Identification of a novel isoform of the retinoic acid receptor gamma expressed in the mouse embryo.

Retinoic acid is known to have profound effects on developmental processes. It has been implicated as a putative morphogen in the developing chick limb bud and regenerating amphibian limb blastema and has been demonstrated to have powerful teratogenic effects in mammals, including humans. Recently, three specific retinoic acid receptors (RARs), RAR alpha, -beta, and -gamma, were identified and shown to be members of the steroid receptor superfamily. We report the identification of a novel RAR gamma isoform, mRAR gamma B, which differs from the previously described mouse RAR gamma at its amino terminus. In addition, we show that both RAR gamma isoforms are expressed maximally at midgestation in structures known to be affected adversely by retinoic acid administration to pregnant mice. Multiple RAR isoforms, each of which may play a unique or combinatorial role as a regulator of mammalian development, are thus expressed in the mouse embryo.     

Cloning of cDNAs encoding retinoic acid receptors RAR gamma 1, RAR gamma 2, and a new splicing variant, RAR gamma 3, from Aambystoma mexicanum and characterization of their expression during early development

To analyze retinoic acid (RA) receptor (RAR) expression during early development in the urodele embryo, we have isolated cDNAs for four members of the axolotl (Ambystoma mexicanum) RAR family, namely RAR alpha (NR1B1), aRAR gamma 1 (NR1B3a), aRAR gamma 2 (NR1B3b), and a new splicing variant of aRAR gamma 2, aRAR gamma 3 (NR1B3c), which contains an insertion of five hydrophobic amino acids in the C-terminal region of the DNA binding domain. The temporal expression pattern of the RAR gamma isoforms was established by RT-PCR using total RNA from embryos of different stages. The expression of aRAR gamma 2 coincides with neurulation and is enhanced in the extremities of the embryo's anteroposterior axis. The aRAR gamma 3 is specifically expressed during gastrulation and early neurulation, whereas aRAR gamma 1 is expressed later during organogenesis. Global aRAR gamma 2 mRNA levels, as well as their spatio-temporal expression pattern in the neurula, were not affected by treatment with RA. These results show that several RARs are expressed in the axolotl embryo during early development, and reveal the existence of a new RAR gamma variant.     

Altered expression of retinoic acid (RA) receptor mRNAs in the fetal mouse secondary palate by all-trans and 13-cis RAs: implications for RA-induced teratogenesis

Retinoic acid (RA) is mandatory for various biological processes and normal embryonic development but is teratogenic at high concentrations. In rodents, one of the major malformations induced by RA is cleft palate (CP). RA mediates its effects by RA receptors (RARs), but the expression patterns of RARs in the developing palate are still unclear. We investigated the normal expression of RAR alpha, beta, and gamma messenger RNAs (mRNAs) in the fetal mouse secondary palate and the effects of all-trans and 13-cis RAs on the expression of RAR mRNAs by Northern blot analysis. RAR alpha (2.8, 3.8 kb), RAR beta (3.3 kb), and RAR gamma (3.7 kb) mRNAs were detected in the fetal palate on gestational days (GD) 12.5-14.5. The expression of RAR alpha and gamma mRNAs did not show apparent sequential changes, but that of RAR beta mRNA increased at GD 13.5. Treatment of pregnant mice with 100 mg/kg all-trans RA induced CP in 94% of the fetuses and elevated the levels of RAR beta and gamma mRNAs in the fetal palate. The up-regulation of RAR beta mRNA by all-trans RA was more marked than that of RAR gamma mRNA. Treatment with 100 mg/kg 13-cis RA induced CP in only 19% of the fetuses. Although 13-cis RA elevated the RAR beta and gamma mRNA levels in fetal palates, its up-regulation was slower and less marked than that induced by all-trans RA. These findings indicate that the induction of RAR beta mRNA in the fetal palate correlates well with the tissue concentration of all-trans RA after RA treatment, and RAR beta may be one of the most influential candidate molecules for RA-induced teratogenesis.     

Phosphorylation of activation functions AF-1 and AF-2 of RAR alpha and RAR gamma is indispensable for differentiation of F9 cells upon retinoic acid and cAMP treatment.

The role of RAR alpha 1 and RAR gamma 2 AF-1 and AF-2 activation functions and of their phosphorylation was investigated during RA-induced primitive and parietal differentiation of F9 cells. We found that: (i) primitive endodermal differentiation requires RAR gamma 2, whereas parietal endodermal differentiation requires both RAR gamma 2 and RAR alpha 1, and in all cases AF-1 and AF-2 must synergize; (ii) primitive endodermal differentiation requires the proline-directed kinase site of RAR gamma 2-AF-1, whereas parietal endodermal differentiation additionally requires that of RAR alpha 1-AF-1; (iii) the cAMP-induced parietal endodermal differentiation also requires the protein kinase A site of RAR alpha-AF-2, but not that of RAR gamma; and (iv) the AF-1-AF-2 synergism and AF-1 phosphorylation site requirements for RA-responsive gene induction are promoter context-dependent. Thus, AF-1 and AF-2 of distinct RARs exert specific cellular and molecular functions in a cell-autonomous system mimicking physiological situations, and their phosphorylation by kinases belonging to two main signalling pathways is required to enable RARs to transduce the RA signal during F9 cell differentiation.     

Cyclic AMP analogs and retinoic acid influence the expression of retinoic acid receptor alpha, beta, and gamma mRNAs in F9 teratocarcinoma cells.

Retinoic acid (RA) receptor alpha (RAR alpha) and RAR gamma steady-state mRNA levels remained relatively constant over time after the addition of RA to F9 teratocarcinoma stem cells. In contrast, the steady-state RAR beta mRNA level started to increase within 12 h after the addition of RA and reached a 20-fold-higher level by 48 h. This RA-associated RAR beta mRNA increase was not prevented by protein synthesis inhibitors but was prevented by the addition of cyclic AMP analogs. In the presence of RA, cyclic AMP analogs also greatly reduced the RAR alpha and RAR gamma mRNA levels, even though cyclic AMP analogs alone did not alter these mRNA levels. The addition of either RA or RA plus cyclic AMP analogs did not result in changes in the three RAR mRNA half-lives. These results suggest that agents which elevate the internal cyclic AMP concentration may also affect the cellular response to RA by altering the expression of the RARs.     

Modulation by retinoids of mRNA levels for nuclear retinoic acid receptors in murine melanoma cells

Retinoids inhibit the growth and enhance the differentiation of murine S91-C2 melanoma cells. Specific alterations in gene expression are a plausible mechanism for these effects. Since nuclear retinoic acid receptors (RAR) are likely mediators of retinoid-induced changes in gene expression, we used Northern blotting to analyze the expression of RAR alpha, RAR beta, and RAR gamma in S91-C2 cells. mRNA for both RAR alpha and RAR gamma was detected in these cells, but no RAR beta mRNA could be found. Treatment with 10(-7) and 10(-6) M beta-all-trans-retinoic acid (RA) for 24 h caused a 1.5- to 2-fold increase in RAR alpha and RAR gamma mRNA, whereas lower concentrations of RA were ineffective. RAR beta mRNA, which was undetectable in untreated cells, was detected after 24 h of treatment with a RA concentration as low as 10(-9) M, and its level increased with up to 10(-6) M RA. At the latter dose, RAR beta mRNA induction occurred by 4 h and increased progressively, reaching a plateau after 24 h of treatment. RAR beta mRNA induction at 4 h was not inhibited by cycloheximide at a concentration that suppressed protein synthesis by more than 90%. Several retinoids and related synthetic compounds, including 13-cis RA, TTNPB, Ch55, Am80, and the trifluoromethyl nonyloxyphenyl analog of RA, also induced RAR beta mRNA, whereas a 24-h treatment with 10(-6) M retinol, TTNP (a decarboxylated analog of TTNPB), or the phenyl analog of RA failed to induce RAR beta mRNA. With the exception of retinol and the trifluoromethyl nonyloxyphenyl analog of RA, the ability of the retinoids to induce RAR beta mRNA and their growth inhibitory effect were correlated. However, S91-C154, a RA-resistant mutant subclone derived from S91-C2 cells, showed mRNA levels of RAR alpha and RAR gamma and induction of RAR beta by RA similar to those detected in the sensitive S91-C2 cells. Like the S91 melanoma cells, two other mouse melanoma cell lines, K-1735P and B16-F1, constitutively expressed RAR alpha and RAR gamma mRNAs. The level of RAR beta mRNA was increased by RA only in B16-F1 cells, although the growth of both was inhibited by RA. These results demonstrate that RA can, directly and rapidly, induce the expression of mRNA for a high affinity nuclear receptor in some murine melanoma cells and that this induction is not sufficient to inhibit growth.     

Expression of retinoic acid receptor genes in neural crest-derived cells during mouse facial development

Retinoic acid (RA) is known as a teratogen that induces abnormalities in facial structures which are made up mainly of neural crest-derived mesenchyme. We investigated expression patterns of RA receptor (RAR) genes (subtypes alpha, beta, gamma) during mouse facial development. The expression of the RAR beta gene is specific for the mesenchyme around developing eyes and nose, whereas the RAR gamma gene is expressed in the mesenchyme differentiating to facial cartilages and bones. In contrast, the RAR alpha gene is expressed weakly and uniformly over the facial region. These results suggest that crucial roles of endogenous RA in facial development depend on differential functions of the RAR subtypes.