Long Non-coding RNA - Conservation of Long NcRNAs

Conservation of Long NcRNAs

Many small RNAs, such as microRNAs or snoRNAs, exhibit strong conservation across diverse species (Bentwich 2005). In contrast, in general long ncRNAs lack strong conservation, which is often cited as evidence of non-functionality (Brosius 2005; Struhl 2007). However, many well-described long ncRNAs, such as Air and Xist, are poorly conserved (Nesterova 2001), suggesting that ncRNAs may be subject to different selection pressures (Pang 2006). Unlike mRNAs, which have to conserve the codon usage and prevent frameshift mutations in a single long ORF, selection may conserve only short regions of long ncRNAs that are constrained by structure or sequence-specific interactions. Therefore, we may see selection act only over small regions of the long ncRNA transcript. Nevertheless, despite low conservation of long ncRNAs in general, it should be noted that many long ncRNAs still contain strongly conserved elements. For example, 19% of highly conserved phastCons elements occur in known introns, and another 32% in unannotated regions (Siepel 2005). Furthermore, a representative set of human long ncRNAs exhibit small, yet significant, reductions in substitution and insertion/deletion rates indicative of purifying selection that conserve the integrity of the transcript at the levels of sequence, promoter and splicing (Ponjavic 2007).

The poor conservation of ncRNAs may be the result of recent and rapid adaptive selection. For instance, ncRNAs may be more pliant to evolutionary pressures than protein-coding genes, as evidenced by the existence of many lineage specific ncRNAs, such as Xist or Air (Pang 2006). Indeed, those conserved regions of the human genome that are subject to recent evolutionary change relative to the chimpanzee genome occurs mainly in non-coding regions, many of which are transcribed (Pollard 2006; Pollard 2006). This includes a ncRNA, HAR1F, which has undergone rapid evolutionary change in humans and is specifically expressed in the Cajal-Retzius cells in the human neocortex (Pollard 2006). The observation that many functionally validated RNAs are evolving quickly (Pang 2006; Smith 2004), may result from these sequences having more plastic structure-function constraints, and we may expect a great deal of evolutionary innovation to occur in such sequences. This is supported by the existence of thousands of sequences in the mammalian genome that show poor conservation at the primary sequence level but have evidence of conserved RNA secondary structures (Torarinsson 2006; Torarinsson 2008).