The dorsal blastopore lip of the early Xenopus laevis gastrula can organize a complete secondary body axis when transplanted to another embryo. A search for potential gene regulatory components specifically expressed in the organizer was undertaken that resulted in the identification of four types of complementary DNAs from homeobox-containing genes that fulfill this criterion. The most abundant of these encodes a DNA-binding specificity similar to that of the Drosophila melanogaster anterior morphogen bicoid. The other three are also homologous to developmentally significant Drosophila genes. These four genes may participate in the regulation of the developmental potential of the organizer.

What is the nature of positional information during embryogenesis? By using Xenopus homeo box genes as anteroposterior (A-P) markers, we confirm the findings of others that mesoderm-inducing growth factors and retinoic acid (RA) can provide positional information along the axis of the body. Xenopus tissue culturemesoderm-inducing factor (XTC-MIF) selectively activates an anteriorly expressed homeo box gene {XlHbox 1), while basic fibroblast growth factor (bFGF) activates selectively a posteriorly expressed homeo box gene {XlHbox 6). RA activates expression of the posterior gene XlHbox 6, but not of XlHbox 1. This activation, however, requires exposure to growth factors. The data suggest that growth factors and RA may cooperate with each other to provide positional information in vertebrates.

Xenopus XlHbox 1 produces two transcripts during early development. One encodes a long open reading frame (ORF) and the other a short ORF sharing the same homeodomain, but differing by an 82 amino acid domain at the amino terminus. The long protein amino terminus is conserved with many other homeodomain proteins, and its absence from the short protein could have functional consequences. Some viral genes also utilize a single ORF to encode transcription factors of antagonistic functions. The overall organization of the homologous genes in frog and man is similar, supporting the notion that both transcripts are of functional significance. Studies on XlHbox 1 function show that the region common to the long and short proteins has a sequence‐specific DNA‐binding activity, and that microinjection of specific antibodies into embryos results in the loss of structures derived from cells normally expressing XlHbox 1.

A cDNA clone (p52) that contains all the protein‐coding region from the maternally expressed XlHbox 2 locus of the frog Xenopus laevis has been isolated and sequenced. A probe containing the exon preceding the homeobox detected transcripts which arise from a splicing event in which the homeobox‐containing exon is replaced by another exon lying 5′ to it in the genome. Both the homeobox‐containing and homeobox‐less splicing event occur in the same tissues, with the homeobox‐less RNA representing the minority of mRNA from this gene. There may therefore be a function for two types of transcript, and hence protein, from this locus. This phenomenon may not be exclusive to the XlHbox 2 gene of Xenopus, but might occur more generally in other homeobox‐containing genes. The protein deduced from the homeobox‐containing cDNA is significantly similar to the yeast mating type factor a1 (MAT‐a1) gene product. In addition to the previously described homology of the homeodomains, the amino‐terminal domains of XlHbox 2 and MAT‐a1 are similar to each other; thus essentially all of the MAT‐a1 protein corresponds to some part of the XlHbox 2 protein. In the case of XlHbox 2, the protein coded for by the homeoboxless mRNA would contain all of the non‐homeobox homology to yeast MAT‐a1.

We review here old experiments that defined the existence of morphogenetic gradient fields in vertebrate embryos. The rather abstract idea of cell fields of organ-forming potential has become less popular among modern developmental and molecular biologists. Results obtained with antibodies directed against homeodomain proteins suggest that gradient fields may indeed be visualized at the level of individual regulatory molecules in vertebrate embryos.