Hox Genes in Vertebrate Development

Robb Krumlauf Laboratory of Developmental Neurobiology Medical Research Council National Institute for Medical Research The Ridgeway Mill Hill London NW7 1AA England Introduction This year marks the centenary anniversary of the work in which W. Bateson first described homeotic mutations (Bateson, 1894) and also marks a decade since the discov- ery of the homeobox motif in genes of the Drosophila ho- meotic (HOM-C) complexes (McGinnis et al., 1984; Scott and Weiner, 1984). These last 10 years represent a pivotal and extremely stimulating period in the study of animal development. New techniques and approaches have rap- idly evolved and enhanced our ability to search for and identify components that establish the basic body plan at the molecular, cellular, and genetic levels. Large numbers of genes involved in regulating basic mechanisms of em- bryogenesis and development have been characterized in many different species. As opposed to working in isola- tion in independent experimental systems, there has also been ageneral reemergence and appreciation of thevalue of comparative biology and evolution in the analysis of developmental problems. It is now almost routinely ex- pected that a regulatory gene or signaling molecule with a particular role in one vertebrate species will have a similar conserved role in other vertebrates. There is also a reason- able chance that such genes or molecules will have analo- gous roles in more diverse species, which can be useful in interpreting their roles in vertebrates. This ability to com- pare and extrapolate among systems has created an inter- active and interdisciplinary atmosphere that is helping to stimulate new areas of research. To a great extent, the analysis of the Drosophila genes containing a homeobox and the discovery of their vertebrate and invertebrate counterparts have provided a useful paradigm for studying the roles of other genes in embryogenesis. This review will discuss some of the advances, problems, and future directions of work on the function and regulation of the four vertebrate Hex gene clusters, which have come to represent one of the best examples of structural and func- tional conservation in animal species. Organization and Evolution of Hox/HOM-C Complexes The genomic structural and organizational similarities be- tween vertebrate Hox and Drosophila HOM-C complexes provided compelling evidence that the Hox/HOM-C clus- ters arose by duplication and divergence from a common ancestral cluster that was speculated to be conserved in an even broader range of animal species (reviewed by Akam, 1989; Scott et al., 1989; Kessel and Gruss, 1990; Boncinelli et al., 1991; Duboule, 1992; Krumlauf, 1992; McGinnis and Krumlauf, 1992; see also review by Kenyon, 1994 [this issue of Ce//)). Initially, there was some difficulty in defining the nature of a primordial complex(s), because several features of HOM-C are not shared with Hox clus- ters. For example, in Drosophila the HOM-C complex is comprised of two separate clusters, bithorax(BX-C; Lewis, 1978) and Antennapedia (An@) complex (Kaufman et al., 1980,1990), and also contains interspersed genes encod- ing an immunoglobulin-related protein (amalgam), a cuti- cle protein, the bicoid maternal-effect homeobox gene, and two zen (zl and 22) homeobox genes involved in dor- soventral patterning (Kaufman et al., 1990). However, the presence of a single Hox/HOM cluster in the red flour bee- tle Tribolium castaneum (Beeman et al., 1989), the nema- tode Caenorhabditis elegans (Kenyon and Wang, 1991; Salser and Kenyon, 1994), the crustacean Artemia (Averof and Akam, 1993), some primitive chordates (Pendle- ton et al., 1993; Holland et al., 1994) is most consistent with the concept of a single ancestral cluster, paradoxically a type more related to the vertebrate Hox clusters than Drosophila HOM-C, on which we have historically based so much of our understanding of homeotic complexes. While the data are still emerging from polymerase chain reaction-based screens and genomic cloning (Holland et al., 1992, 1994; Pendleton et 1993) it appears that the origins of multiple Hox clusters within a species is property associated with evolution of primitive verte- brates. In vertebrates, analysis of the mouse and human Hox complexes is the most detailed and indicates that there are 38 genes organized in four different chromosomal complexes approximately 120 kb in length and that the genes in each cluster are all oriented the same 5’ to 3’ direction of transcription (Kessel and Gruss, 1990; Du- boule, 1992; Krumlauf, McGinnis and 1992). To date, similar studies in all other vertebrates indi- cate an identical Hox gene organization, which has prompted the adoption of a common vertebrate nomencla- ture (Scott, 1992; Duboule, 1994). Figure 1 summarizes the organization and homology relationships between four Hox clusters and HOM-C. Alignments are made on the basis of multiple domains of sequence identity, in addi- tion to the homeodomain itself, and on the relative posi- tions of the genes within respective complexes. There are 13 different sets of genes with shared properties, termed paralogous groups. The strongest homologies to HOM-C members are labial (group l), proboscipedia (group 2) Deformed (group 4) Sex combs reduced (SW) (group 5) and AbdominalB (Abd-8) (groups 9-13). Group 3 has no Drosophila counterpart, and the comparable de- gree of similarity among An@, Ultrabithorax (Ubx), and abdA and members of groups 8-8 make it impossible to determine whether they are homologs or independent duplication events of an Antplike gene from the primordial cluster. The vertebrate clusters arose from an ancestral complex