Annie Rowe

EXPRESSION OF THE CHICKEN RETINOID X RECEPTOR-GAMMA GENE IN MIGRATING CRANIAL NEURAL CREST CELLS

We have used in situ hybridisation to whole chick embryos with digoxygenin-labelled probes to investigate the distribution of RXR-γ transcripts during neural crest cell migration in the developing head and the anterior of the trunk (the vagal region), where neural crest cells make a substantial contribution. We have found that RXR-γ transcripts are a good marker for migrating neural crest cells in the chick embryo. RXR-γ transcripts were first detected in cells that had recently emerged from the neural crest, providing an earlier marker for neural crest cells than the HNK-1 epitope. The pattern of RXR-γ transcript distribution is dynamic in the developing chick head, and changes in a pattern which is coincident with the migration of cells containing RXR-γ transcripts and the gradual restriction of RXR-γ transcripts to specific differentiating neural crest derivatives. Transcripts appeared to be present initially in migrating neural crest cells thoughout the developing head, but gradually became restricted to some crest-derived populations and absent from others. By stage 15, RXR-γ transcripts were not detectable in neural-crest-derived ectomesenchymal cells, although they were still found in cells contributing to the cranial ganglia and their roots.

The chicken retinoid-X-receptor-? (RXR-?) gene and its expression in the developing limb

Antero-posterior (a-p) patterning of the vertebrate limb bud is controlled by signals from the polarizing region, a group of cells in the posterior mesenchyme of the bud. Application of retinoic acid to the anterior margin of the chick wing bud induces polarizing region activity in anterior mesenchyme cells, resulting in digit duplications. Retinoic acid acts by binding to nuclear retinoid receptors, and so regulating expression of target genes. Retinoid receptors of the RXR class are essential for this activity. We have previously described a chicken RXR-γ cDNA clone (Rowe et al. 1991a). In this paper we report the isolation and characterization of a chicken RXR-α cDNA clone, and show by northern blotting that an RXR-α mRNA of approximately 5 kilobases is present in a range of tissues in embryonic and adult chickens. In situ hybridization experiments showed that RXR-α transcripts were present throughout the epithelium and mesenchyme of the chick wing bud at stages when retinoic acid can affect a-p patterning. In contrast, RXR-γ transcripts were undetectable in these cells, being restricted to peripheral nervous tissue in the bud. These data suggest that RXR-α, but not RXR-γ, could mediate the effects of locally applied retinoic acid on a-p patterning in the chick wing bud.

Position-dependence of retinoic acid receptor-β gene expression in the chick limb bud

Retinoic acid and 3,4-didehydroretinoic acid are metabolites of vitamin A that can induce duplications and other malformations when locally applied to the anterior margin of the chick limb bud. There is evidence that they may be natural signaling substances in the limb bud. Both compounds are thought to act by binding to ligand-dependent transcription factors that belong to the steroid/thyroid hormone nuclear receptor superfamily. In situ hybridization analyses show that in the mesenchyme of the chick wing bud between embryonic stages 20 and 27, retinoic acid receptor-beta (RAR-beta) transcripts are restricted to the proximal region of the bud and are present at highest levels in the region of the limb bud mesenchyme that contributes to the shoulder. We have performed grafting experiments in order to examine whether RAR-beta gene expression in limb bud mesenchyme cells is cell-autonomous or whether it is dependent upon the cells position within the limb bud. When tissue from the proximal region of the stage 22 wing bud, which contains high levels of RAR-beta transcripts, was grafted to the distal tip of the bud, RAR-beta transcripts were undetectable in the graft 6 hr later. When tissue from the distal tip of the bud was grafted to a proximal site, most of the grafts exhibited a slight increase in the level of RAR-beta transcripts, which was detectable 6 hr after grafting. However, the levels of RAR-beta transcripts in these grafts never approached those found in the proximal core of the bud. These data indicate that RAR-beta gene expression in the chick wing bud is position-dependent in that it is repressed at the distal tip of the bud and partially activated by grafting distal tissue to a proximal site. However, accumulation of RAR-beta transcripts to high levels appears to be a characteristic of mesenchyme that was initially specified to form proximal structures.

A member of the chicken RXR family of nuclear receptors activates transcription in response to retinoic acid

The chicken cRXR nuclear receptor is a member of the steroid/thyroid hormone receptor superfamily. In this paper we show that cRXR can transactivate gene expression in response to retinoic acid, but that its sensitivity to retinoic acid is lower than that of the hRAR‐β receptor. We have also compared the ability of cRXR and hRAR‐β to respond to a panel of other retinoids. Unlike hRAR‐β, cRXR failed to respond to the naturally occurring retinoid 3,4‐didehydro‐retinoic acid or to the synthetic retinoid TTNPB, both of which share the ability of retinoic acid to induce digit duplications when locally applied to chick limb buds.

Differential expression of RAR-ß and RXR-γ transcripts in cultured cranial neural crest cells

In situ hybridization reveals that RAR-β and RXR-γ genes in mesencephalic neural crest cultures are independently regulated. RAR-β transcripts are found in all cells, with a slight increase in signal/cell with time in culture. In contrast, the distribution of RXR-γ transcript is initially uniform but becomes increasingly heterogeneous, so that after 72 h in culture, a significant proportion of cells lack transcripts while a small subpopulation contains very high levels of message. These differences in the behaviour of the RARβ and RXRγ genes in vitro can be related to differences in their expression patterns in vivo.

Expression Pattern of an RXR Nuclear Receptor Gene in the Chick Embryo

Retinoic Acid (RA) affects differentiation and morphogenesis in various developmental systems, including the vertebrate limb (Brickell and Tickle, 1989; Brookes, 1989), and acts by binding to nuclear RA receptors which are members of the steroid/thyroid hormone family of ligand-binding transcription factors (Green and Chambon, 1988; de The et al, 1990). Three closely-related RA receptors (RAR-alpha, RAR-beta and RAR-gamma), encoded by distinct genes and with distinct patterns of expression (Ruberte et al, 1990), have been identified in humans and mice, and a fourth RA receptor (hRXR-alpha) was discovered recently in humans (Mangelsdorf et al, 1990). Whilst still a member of the steroid/thyroid hormone receptor family, hRXR-alpha shares no significant homology with the RA-binding domain of RAR-alpha, RAR-beta or RAR-gamma and appears to represent an evolutionarily distinct RA response pathway (Mangelsdorf et al, 1990). We have isolated a cDNA clone encoding a chicken homologue of hRXR-alpha (cRXR) and have used this to examine the distribution of cRXR transcripts in chick embryos.

Retinoic Acid Treatment Alters the Pattern of Retinoic Acid Receptor Beta Expression in the Embryonic Chick Limb

Local application of retinoic acid (RA) to the anterior margin of the chick wing bud can affect patterning in both the developing limb and face, and there is evidence that RA may be a natural signalling substance in the limb (reviewed in Eichele, 1989; Brickell and Tickle, 1990, and Wedden et al, 1988). In the limb, local application of RA produces a mirror image duplication of structures across its anterior-posterior (A-P) axis. A second effect of RA application to the bud is inhibition of outgrowth of the upper beak primordium (the frontonasal mass), resulting in clefting of the primary palate.