Sense-antisense gene overlap is probably a cause for retaining the few introns in Giardia genome and the implications

Abstract

Background: It is widely accepted that the last eukaryotic common ancestor and early eukaryotes were intron-rich and intron loss dominated subsequent evolution, thus the presence of only very few introns in some modern eukaryotes must be the consequence of massive loss. But it is striking that few eukaryotes were found to have completely lost introns. Despite extensive research, the causes of massive intron losses remain elusive. Actually the reverse question -- how the few introns can be retained under the evolutionary selection pressure of intron loss -- is equally significant but was rarely studied, except that it was conjectured that the essential functions of some introns prevent their loss. The situation that extremely few (eight) spliceosome-mediated cis-spliced introns present in the relatively simple genome of Giardia lamblia provides an excellent opportunity to explore this question.

Results: Our investigation found three types of distribution patterns of the few introns in the intron-containing genes: ancient intron in ancient gene, later-evolved intron in ancient gene, and later-evolved intron in later-evolved gene, which can reflect to some extent the dynamic evolution of introns in Giardia. Without finding any special features or functional importance of these introns responsible for their retention, we noticed and experimentally verified that some intron-containing genes form sense-antisense gene pairs with transcribable genes on their complementary strands, and that the introns just reside in the overlapping regions.

Conclusions: In Giardia's evolution, despite constant evolutionary selection pressure of intron loss, intron gain can still occur in both ancient and later-evolved genes, but only a few introns are retained; at least the evolutionary retention of some of the introns might not be due to the functional constraint of the introns themselves but the causes outside of introns, such as the constraints imposed by other genomic functional elements overlapping with the introns. These findings can not only provide some clues to find new genomic functional elements -- in the areas overlapping with introns, but suggest that "functional constraint" of introns may not be necessarily directly associated with intron loss and gain, and that the real functions are probably still outside of our current knowledge.

Reviewers: This article was reviewed by Mikhail Gelfand, Michael Gray, and Igor Rogozin.

Keywords: Evolutionary retention of introns; Gene overlap; Genome evolution; Giardia lamblia; Intron evolution.

Fig. 1

Results of strand-specific RT-PCR and sequencing of the two antisense genes and their corresponding schematic diagram. a Lane 1, Strand-specific RT-PCR product of the GL50803–20429; Lane 4, the Strand-specific RT-PCR product of the GL50803–28204; Lane 2 and lane 3, negative controls (with no RTase) corresponding to lane 1 and lane 4, respectively; M, molecular markers. b Nucleotide sequence of GL50803–28204 gene acquired by Strand-specific RT-PCR and sequencing. The locations of the primers are underlined. c Nucleotide sequence of GL50803–20429 gene acquired by Strand- specific RT-PCR and sequencing. d Schematic diagram of the two SAS gene pairs. The sequence lengths of GL50803–28204 and GL50803–20429 are according to the Strand-specific RT-PCR products, and the lengths of GL50803–37070 and GL50803–17244 are based on the GiardiaDB database. Arrow represents the orientation of transcription; and the dashed box and solid lines represent introns and exons, respectively