Evidence for at least six Hox clusters in the Japanese lamprey (Lethenteron japonicum)

Abstract

Cyclostomes, comprising jawless vertebrates such as lampreys and hagfishes, are the sister group of living jawed vertebrates (gnathostomes) and hence an important group for understanding the origin and diversity of vertebrates. In vertebrates and other metazoans, Hox genes determine cell fate along the anteroposterior axis of embryos and are implicated in driving morphological diversity. Invertebrates contain a single Hox cluster (either intact or fragmented), whereas elephant shark, coelacanth, and tetrapods contain four Hox clusters owing to two rounds of whole-genome duplication ("1R" and "2R") during early vertebrate evolution. By contrast, most teleost fishes contain up to eight Hox clusters because of an additional "teleost-specific" genome duplication event. By sequencing bacterial artificial chromosome (BAC) clones and the whole genome, here we provide evidence for at least six Hox clusters in the Japanese lamprey (Lethenteron japonicum). This suggests that the lamprey lineage has experienced an additional genome duplication after 1R and 2R. The relative age of lamprey and human paralogs supports this hypothesis. Compared with gnathostome Hox clusters, lamprey Hox clusters are unusually large. Several conserved noncoding elements (CNEs) were predicted in the Hox clusters of lamprey, elephant shark, and human. Transgenic zebrafish assay indicated the potential of CNEs to function as enhancers. Interestingly, CNEs in individual lamprey Hox clusters are frequently conserved in multiple Hox clusters in elephant shark and human, implying a many-to-many orthology relationship between lamprey and gnathostome Hox clusters. Such a relationship suggests that the first two rounds of genome duplication may have occurred independently in the lamprey and gnathostome lineages.

Fig. 1.

Hox gene loci in the Japanese lamprey. Genes are represented as boxes and arrows denote the direction of transcription. Pseudogenes are denoted by the ψ symbol. Hox gene pairs Hox-ε8/ε7 and Hox-ε4/ε1 are putatively assigned to be part of Hox-ε locus comprising Hox-ε14 to Hox-ε9 genes. Likewise, Hox genes Hox-ζ13, Hox-ζ9/ζ4, and Hox-ζ3 are putatively assigned to be part of a single locus. More data are required to confirm whether they are really part of such loci/clusters. e2, second exon (only the second exon could be identified for these genes).

Fig. 2.

Expression patterns driven by lamprey Hox-αCNE2 and its human homolog in 3 dpf F1 generation zebrafish embryos. (A, B, D, and E) Lateral views. (C and F) Dorsal views. fb, forebrain; hb, hindbrain; mb, midbrain; oc, otic capsule; pf, pectoral fin; sc, spinal cord.

Fig. 3.

The 4DTv of paralogs from lamprey and other vertebrates compared with human. (A) Japanese lamprey and human; (B) stickleback and human; (C) coelacanth and human. The 4DTv rates of paralogs in each genome showed a bimodal distribution, with 77–92% of the gene pairs having a 4DTv value of 0.2 or more. These gene pairs are composed mainly of ancient duplicate genes. The median 4DTv values for these gene pairs in human, lamprey, stickleback, and coelacanth are 0.437, 0.402, 0.409, and 0.444, respectively. The 4DTv rates of coelacanth are similar to that of human, consistent with the notion that human and coelacanth shared the last round of whole-genome duplication (i.e., 2R). On the other hand, the 4DTv rates of lamprey and stickleback are lower than that of human. For stickleback, this result is consistent with the teleost-specific additional genome duplication. For lamprey, this implies that there exists one round of whole-genome duplication that is more recent than 2R.