Are transposition events at the origin of the bilaterian Hox complexes

jean s. deutsch and philippe lopez

The genome sequences of two non-bilaterian animals, the cnidarians NematosteUa vectensis and Hydra magnipapillata, have been recently completed. These new data lead to the fascinating result that the complement of Hox genes in the cnidarian ancestor is considerably lower than that in the bilaterians, although the complexity of their genome is otherwise similar (Technau et al. 2005). Thus, there is a correlation between the radiation of the Bilateria and the expansion of the Hox complex.

In the first part of this chapter, we shall present and discuss these data. In the second part, we shall present a novel hypothesis accounting for this phenomenon. In short, we surmise that the expansion of the Hox complex at the base of the Bilateria was due to a series of transposition events. Indeed, we hypothesise that the Hox genes themselves originate from transposons. The main support for this hypothesis is provided by the similarity between the homeodomain and the DNA-binding domain of bacterial integrases and eukaryotic transposases. We also examine some very precise rearrangements of the Hox complex in the Drosophilidae lineage. In the third part, we propose a scenario for the evolution of the Hox complex from the basic complement of Hox genes in the common ancestor of cnidarian and bilaterian animals. This scenario, based on our transposition hypothesis, accounts for several properties of the extant Hox genes.

to set the scene: the hox explosion

The homeobox is a conserved motif found in a huge variety of eukaryotic genes, encoding a DNA-binding domain. Although

Evolving Pathways: Key Themes in Evolutionary Developmental Biology, ed. Alessandro Minelli and Giuseppe Fusco. Published by Cambridge University Press. # Cambridge University Press 2008.

homeobox-containing genes are known from various branches of the eukaryote tree, such as fungi and plants, ANTP-class genes, to which Hox and paraHox genes belong, form a monophyletic group only known from animals (Bharathan et al. 1997, Holland and Takahashi 2005).

The Metazoa (i.e. Animalia) comprise non-bilaterian and bilaterally symmetric animals. Four extant non-bilaterian phyla are known: Porifera (sponges), Ctenophora (comb jellies), Cnidaria (corals, sea anemones, jellyfish and their kin) and Placozoa. There are no nerve cells in sponges, so that the presence of well-characterised nerve cells is a clear synapomorphy unifying Ctenophora and Cnidaria with the Bila-teria, in a clade called 'Eumetazoa'. Placozoa is a problematic phylum, comprising a single species, Trichoplax adhaerens, with a very simple morphology. Schierwater (2005) strongly advocates for its origin from the base of the Metazoa, before the emergence of the Porifera. However, it shares certain synapomorphies, such as belt desmosomes and neuropeptides, with Eumetazoa (Schuchert 1993). We thus think that it is a derived, secondarily simplified, eumetazoan.

The so-called 'new phylogeny' dispatched the bilaterian phyla into three 'super-phyla', Deuterostomia, Lophotrochozoa and Ecdysozoa (Aguinaldo et al. 1997). Two phyla may not fit in this classification: the Chaetognatha and the Acoelomorpha. Chaetognatha have been long considered as 'incertae sedis' (Ball and Miller 2006). They are now either regarded as a sister group to all Protostomes (Marletaz et al. 2006) or included within the Lophotrochozoa (Matus et al. 2006). Acoelomorpha, a phylum recently created following the exclusion of the Acoela and the Nemertodermatida from the phylum Platyhelminthes, is assumed to stem from the very base of the Bilateria, before the split into the three super-phyla (Baguna and Riutort 2004). For both these problematic taxa, Chaetognatha and Acoelomorpha, more data are needed, on a larger panel of genes and taxa, until a firm phylogenetic conclusion can be drawn.

Almost all bilaterian phyla suddenly appear in the fossil record at the base of the Cambrian within a short length of time, c. 540 to 550 million years (Myr) before present. This was called the 'Cambrian explosion'. In parallel, there was a sudden increase in the number of Hox genes.

Indeed, there is no Hox gene in sponges (Manuel and Le Parco 2000). As for ctenophores, a single small fragment of a putative Hox gene isolated from Beroe ovata has been withdrawn from the data banks as a contaminant at the request of the authors. In contrast, Hox and Hox-related genes have been studied from several cnidarian species (Gauchat et al. 2000, Chourrout et al. 2006, Kamm et al. 2006, Ryan et al. 2006). Summarising, we can draw the following conclusions: (1) the number of Hox genes in the repertoire of the ancestral cnidarian is low, not more than two or three; (2) these primitive Hox genes are more similar to the Hox1-2, Hox3 and maybe the posterior Hox9-14 of so-called 'paralogy groups' (PG) of bilaterian Hox genes; (3) lineage-specific duplications have increased this primitive number in several cnidarian taxa; (4) the cnidarian Hox complex, if it ever existed primitively, has been disrupted and reorganised during the evolution of the Cnidaria.

Since the work of de Rosa et al. (1999), who examined the repertoire of Hox genes in a diversity of bilaterian taxa, a number of data have been added, all supporting the main conclusion of this pioneer work: on a qualitative and quantitative basis (type and number of Hox genes), the repertoire of Hox genes supports the classification of the Bilateria into the three 'super-phyla' first proposed by Aguinaldo et al. (1997). These data now allow confident conclusions to be drawn about what the complement of Hox genes was in the ancestor of several phyla and, in the best cases, what their genomic organisation was.

In the Arthropoda, the Hox genes' basic complement comprises 10 genes, the two sister genes lab and pb, orthologous to the paralogy groups (PG) PG1 and PG2, respectively; a single zen gene orthologous to PG3; six genes belonging to the 'median' group PG4-8, namely Dfd, Scr, ftz, Antp, Ubx and abdA, and a single 'posterior' gene, AbdB, corresponding to PG9-14 in deuterostomes. From all available genomic data, we can infer that these ten Hox genes were primitively grouped in a single cluster, despite some breaks and rearrangements that occurred during the evolution of long germ-band insects.

Among the Ecdysozoa, the Onychophora, a phylum closely related to the Arthropoda, and the Priapulida fit the arthropod scheme (de Rosa et al. 1999), with the possible exception of a duplication of the posterior gene. In contrast, in various nematodes the Hox complement appears quite reduced and derived: some genes are missing, others are derived, mosaic or duplicated, and the Hox complex is profoundly rearranged and disrupted (Aboobaker and Blaxter 2003). This seems specific to the Nematoda, since a species belonging to the Nematomorpha (probably the closest relatives to the nematodes) has a full complement of arthropod-like Hox genes. We can thus infer that the Hox complement present in the ecdy-sozoan ancestor comprised ten different genes, or at least nine if the duplication leading to the sister genes Ubx/abdA were specific to the Arthropoda. They were most probably arranged in a single complex.

As for the Lophotrochozoa, data are less complete than those for the Ecdysozoa, both in terms of the number of taxa studied and the structure of the genes and complexes. Lophotrochozoan species possess clear orthologues of the PG1/lab, PG2/pb, and PG3/zen. As 'median' genes, they share PG4/Dfd and PG5/Scr. Telford (2000) hypothesised an orthology relationship between the lophotrochozoan median gene Lox5, the arthropodftz and PG6. They also possess clear orthologues of Antp (possibly a member of PG7). Lox2 and Lox4 are two sister genes, arising from a different duplication from the one that generated Ubx and abdA in the arthropod lineage (Wong et al. 1995). This amounts to six 'median' Hox genes. In addition, lophotrochozoans have two specific 'posterior' genes, Post1 and Post2 (de Rosa et al 1999). The two latter genes, together with the couple Lox2/Lox4, constitute characteristic signatures of the lophotrochozoan lineage. Platyhelminths show a disturbed panel of Hox genes with derived and duplicated genes. In the parasite platyhelminth Schistosoma mansoni, the Hox complex is disintegrated and dispersed in the genome (Pierce et al. 2005). We can derive a figure of 11 Hox genes as the complement of Hox genes in the primitive lophotrochozoan. Whether they were clustered in a complex is still an open issue.

The Deuterostomia includes two branches: the Ambulacraria, uniting the Echinodermata and the Hemichordata, and the Chordata, comprising the Cephalochordata, the Urochordata and the Vertebrata. The vertebrates have undergone several whole genome duplications during their evolution, leading to up to four paralogous Hox clusters in the Tetrapoda and (primitively) up to eight in the Teleostei. From sequence comparisons between the four Hox clusters in mammals, a primitive complex of 13 Hox genes was derived (McGinnis and Krumlauf 1992). The discovery of a 14th Hox gene in the cephalochordate Bran-chiostoma floridae (Ferrier et al. 2000), in the coelacanth Latimeria chalum-nae and in the shark Heterodontus francisci (Powers and Amemiya 2004) added one more posterior gene to the ancestral chordate Hox complex. The grouping in a single cluster of the Hox genes in the amphioxus and the tight clustering of the Hox genes in vertebrates led to the hypothesis of a single complex of 14 Hox genes in the chordate ancestor. In the Urochordata, losses and rearrangements yielded a disorganised and derived Hox cluster, variable among taxa.

Summing up data from a number of species belonging to diverse classes among the hemichordates and the echinoderms, a Hox complex orthologous to that of the chordates can be derived in the common ancestor of the Ambulacraria, possibly with a smaller number of posterior genes. The Hox cluster of the sea urchin Strongylo-centrotus purpuratus is profoundly perturbed, maybe in line with the huge modification of the echinoderm body plan (Cameron et al. 2006). In total, a single cluster of 14 genes comprising at least six posterior genes can reliably be postulated to have been present in the deuterostome ancestor (Monteiro and Ferrier 2006).

In both Chaetognatha and Acoelomorpha, a reduced number of Hox genes have been reported, with some of them showing no clear orthology with known Hox genes from other bilaterian taxa (Papillon et al. 2003, Cook et al. 2004). This has been attributed to the 'primitive' nature of these Hox genes and phyla. However, 'mosaic' homeodomain sequences could result from divergent evolution after loss of some Hox genes as well. This kind of evolution of remaining Hox genes is exemplified in the case of echinoderms, such as Hox4 and Hox5 from starfish compared with Hox genes from echinoid species that have lost one median gene (Long et al. 2003).

The most parsimonious figure for the number of Hox genes present in the common ancestor of the three bilaterian branches is nine genes: two anterior genes, orthologous to PG1 and PG2, one anterior-median (PG3), five median (PG4 to PG8) and one posterior gene, to which we refer in the following as PG9*, being the ancestor of PG9 to PG14. Hence, the number of Hox genes suddenly jumped from two to three genes as present in the common ancestor of the Cni-daria and the Bilateria (Ferrier and Holland 2001, Garcia-Fernandez 2005, Chourrout et al. 2006) to nine in the common ancestor of extant bilaterians, with a further increase to 10 in the Ecdysozoa (duplication of PG8 to Ubx and abdA), to 11 in the Lophotrochozoa (duplication of PG8 to Lox2 and Lox4 and of PG9* to Post1 and Post2) and to 14 in the deuterostome ancestor (duplications of PG9* in PG9 to PG14). This sudden increase is what we call 'the Hox explosion' that paralleled the radiation of the Bilateria.

This observation needs an explanation. Here we hypothesise that the 'Hox explosion' is due to a burst of transposition events and that the Hox genes themselves are primitive transposons that have been 'domesticated' during further evolution of the metazoan genome.

the homeo domain protein as a transposase

We shall now present the first piece of evidence supporting the transposition mechanism that, as we suggest, has operated at the origin of the bilaterian Hox complex. We review the current literature presenting evidence that the RAG1 gene, involved in the recombination events leading to the diversity of the vertebrates' immune response, is derived from a transposon. We show that metazoan homeodomains are very similar to the DNA-binding domain of the RAG1 protein, similarity being the greatest for Hox homeodomains. Hence we surmise that the Hox genes are also issued from a transfer of DNA by a transposon of a similar kind.

Schatz et al. (1989) discovered the RAG1 protein as a main player for V(D)J recombination of immunoglobin and receptor genes. RAG1 interacts with its partner RAG2, encoded by a neighbouring gene, and with HMG proteins in a multimeric complex. In this complex, both the DNA-binding domains and the critical DDE acidic residues active in recombination are located within the RAG1 moiety (De and Rodgers 2004).

Thompson (1995) first suggested that the RAG locus has evolved from a transposase. Spanopolou et al. (1996) discussed the parallel between V(D)J and bacterial recombination. Bernstein et al. (1996) underlined the structural similarity between RAG1 and bacterial integrases. This 'transposon hypothesis' on the origin of the V(D)J system has gained support from evidence provided by Hiom et al. (1998) and Reddy et al. (2006) that the RAG proteins are able to generate transpositions in vitro and in vivo. Last but not least, Kapitonov and Jurka (2005) revealed sequence similarity between the RAG1 protein and the transposase and between the V(D)J recombination signals and the target of a new DNA transposon, called 'Transib'.

We aligned the DNA nonamer-binding domain of RAG1, highly conserved throughout gnathostome evolution, to a sample of homeodo-mains from metazoan proteins representative of the diversity of this group of transcription factors. A part of this alignment is shown in Figure 13.1, with homeodomains of Hox genes from Drosophila melanoga-ster (fly) as an ecdysozoan representative, Nereis virens (polychaete worm) for lophotrochozoans and Mus musculus (mouse) for deuterostomes. So-called 'posterior genes', i.e. AbdB, Post1, Post2 and Hox9 to Hox13, have been discarded because of rapid evolutionary rate (Chourrout et al. 2006). For the same reason, Drosophila genes corresponding to Hox3 (i.e. zen, zen2, bicoid) as well as ftz have not been taken into account.

Cle Dre Pwa RAG1 Gga Mmu Hsa

Was this article helpful?

0 0

Responses

  • Ciriaca
    Are transposition events at the origin of the bilaterian Hoxcomplexes?
    8 years ago

Post a comment