The "fish-specific" Hox cluster duplication is coincident with the origin of teleosts

The Hox gene complement of zebrafish, medaka, and fugu differs from that of other gnathostome vertebrates. These fishes have seven to eight Hox clusters compared to the four Hox clusters described in sarcopterygians and shark. The clusters in different teleost lineages are orthologous, implying that a "fish-specific" Hox cluster duplication has occurred in the stem lineage leading to the most recent common ancestor of zebrafish and fugu. The timing of this event, however, is unknown. To address this question, we sequenced four Hox genes from taxa representing basal actinopterygian and teleost lineages and compared them to known sequences from shark, coelacanth, zebrafish, and other teleosts. The resulting gene genealogies suggest that the fish-specific Hox cluster duplication occurred coincident with the origin of crown group teleosts. In addition, we obtained evidence for an independent Hox cluster duplication in the sturgeon lineage (Acipenseriformes). Finally, results from HoxA11 suggest that duplicated Hox genes have experienced diversifying selection immediately after the duplication event. Taken together, these results support the notion that the duplicated Hox genes of teleosts were causally relevant to adaptive evolution during the initial teleost radiation.     

Genome Duplications and Accelerated Evolution of Hox Genes and Cluster Architecture in Teleost Fishes

The early origin of four vertebrate Hox gene clusters duringthe evolution of gnathostomes was likely caused by two consecutiveduplications of the entire genome and the subsequent loss ofindividual genes. The presumed conserved and important rolesof these genes in tetrapods during development led to the generalassumption that Hox cluster architecture had remained unchangedsince the last common ancestor of all jawed vertebrates. Butrecent data from teleost fishes reveals that this is not thecase. Here, we present an analysis of the evolution of vertebrateHox genes and clusters, with emphasis on the differences betweenthe Hox A clusters of fish (actinopterygian) and tetrapod (sarcopterygian)lineages. In contrast to the general conservation of genomicarchitecture and gene sequence observed in sarcopterygians,the evolutionary history of actinopterygian Hox clusters likelyincludes an additional (third) genome duplication that initiallyincreased the number of clusters from four to eight. We document,for the first time, higher rates of gene loss and gene sequenceevolution in the Hox genes of fishes compared to those of landvertebrates. These two observations might suggest that two differentmolecular evolutionary strategies exist in the two major vertebratelineages. Preliminary data from the African cichlid fish Oreochromisniloticus compared to those of the pufferfish and zebrafishreveal important differences in Hox cluster architecture amongfishes and, together with genetic mapping data from Medaka,indicate that the third genome duplication was not zebrafish-specific,but probably occurred early in the history of fishes. Each descendingfish lineage that has been characterized so far, distinctivelymodified its Hox cluster architecture through independent secondarylosses. This variation is related to the large body plan differencesobserved among fishes, such as the loss of entire sets of appendagesand ribs in some lineages.     

Revisiting the origin of the vertebrate Hox14 by including its relict sarcopterygian members

Bilaterian Hox genes play pivotal roles in the specification of positional identities along the anteroposterior axis. Particularly in vertebrates, their regulation is tightly coordinated by tandem arrays of genes [paralogy groups (PGs)] in four gene clusters (HoxA-D). Traditionally, the uninterrupted Hox cluster (Hox1-14) of the invertebrate chordate amphioxus was regarded as an archetype of the vertebrate Hox clusters. In contrast to Hox1-13 that are globally regulated by the "Hox code" and are often phylogenetically conserved, vertebrate Hox14 members were only recently revealed to be present in an African lungfish, a coelacanth, chondrichthyans and a lamprey, and decoupled from the Hox code. In this study we performed a PCR-based search of Hox14 members from diverse vertebrates, and identified one in the Australian lungfish, Neoceratodus forsteri. Based on a molecular phylogenetic analysis, this gene was designated NfHoxA14. Our real-time RT-PCR suggested its hindgut-associated expression, previously observed also in cloudy catshark HoxD14 and lamprey Hox14α. It is likely that this altered expression scheme was established before the Hox cluster quadruplication, probably at the base of extant vertebrates. To investigate the origin of vertebrate Hox14, by including this sarcopterygian Hox14 member, we performed focused phylogenetic analyses on its relationship with other vertebrate posterior Hox PGs (Hox9-13) as well as amphioxus posterior Hox genes. Our results confirmed the hypotheses previously proposed by other studies that vertebrate Hox14 does not have any amphioxus ortholog, and that none of 1-to-1 pairs of vertebrate and amphioxus posterior Hox genes, based on their relative location in the clusters, is orthologous.     

Hox clusters of the bichir (Actinopterygii, Polypterus senegalus) highlight unique patterns of sequence evolution in gnathostome phylogeny

Teleost fishes have extra Hox gene clusters owing to shared or lineage-specific genome duplication events in rayfinned fish (actinopterygian) phylogeny. Hence, extrapolating between genome function of teleosts and human or even between different fish species is difficult. We have sequenced and analyzed Hox gene clusters of the Senegal bichir (Polypterus senegalus), an extant representative of the most basal actinopterygian lineage. Bichir possesses four Hox gene clusters (A, B, C, D); phylogenetic analysis supports their orthology to the four Hox gene clusters of the gnathostome ancestor. We have generated a comprehensive database of conserved Hox noncoding sequences that include cartilaginous, lobe-finned, and ray-finned fishes (bichir and teleosts). Our analysis identified putative and known Hox cis-regulatory sequences with differing depths of conservation in Gnathostoma. We found that although bichir possesses four Hox gene clusters, its pattern of conservation of noncoding sequences is mosaic between outgroups, such as human, coelacanth, and shark, with four Hox gene clusters and teleosts, such as zebrafish and pufferfish, with seven or eight Hox gene clusters. Notably, bichir Hox gene clusters have been invaded by DNA transposons and this trend is further exemplified in teleosts, suggesting an as yet unrecognized mechanism of genome evolution that may explain Hox cluster plasticity in actinopterygians. Taken together, our results suggest that actinopterygian Hox gene clusters experienced a reduction in selective constraints that surprisingly predates the teleost-specific genome duplication.     

Characterization of Hox genes in the bichir,Polypterus palmas

It has been suggested that the increase in the number of Hox genes may have been one of the key events in vertebrate evolution. Invertebrates have one Hox cluster, while mammals have four. Interestingly, the number of Hox gene clusters is greater in the teleost fishes, zebrafish and medaka, than in mouse and human. The greater number of Hox clusters in the teleosts suggests that Hox gene duplication events have occurred during the radiation of ray-finned fishes. The question is when the Hox gene duplication event(s) that lead to seven Hox clusters in the teleosts actually occurred.We have addressed this question by studying the Hox genes in the bichir, Polypterus palmas. A preliminary PCR-estimation of the number of Hox genes suggests that Polypterus has five different Hox9 cognate group genes, which may be an indication of more than four Hox clusters in the bichir.     

Consequences of Hox gene duplication in the vertebrates: an investigation of the zebrafish Hox paralogue group 1 genes

As a result of a whole genome duplication event in the lineage leading to teleosts, the zebrafish has seven clusters of Hox patterning genes, rather than four, as described for tetrapod vertebrates. To investigate the consequences of this genome duplication, we have carried out a detailed comparison of genes from a single Hox paralogue group, paralogue group (PG) 1. We have analyzed the sequences, expression patterns and potential functions of all four of the zebrafish PG1 Hox genes, and compared our data with that available for the three mouse genes. As the basic functions of Hox genes appear to be tightly constrained, comparison with mouse data has allowed us to identify specific changes in the developmental roles of Hox genes that have occurred during vertebrate evolution. We have found variation in expression patterns, amino acid sequences within functional domains, and potential gene functions both within the PG1 genes of zebrafish, and in comparison to mouse PG1 genes. We observed novel expression patterns in the midbrain, such that zebrafish hoxa1a and hoxc1a are expressed anterior to the domain traditionally thought to be under Hox patterning control. The hoxc1a gene shows significant coding sequence changes in known functional domains, which correlate with a reduced capacity to cause posteriorizing transformations. Moreover, the hoxb1 duplicate genes have differing functional capacities, suggesting divergence after duplication. We also find that an intriguing function 'shuffling' between paralogues has occurred, such that one of the zebrafish hoxb1 duplicates, hoxb1b, performs the role in hindbrain patterning played in mouse by the non-orthologous Hoxa1 gene.     

Genomic analysis of Hox clusters in the sea lamprey Petromyzon marinus

The sea lamprey Petromyzon marinus is among the most primitive of extant vertebrates. We are interested in the organization of its Hox gene clusters, because, as a close relative of the gnathostomes, this information would help to infer Hox cluster organization at the base of the gnathostome radiation. We have partially mapped the P. marinus Hox clusters using phage, cosmid, and P1 artificial chromosome libraries. Complete homeobox sequences were obtained for the 22 Hox genes recovered in the genomic library screens and analyzed for cognate group identity. We estimate that the clusters are somewhat larger than those of mammals (roughly 140 kbp vs. 105 kbp) but much smaller than the single Hox cluster of the cephalochordate amphioxus (at more than 260 kb). We never obtained more than three genes from any single cognate group from the genomic library screens, although it is unlikely that our screen was exhaustive, and therefore conclude that P. marinus has a total of either three or four Hox clusters. We also identify four highly conserved non-coding sequence motifs shared with higher vertebrates in a genomic comparison of Hox 10 genes.     

The NK homeobox gene cluster predates the origin of Hox genes

Hox and other Antennapedia (ANTP)-like homeobox gene subclasses - ParaHox, EHGbox, and NK-like - contribute to key developmental events in bilaterians [1-4]. Evidence of physical clustering of ANTP genes in multiple animal genomes [4-9] suggests that all four subclasses arose via sequential cis-duplication events. Here, we show that Hox genes' origin occurred after the divergence of sponge and eumetazoan lineages and occurred concomitantly with a major evolutionary transition in animal body-plan complexity. By using whole genome information from the demosponge Amphimedon queenslandica, we provide the first conclusive evidence that the earliest metazoans possessed multiple NK-like genes but no Hox, ParaHox, or EHGbox genes. Six of the eight NK-like genes present in the Amphimedon genome are clustered within 71 kb in an order akin to bilaterian NK clusters. We infer that the NK cluster in the last common ancestor to sponges, cnidarians, and bilaterians consisted of at least five genes. It appears that the ProtoHox gene originated from within this ancestral cluster after the divergence of sponge and eumetazoan lineages. The maintenance of the NK cluster in sponges and bilaterians for greater than 550 million years is likely to reflect regulatory constraints inherent to the organization of this ancient cluster.     

Chromosomal mapping of ANTP class homeobox genes in amphioxus: piecing together ancestral genomes

Homeobox genes encode DNA-binding proteins, many of which are implicated in the control of embryonic development. Evolutionarily, most homeobox genes fall into two related clades: the ANTP and the PRD classes. Some genes in ANTP class, notably Hox, ParaHox, and NK genes, have an intriguing arrangement into physical clusters. To investigate the evolutionary history of these gene clusters, we examined homeobox gene chromosomal locations in the cephalochordate amphioxus, Branchiostoma floridae. We deduce that 22 amphioxus ANTP class homeobox genes localize in just three chromosomes. One contains the Hox cluster plus AmphiEn, AmphiMnx, and AmphiDll. The ParaHox cluster resides in another chromosome, whereas a third chromosome contains the NK type homeobox genes, including AmphiMsx and AmphiTlx. By comparative analysis we infer that clustering of ANTP class homeobox genes evolved just once, during a series of extensive cis-duplication events of genes early in animal evolution. A trans-duplication event occurred later to yield the Hox and ParaHox gene clusters on different chromosomes. The results obtained have implications for understanding the origin of homeobox gene clustering, the diversification of the ANTP class of homeobox genes, and the evolution of animal genomes.     

Evolution of Conserved Non-Coding Sequences Within the Vertebrate Hox Clusters Through the Two-Round Whole Genome Duplications Revealed by Phylogenetic Footprinting Analysis

As a result of two-round whole genome duplications, four or more paralogous Hox clusters exist in vertebrate genomes. The paralogous genes in the Hox clusters show similar expression patterns, implying shared regulatory mechanisms for expression of these genes. Previous studies partly revealed the expression mechanisms of Hox genes. However, cis-regulatory elements that control these paralogous gene expression are still poorly understood. Toward solving this problem, the authors searched conserved non-coding sequences (CNSs), which are candidates of cis-regulatory elements. When comparing orthologous Hox clusters of 19 vertebrate species, 208 intergenic conserved regions were found. The authors then searched for CNSs that were conserved not only between orthologous clusters but also among the four paralogous Hox clusters. The authors found three regions that are conserved among all the four clusters and eight regions that are conserved between intergenic regions of two paralogous Hox clusters. In total, 28 CNSs were identified in the paralogous Hox clusters, and nine of them were newly found in this study. One of these novel regions bears a RARE motif. These CNSs are candidates for gene expression regulatory regions among paralogous Hox clusters. The authors also compared vertebrate CNSs with amphioxus CNSs within the Hox cluster, and found that two CNSs in the HoxA and HoxB clusters retain homology with amphioxus CNSs through the two-round whole genome duplications.     

Hox gene complexity in medaka fish may Be similar to that in pufferfish rather than zebrafish.

Changes in number and the genomic organization of Hox genes have played an important role in metazoan body-plan evolution. They make cluster(s), and in vertebrates, each cluster contains different number of Hox genes that have been classified into 13 groups. There are 39 Hox genes in four clusters on different chromosomes in the mammalian genome. In the fish, while 31 Hox genes in four clusters have been identified in pufferfish Fugu rubripes, 47 Hox genes in seven clusters exist in the zebrafish Danio rerio. To estimate the evolutionary origin of Hox organization in ray-finned fishes, we searched for Hox genes in the medaka fish Oryzias latipes, with a taxon thought to be widely separated from those of pufferfish and zebrafish. We synthesized various mixed oligonucleotides that can work as group-specific primers for PCR, then cloned and sequenced amplified fragments. Numbers of Hox genes identified in the present study were 2 for group 1, 2 for group 2, 1 for group 3, 3 for group 4, 6 for groups 5-7, 2 for group 8, 4 for group 9, 3 for group 10, 1 for group 12, and 3 for group 13. The primers specific for group 11 did not function in this study. Thus, at least 27 Hox genes are present in medaka genome, suggesting that the Hox gene complexity of the medaka genome is similar to that of the pufferfish rather than the zebrafish.     

HOX Pro: a specialized database for clusters and networks of homeobox genes

It is now clear that the homeobox motif is well conserved across metazoan phyla. It has been established experimentally that a subset of genes containing this motif plays key roles in the orchestration of gene expression during development. Auto- and cross-regulatory functional interactions join homeobox genes into genetic networks. We have developed a specialized database HOX-Pro in order to arrange all available data on structure, function, phylogeny and evolution of Hox genes, Hox clusters and Hox networks. Its primary location is http://www.iephb.nw.ru/hoxpro. The database is also mirrored at http://www.mssm.edu/molbio/hoxpro. The HOX-Pro database is aimed at: (i) analysis and classification of regulatory and coding regions in diverse homeobox and related genes; (ii) comparative analysis of organization of 'Hox-based' genetic networks in the sea urchin Strongylocentrotus purpuratus, the fruit fly Drosophila melanogaster and the mouse Mus musculus; and (iii) analysis of phylogeny and evolution of homeobox genes and clusters.     

Chromosomal localization of homeobox genes and associated markers on porcine Chromosomes 3, 5, 12, 15, 16 and 18: comparative mapping study with human and mouse

Four homeobox genes that belong to the four homeobox gene clusters known in mammals have been regionally assigned to four distinct porcine chromosomes in conserved regions between human and pig. HOXA11, HOXB6, HOXC8, and HOXD4 genes were mapped by radioactive in situ hybridization to porcine Chromosomes (Chrs) 18q21-24 (with a secondary signal in 16q14-21), 12p11-12, 5p11-12, and 15q22-23 respectively. Besides, we have also revealed the presence of a porcine homeobox (pig Hbx24) which, although showing DNA sequence homology with a mouse gene of HOXB cluster, was located on porcine Chr 3 (3p14-13) outside the Hox clusters. To support the identity of the homeobox gene clusters analyzed and in the light of the high sequence similarity among homeobox genes, we also localized markers known to be mapped near each Hox cluster in human. In this way, four genes were also mapped in pig: GAPD (5q12-21), GAD1 (15q21-22), INHBA (18q24), and IGFBP3 (18q24). Mapping of HOXA11, INHBA, and IGFBP3 on pig Chr 18 constitutes the first assignments of genes on this small chromosome. These new localizations extend the information on the conservation of four human chromosomal regions in the pig genome. Received: 7 August 1995 / Accepted: 16 October 1995     

Isolation of Hox genes from the scyphozoan Cassiopeia xamachana: implications for the early evolution of Hox genes.

The isolation of Hox genes from two cnidarian groups, the Hydrozoa and Anthozoa, has sparked hypotheses on the early evolution of Hox genes and a conserved role for these genes for defining a main body axis in all metazoan animals. We have isolated the first five Hox genes, Scox-1 to Scox-5, from the third cnidarian class, the Scyphozoa. For all but one gene, we report full-length homeobox plus flanking sequences. Four of the five genes show close relationship to previously reported Cnox-1 genes from Hydrozoa and Anthozoa. One gene, Scox-2, is an unambiguous homologue of Cnox-2 genes known from Hydrozoa, Anthozoa, and also Placozoa. Based on sequence similarity and phylogenetic analyses of the homeobox and homeodomain sequences of known Hox genes from cnidarians, we suggest the presence of at least five distinct Hox gene families in this phylum, and conclude that the last common ancestor of the Recent cnidarian classes likely possessed a set of Hox genes representing three different families, the Cnox-1, Cnox-2, and Cnox-5 families. The data presented are consistent with the idea that multiple duplication events of genes have occurred within one family at the expense of conservation of the original set of genes, which represent the three ancestral Hox gene families.     

Trends in the evolution of the prodynorphin gene in teleosts: cloning of eel and tilapia prodynorphin cDNAs

The detection of the prodynorphin gene in anuran amphibians and lungfishes may indicate that this gene arose as a result of the duplication of the proenkephalin gene early during the divergence of the Sarcopterygii, or that this gene may predate the divergence of the ray-finned fish and the lobe-finned fish. The cloning of prodynorphin-related genes from the pufferfish and zebrafish supports the latter hypothesis. This study analyzes trends in the radiation of the prodynorphin gene in teleosts. Prodynorphin cDNAs were cloned from the brain of the eel Anguilla rostrata and the Nile tilapia, Oreochromis niloticus. These teleost prodynorphin sequences have distinct alpha-neoendorphin, dynorphin A, and dynorphin B sequences, and a novel opioid sequence, YGGFI. The relationship of these teleost prodynorphin sequences to other actinopterygian and sarcopterygian prodynorphin sequences will be discussed.     

Expression pattern of the homeobox gene Not in the basal metazoan Trichoplax adhaerens

The homeobox gene Not is highly conserved in Xenopus, chicken and zebrafish with an apparent role in notochord formation, which inspired the name of this distinct subfamily. Interestingly, Not genes are also well conserved in animals without notochord such as sea urchins, Drosophila or even Hydra, but appear to be highly derived in mammals. A search for homeobox genes in the placozoan Trichoplax adhaerens, one of the simplest organisms available today, revealed only two homeobox genes: a Not homologue and the previously described gene Trox-2, which is most similar to the Gsx subfamily of the Hox/ParaHox cluster genes. Not has a unique expression profile in Trichoplax. It is highly expressed in folds of intact animals and in the wounds of regenerating animals. The dynamic expression pattern of Trichoplax Not is discussed in comparison with the invariable expression pattern of Trox-2 and the putative secreted protein Secp1. The high sequence conservation of Not from Trichoplax to lower vertebrates, but not to mammals, represents a rare example of an apparent gene decay in the lineage leading to humans.