) during establishment of left–right symmetry, and that retinoic-acid signalling does not occur at higher levels in mesoderm on the left side of the embryo. This finding contrasts with a study that reports the use of antibodies to detect retinoic acid on the surface of nodal vesicular parcels released from the ventral node and transported by leftward nodal flow16. This model predicts higher retinoic-acid signalling in left mesoderm, but this was not observed in our studies. Previous studies have also indicated a role for retinoic-acid signalling in the positioning of somite boundaries2,3 and somite left–right patterning46, presuming that retinoic acid functions in the presomitic mesoderm during each cycle of somite formation. As Raldh2 is expressed in the presomitic mesoderm it was reasonable to hypothesize that retinoic acid may function in somite patterning in this tissue. However, our studies demonstrate that retinoic acid is needed only during the early somite stages (when the node is present) and that retinoic acid synthesized in the presomitic mesoderm by RALDH2 travels to the adjacent neural plate and node ectoderm where it regulates early somite patterning. Also, as Fgf8 is expressed in both presomitic mesoderm and epiblast ectoderm it was reasonable to assume that retinoic-acid antagonism of Fgf8 expression may occur in presomitic mesoderm. However, our data indicate that antagonism occurs only in ectoderm. Thus, our conclusion that retinoic acid acts on ectoderm rather than presomitic mesoderm to exert its effects on left–right patterning is based, not only on RARE–lacZ expression in the ectoderm of rescued Raldh2−/− embryos, but also on ectopic expression of Fgf8 in the same ectodermal tissue of unrescued mutants. This conclusion is also supported by our observation of selective expression of Crabp2 (a stimulator of retinoic-acid signalling) in ectoderm and expression of COUP–TFII (an antagonist of retinoic-acid signalling) in mesoderm. It has also been observed that conditional loss of Fgf8 expression in presomitic mesoderm does not have an effect on somitogenesis24.
As Fgf8 is essential for regulation of left–right patterning25 and given our data indicate that retinoic acid functions in the ectoderm during somite left–right patterning, we propose that retinoic acid controls left–right patterning through its ability to control the anterior limit of Fgf8 mRNA in the posterior ectoderm. The effects seen on the mesodermal expression of Mesp2, Papc, Uncx4.1, and Tbx18 in Raldh2−/− embryos may be attributed to the anterior encroachment of ectodermal Fgf8 mRNA; this may lead to production and release of FGF8, which increases FGF signalling in adjacent presomitic mesoderm and thus pushes somite formation further anterior. By correctly establishing the antero-posterior limit of ectodermal Fgf8 expression during the node stages, retinoic acid signalling may ensure that Fgf8 expression in presomitic mesoderm does not become subject to later signals that would push it to asymmetry along the left–right axis and thus result in asymmetric somite formation. As the requirement for retinoic acid in somitogenesis ends after the node has regressed, this suggests that retinoic-acid signalling coordinates the early simultaneous functions of FGF8 in controlling somite formation (through action in the presomitic mesoderm) and left–right asymmetry (through action in the node). These findings also provide more evidence that retinoic acid synthesized in the paraxial mesoderm does not function in this tissue, but instead operates in a non-cell autonomous manner, as a signal to adjacent tissues (including hindbrain neuroectoderm19, spinal cord neuroectoderm9, foregut endoderm26) and to posterior neural plate and node ectoderm as detailed here.
Generation of Raldh2−/− embryos and rescue with a physiological dose of retinoic acid
Raldh2−/− mice have been previously described13. Embryos derived from timed matings of Raldh2+/− mice were genotyped by PCR analysis of yolk-sac DNA. After mating, we designated noon on the day of vaginal plug detection as E0.5. Embryos were staged according to somite number. Rescue of Raldh2−/− embryos by maternal dietary retinoic-acid supplementation was performed as previously described with a retinoic-acid dose that has been demonstrated to be within the normal physiological range14. This dose stimulates development of the heart, somites and other structures up to E13.5 (ref. 10). Briefly, all-trans-retinoic acid (Sigma, St Louis, MO) was dissolved in dimethylsulphoxide (50 mg ml), diluted 1:10 with corn oil, and mixed with powdered mouse chow to provide a final retinoic-acid concentration of 0.1 mg g−1 for treatment from E6.75–E8.25. Food was prepared fresh twice each day (morning and evening) and provided ad libitum. In some cases embryos were analysed at E8.25 when the mother was still receiving the retinoic-acid-supplemented diet. For embryos analysed after E8.25, mothers received standard mouse chow from E8.25 until the point of analysis at E9.5.
Retinoic-acid detection and in situ hybridization
Detection of retinoic-acid activity was performed in embryos carrying the RARE–lacZ–retinoic-acid-reporter transgene, which places lacZ (encoding β-galactosidase) under the transcriptional control of a retinoic-acid response element (RARE)15. β-galactosidase activity was detected using either X-gal (5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside) to produce a blue reaction product or salmon-gal (6-chloro-3-indolyl-β-D-galactopyranoside; Labscientific, Inc., Livingstone, NJ) to produce a red reaction product19. Whole-mount in situ hybridization was performed as previously described19. Double-staining to examine both retinoic-acid activity (RARE–lacZ) and mRNA localization was performed by first staining for 1 h for β-galactosidase activity, followed by processing for whole-mount in situ hybridization as previously reported19. Stained embryos were embedded in 3% agarose and sectioned at 20–50 μm with a vibratome.
Acknowledgments
We thank the following for mouse cDNAs used to prepare in situ hybridization probes: G. Martin (Fgf8), Y. Saga (Mesp2), E. DeRobertis (Papc), A. Kispert (Tbx18), V. Giguere (Crabp2), M. Tsai (COUP-TFII) and A. Mansouri (Uncx4.1). We also thank J. Rossant for providing RARE–lacZ mice, and A. Molotkov, N. Molotkova and X. Zhao for helpful discussions. This work was funded by National Institutes of Health grant GM62848.
Footnotes
COMPETING FINANCIAL INTERESTS
The authors declare that they have no competing financial interests.
Published online at http://www.nature.com/naturecellbiology
Reprints and permissions information is available online at http://npg.nature.com/reprintsandpermissions/
1. Pourquié O. The segmentation clock: converting embryonic time into spatial pattern. Science. 2003;301:328–330. [PubMed]
2. Del Corral RD, et al. Opposing FGF and retinoid pathways control ventral neural pattern, neuronal differentiation, and segmentation during body axis extension. Neuron. 2003;40:65–79. [PubMed]
3. Moreno TA, Kintner C. Regulation of segmental patterning by retinoic acid signaling during Xenopus somitogenesis. Dev Cell. 2004;6:205–218. [PubMed]
4. Vermot J, et al. Retinoic acid controls the bilateral symmetry of somite formation in the mouse embryo. Science. 2005;308:563–566. [PubMed]
5. Vermot J, Pourquié O. Retinoic acid coordinates somitogenesis and left-right patterning in vertebrate embryos. Nature. 2005;435:215–220. [PubMed]
6. Kawakami Y, et al. Retinoic acid signalling links left-right asymmetric patterning and bilaterally symmetric somitogenesis in the zebrafish embryo. Nature. 2005;435:165–171. [PubMed]
7. Dubrulle J, McGrew MJ, Pourquié O. FGF signaling controls somite boundary position and regulates segmentation clock control of spatiotemporal Hox gene activation. Cell. 2001;106:219–232. [PubMed]
8. Aulehla A, Herrmann BG. Segmentation in vertebrates: clock and gradient finally joined. Genes Dev. 2004;18:2060–2067. [PubMed]
9. Molotkova N, Molotkov A, Sirbu IO, Duester G. Requirement of mesodermal retinoic acid generated by Raldh2 for posterior neural transformation. Mech Dev. 2005;122:145–155. [PMC free article] [PubMed]
10. Niederreither K, et al. Embryonic retinoic acid synthesis is essential for heart morphogenesis in the mouse. Development. 2001;128:1019–1031. [PubMed]
11. Molotkov A, et al. Stimulation of retinoic acid production and growth by ubiquitously-expressed alcohol dehydrogenase. Adh3 Proc Natl Acad Sci USA. 2002;99:5337–5342. [PubMed]
12. Niederreither K, Subbarayan V, Dollé P, Chambon P. Embryonic retinoic acid synthesis is essential for early mouse post-implantation development. Nature Genet. 1999;21:444–448. [PubMed]
13. Mic FA, Haselbeck RJ, Cuenca AE, Duester G. Novel retinoic acid generating activities in the neural tube and heart identified by conditional rescue of Raldh2 null mutant mice. Development. 2002;129:2271–2282. [PMC free article] [PubMed]
14. Mic FA, Molotkov A, Benbrook DM, Duester G. Retinoid activation of retinoic acid receptor but not retinoid X receptor is sufficient to rescue lethal defect in retinoic acid synthesis. Proc Natl Acad Sci USA. 2003;100:7135–7140. [PubMed]
15. Rossant J, et al. Expression of a retinoic acid response element-hsplacZ transgene defines specific domains of transcriptional activity during mouse embryogenesis. Genes Dev. 1991;5:1333–1344. [PubMed]
16. Tanaka Y, Okada Y, Hirokawa N. FGF-induced vesicular release of Sonic hedgehog and retinoic acid in leftward nodal flow is critical for left–right determination. Nature. 2005;435:172–177. [PubMed]
17. Morimoto M, Takahashi Y, Endo M, Saga Y. The Mesp2 transcription factor establishes segmental borders by suppressing Notch activity. Nature. 2005;435:354–359. [PubMed]
18. Bussen M, et al. The T-box transcription factor Tbx18 maintains the separation of anterior and posterior somite compartments. Genes Dev. 2004;18:1209–1221. [PubMed]
19. Sirbu IO, Gresh L, Barra J, Duester G. Shifting boundaries of retinoic acid activity control hindbrain segmental gene expression. Development. 2005;132:2611–2622. [PMC free article] [PubMed]
20. MacLean G, et al. Cloning of a novel retinoic-acid metabolizing cytochrome P450, Cyp26B1, and comparative expression analysis with Cyp26A1 during early murine development. Mech Dev. 2001;107:195–201. [PubMed]
21. Noy N. Retinoid-binding proteins: mediators of retinoid action. Biochem J. 2000;348:481–495. [PubMed]
22. Ruberte E, Dolle P, Chambon P, Morriss–Kay G. Retinoic acid receptors and cellular retinoid binding proteins. II. Their differential pattern of transcription during early morphogenesis in mouse embryos. Development. 1991;111:45–60. [PubMed]
23. Qiu Y, Tsai SY, Tsai M-JCOUP-TF. An orphan member of the steroid–thyroid hormone receptor superfamily. Trends Endocrinol Metab. 1994;5:234–239. [PubMed]
24. Perantoni AO, et al. Inactivation of FGF8 in early mesoderm reveals an essential role in kidney development. Development. 2005;132:3859–3871. [PubMed]
25. Meyers EN, Martin GR. Differences in left–right axis pathways in mouse and chick: functions of FGF8 and SHH. Science. 1999;285:403–406. [PubMed]
26. Molotkov A, Molotkova N, Duester G. Retinoic acid generated by Raldh2 in mesoderm is required for mouse dorsal endodermal pancreas development. Dev Dyn. 2005;232:950–957. [PubMed]
27. Mansouri A, et al. Paired-related murine homeobox gene expressed in the developing sclerotome, kidney, and nervous system. Dev Dyn. 1997;210:53–65. [PubMed]
28. Kim SH, et al. The role of paraxial protocadherin in selective adhesion and cell movements of the mesoderm during Xenopus gastrulation. Development. 1998;125:4681–4691. [PubMed]