QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 6 June 2007
doi: 10.1242/dev.001065

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Duboule, D. Search for Related Content PubMed PubMed Citation Articles by Duboule, D. Social Bookmarking                         What's this?

The rise and fall of Hox gene clusters

National Research Centre `Frontiers in Genetics', Department of Zoology and Animal Biology, University of Geneva, Sciences III, Switzerland. School of Life Sciences, Ecole Polytechnique Fédérale, Lausanne, Switzerland.

View larger version (46K): [in this window] [in a new window]   Fig. 1. Conventional and to-scale representations of Hox clusters. The colored boxes correspond to the various paralogy groups - those genes that are most closely related in sequence, and hence are derived from a common ancestral Hox gene. (A) Schematic depicting Hox clusters as they are usually represented in the literature and textbooks. It shows the respective positions of paralogy groups 1 to 14 for the cephalochordate amphioxus and a vertebrate prototypic cluster, and their corresponding Drosophila genes located on both ANT-C and BX-C. (B) A more precise representation of the organization within Hox clusters, with the correct relative distances and a clear separation present between the two Drosophila sub-clusters. The comparison between A and B highlights the differences that exist between a structural reality and its conceptual interpretation. Only the former, as shown in B, should be considered when discussing the structural and functional evolution of Hox clusters. ANT-C, Antennapedia complex; BX-C, Bithorax complex.

View larger version (8K): [in this window] [in a new window]   Fig. 2. Structural classification of Hox clusters. Type O (organized) clusters are well organized, with genes tightly arranged and all encoded by the same DNA strand. They are devoid of both `foreign' genes and repeats, yet they may contain non-coding RNAs and miRNAs. Vertebrate clusters provide, so far, the sole example of this organization. Type D (disorganized) clusters are much larger and may contain mixed-up Hox genes (black boxes), or genes in opposite orientations, in addition to non-Hox genes (white boxes) and repeats. Type D examples are found in amphioxus and in sea urchins. Split (type S) clusters can have type O or type D features in each of their sub-clusters, such as in Diptera, whereas the type A (atomized) `cluster' represents the `no-cluster' situation, in which genes are found, at best, in pairs at scattered genomic loci (e.g. in Oikopleura).

View larger version (23K): [in this window] [in a new window]   Fig. 3. Cluster types in various animal species. A recent animal phylogeny with deuterostomes shown in orange, protostomes in green (lophotrochozoans) and yellow (ecdyzozoans), and ctenophorans, cniderians and poriferans arbitrarily shown in blue, as outgroups of bilateria. Cluster type is indicated to the right. Amongst the deuterostomes, cephalochordates are still positioned closer to vertebrates than are tunicates (urochordates), contrary to the proposal from Delsuc et al. (Delsuc et al., 2006). This uncertainty does not change the nature of the argument regarding the evolution of chordate Hox clusters (see text). Whereas type A, S and D clusters are found throughout these large groups, genuine type O clusters are only described in vertebrates (see text for a discussion of the type D/O cluster found in cephalochordates).

View larger version (18K): [in this window] [in a new window]   Fig. 4. The dilemma of bilaterian Hox cluster evolution. Two possible models can account for the restriction of type O clusters to vertebrates. (A) In the `vertebro-centrist' view, the ancestral bilaterian animal had a type O cluster, which was maintained all the way through to the vertebrate lineage, as illustrated in blue. All other bilaterian animals are derived from this ancestral form, first through a type D cluster (red) and, subsequently, with type S or A clusters. Although this view provides the easiest mechanistic explanation (as it progresses towards disorganization only), it implies that vertebrates are `direct descendants' of this ancestor, which is at odds with our modern view of animal evolution. (B) In the `consolidation model', the ancestral cluster was of type D, as illustrated in red, which implies that a Hox cluster, at some point during the evolution of deuterostomes, underwent a phase of consolidation (that is, an increased organization), as suggested by the blue color on the top right. Although this view fits better with our current understanding of animal phylogeny, it requires some conceptual tools to explain how such an `organization' can occur and be selected for.

View larger version (31K): [in this window] [in a new window]   Fig. 5. Transitions in the structural evolution of Hox clusters. From the starting type D cluster (red, as in Fig. 4), transitions to type S and, subsequently, to type A `clusters' (downward) are easy to imagine, and could occur concomitantly with the loss of those constraints that keep genes together. These transitions do not call for any particular mechanism other than the occurrence of chromosomal breakpoints. In some cases, these breaks might be beneficial, along with the evolution of an organism towards a different developmental mode for which clustered Hox genes might be an obstacle rather than a solution (see the text). The arrows towards the top illustrate the mechanism of cluster consolidation, which might occur together with genome duplication. Duplication of the locus allows for more (or for different) global regulations to evolve, which in turn further consolidate the various clusters. In this scheme, a meta-cis structure must exist before genome duplication to provide the basis for further co-options. The color code illustrates the passage from a type D to a type O cluster (see also the `consolidation model' in Fig. 4).

View larger version (24K): [in this window] [in a new window]   Fig. 6. Meta-cis regulation as a mechanism that triggers vertebrate Hox cluster consolidation. (A) Schematic of a type D cluster present in the ancestral bilaterian animal. Hox genes (black boxes) are present in various transcriptional orientations (arrows), dispersed over a large fragment, which also includes foreign (non-Hox) genes (white boxes). Genes are kept together owing to the presence of cis-acting regulatory sequences (blue circles), which are shared between several neighboring genes. (B) A type D/O cluster present in an ancestral chordate. Owing to the emergence of a meta-cis regulatory sequence in a control region (CR, brown), the genomic organization of target genes is improved as a consequence of positive selection. Progressively, foreign genes are lost together with some large intergenic fragments, which optimizes the meta-cis control. (C) After genome duplication, consolidation is favored owing to the increasing possibility to recruit novel meta-cis regulation (red, yellow, green boxes and arrows), in particular through the intrinsic capacity of a potent control region (CR in B) to generate regulatory novelties by `regulatory priming' (red and dark green in C; see text). This leads to a concentration of large-scale enhancers to form a `global control region' (GCR), which might continue to recruit and/or evolve novel regulatory elements owing to its intrinsic properties. As a consequence, duplicated clusters are further consolidated.

View larger version (13K): [in this window] [in a new window]   Fig. 7. Comparison between the mouse and zebrafish Hox gene clusters. Only the numbers of genes and their paralogy groups are shown for comparison, along with the sizes (scale bar on top) of the various clusters. The four murine (M) clusters are shown with Hox genes in black and, below, their zebrafish (ZF) Hox gene counterparts in gray. Mouse Hoxa, Hoxb and Hoxc clusters have two zebrafish counterparts, whereas the mouse Hoxd cluster has a single orthologous cluster in zebrafish. On the right, both the number of genes per cluster and the sizes (kb) of the respective clusters are shown. Totals are given at the bottom, which show that despite zebrafish having the higher number of clusters and a slight increase in Hox gene number, the overall lengths obtained when all clusters are artificially considered together are virtually identical in the two animals. *, The position of the mouse Hoxb13 gene is not shown at the correct scale, as it is located 100 kb from Hoxb9 (see text). The Hoxb cluster is 200 kb, which makes the overall length of the four murine clusters even larger than that of the seven fish counterparts. , A ZFHoxDb `cluster' is shown, even in the absence of Hox genes, as this cluster was selectively lost, as indicated by the structure of the remaining genomic locus (Woltering and Durston, 2006).

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter