Contrasting 5' and 3' Evolutionary Histories and Frequent Evolutionary Convergence in Meis/hth Gene Structures

Abstract

Organisms show striking differences in genome structure; however, the functional implications and fundamental forces that govern these differences remain obscure. The intron–exon organization of nuclear genes is involved in a particularly large variety of structures and functional roles. We performed a 22-species study of Meis/hth genes, intron-rich homeodomain-containing transcription factors involved in a wide range of developmental processes. Our study revealed three surprising results that suggest important and very different functions for Meis intron–exon structures. First, we find unexpected conservation across species of intron positions and lengths along most of the Meis locus. This contrasts with the high degree of structural divergence found in genome-wide studies and may attest to conserved regulatory elements residing within these conserved introns. Second, we find very different evolutionary histories for the 5′ and 3′ regions of the gene. The 5′-most 10 exons, which encode the highly conserved Meis domain and homeodomain, show striking conservation. By contrast, the 3′ of the gene, which encodes several domains implicated in transcriptional activation and response to cell signaling, shows a remarkably active evolutionary history, with diverse isoforms and frequent creation and loss of new exons and splice sites. This region-specific diversity suggests evolutionary “tinkering,” with alternative splicing allowing for more subtle regulation of protein function. Third, we find a large number of cases of convergent evolution in the 3′ region, including 1) parallel losses of ancestral coding sequence, 2) parallel gains of external and internal splice sites, and 3) recurrent truncation of C-terminal coding regions. These results attest to the importance of locus-specific splicing functions in differences in structural evolution across genes, as well as to commonalities of forces shaping the evolution of individual genes along different lineages.

Origin and evolution of spliceosomal introns

“…Some long introns, especially those in 5’-terminal regions of coding sequences, might be enriched for various regulatory elements, and consequently, could be subject to purifying selection [160,247-253]. Long introns in several genes of Oikopleura have been shown to contain key developmental regulators [131], and similar observations have been reported for genes involved in development of diverse metazoans [254-257] as well as associated “bystander” genes that are not known to be directly involved in development [258-261]. …”

Origin of Spliceosomal Introns and Alternative Splicing

“…These include some plants (Jiang and Goertzen 2011), brown algae (Cock et al 2010), and, especially, vertebrates, with averages one order of magnitude higher than most species, and with some introns reaching up to one megabase. These expansions seem likely to be associated with insertion of transposable elements into intronic sequences (Jiang and Goertzen 2011), although, in some cases, exceptionally long introns may be associated with the presence of large numbers of cis-regulatory elements (Irimia et al 2011). …”

Identification and embryonic expression of a highly conserved Meis-linked gene

“…This was an unexpected finding because, although the homeobox sequence of the Meis genes is remarkably similar to that of the Pbx genes, the latter differs in that its protein product follows the canonical homeodomain DNA-binding pattern, recognizing the tetrameric TAAT core sequence (Chang et al, 1997;Lawrence and Largman, 1992). Also, whereas the majority of homeobox genes have few introns, in most organisms the Meis genes contain an unusually high 10 to 11 introns (Irimia et al, 2011). In addition to the ability of the Meis proteins to bind their target hexameric sequence alone, they have also been shown to cooperatively bind DNA in conjunction with other homedomain proteins such as Hox or Pbx.…”

“…The lack of meis1.1 causes a reduction in the expression of key hindbrain patterning genes such as krox20 and various hox members (Waskiewicz et al, 2001). Interestingly, the zebrafish meis1.2 seems to have been lost, resulting in meis1.1 being largely referred to in the literature simply as meis1 (Irimia et al, 2011;Minehata et al, 2008).…”