The Human Hox-bearing Chromosome Regions Did Arise by Block or Chromosome (or Even Genome) Duplications

Abstract

Many chromosome regions in the human genome exist in four similar copies, suggesting that the entire genome was duplicated twice in early vertebrate evolution, a concept called the 2R hypothesis. Forty-two gene families on the four Hox-bearing chromosomes were recently analyzed by others, and 32 of these were reported to have evolutionary histories incompatible with duplications concomitant with the Hox clusters, thereby contradicting the 2R hypothesis. However, we show here that nine of the families have probably been translocated to the Hox-bearing chromosomes more recently, and that three of these belong to other chromosome quartets where they actually support the 2R hypothesis. We consider 13 families too complex to shed light on the chromosome duplication hypothesis. Among the remaining 20 families, 14 display phylogenies that support or are at least consistent with the Hox-cluster duplications. Only six families seem to have other phylogenies, but these trees are highly uncertain due to shortage of sequence information. We conclude that all relevant and analyzable families support or are consistent with block/chromosome duplications and that none clearly contradicts the 2R hypothesis.

The duplication of the Hox gene clusters in teleost fishes*1

“…Whether the duplicates of genes in vertebrates (in comparison to invertebrates) and in higher teleosts (in comparison to sarcopterygians) have arisen by means of genome duplication(s) and subsequent massiv gene loss, or whether a large number of local duplication occured has been the subject of debate. We refer to [3,32,34,30,47,70,71] for the genome-duplication point of view and to [51,20] for the local duplication viewpoint. In both scenarios it is undisputed that each of the Hox clusters was duplicated as a unit.…”

Timing and mechanism of ancient vertebrate genome duplications – the adventure of a hypothesis

“…Finally, they point out that the phylogenetic-tree-topology approach is inaccurate in several cases and therefore of limited use in deciding if the members of neighbor gene families on the paralogons were generated by the same sequence of duplication events (Box 1). Larhammar et al [41] concluded that 14 of the 20 families on the human Hox-cluster-bearing chromosomes were co-duplicated. These regions are significantly large and they cover 14.6%, 13.3%, 6.7% and 28.3% of the chromosomes 7, 17, 12, and 2, respectively.…”

“…Moreover, they argued that even these genes were not co-duplicated with the Hox clusters because their order of duplication [as indicated by the topology of the relevant phylogenetic trees (AB)(CD) measure (Box 1)] is not the same as that of the Hox cluster genes. Larhammar et al [41] stress that only genes that are ancestrally linked to the Hox clusters and not those purportedly transported on the Hox-bearing chromosomes at a later stage should be considered, because the present gene order on the Hox-bearing chromosomes is affected by the rearrangement history of the specific chromosomes. For example, Hsa 2 resulted from the fusion of two different chromosomes in the primate lineage and Hsa 12 was rearranged during primate evolution [42].…”

The zebrafish genome in context: ohnologs gone missing

“…Evx genes: instructive example of an R2 ohnolog gone missing Chromosome regions surrounding the four tetrapod Hox clusters provided early evidence for two rounds of genome duplication in vertebrate phylogeny (Lundin, '93;Bailey et al, '97b;Larhammar et al, 2002). Adjacent to two vertebrate Hox clusters lie Evx paralogs, which contain a homeobox related to the Drosophila even-skipped segmentation gene and which can act as transcriptional repressors during development.…”