In a paper from 2004, Peaston et al reported on the expression of various retrotransposons (RTEs) in the mouse oocyte and pre-implantation embryo, finding widespread RTE transcription and the presence of chimeric transcripts composed of host genes and RTEs.
In a cDNA library constructed from full grown oocyte (FGO) transcripts, 12% of sequences were derived from MT (mouse transcript, a member of the MaLR family of nonautonomous LTR class III retrotransposons), whilst in a library from 2 cell stage embryos, 3% of cDNAs were derived from murine ERV-L (another class III LTR RTE). Expression of these and other RTEs tailed off to nothing by the blastocyst stage. The differential developmental expression profile of these RTEs is interesting: MT is a large component of the maternally contributed RNA in the oocyte, whilst MuERV-L must be expressed zygotically very early in development.
The most important finding of this paper was that the cDNA libraries from FGO and 2 cell embryos contained many chimeric gene transcripts in which the 5′ sequence was derived from retrotransposons. These chimeric mRNAs made up 3% of the FGO library and 1.4% of the 2 cell stage embryo library. A large variety of RTEs contribute to chimeras in the FGO library but 51% of them involved MT. 56% of chimeric transcripts in the 2 cell stage had 5′ contributions from MuERV-L and it’s relatives, so RTE composition of the chimeric transcripts correlated with specific RTE abundance. The genes expressed as chimeric transcripts didn’t show any particular functional bias.
When the chimeric transcripts were compared with genomic sequence it was found that the cognate RTEs were either located within the gene locus or upstream of it. If the RTE was encoded within the gene, the chimeric transcript lacked any exons upstream of it. When the RTE was located upstream of the gene, the chimeric mRNA often lacked one or more 5′ exons (2/3rds of the time).
Therefore it appears that RTE sequences act as cis-regulatory elements driving oocyte and pre-implantation embryo specific expression of a population of alternatively spliced transcripts encoding (generally) variant proteins. The notion of RTEs as alternative promoters is close to that of transposons as “controlling elements” put forward by their discoverer Barbara McClintock. The authors note that RTE insertions could give rise to co-regulated gene expression and that RTE driven transcription of multiple host genes “provides grounds for selection of new modes of gene regulation by introducing variation”.
In a review of this work Shapiro uses this as evidence for a “functionalist” perspective, in which he regards mobile elements as “distributed genomic control modules”. This does seem to overstate the purposiveness of TE insertion. One potentially forgets all the cases of deleterious mutations leaving no issue. However, there is no doubt that through evolutionary time, host/parasite arms races can become coevolved integrated functions. An interesting finding in Peaston et al was that sense and antisense transcripts were found in relatively equal ratio when MuERV-L was expressed. This suggested that dsRNA would be formed, triggering RNAi that could seed heterochromatin formation to repress RTE expression. This is again open to a dichotomy of interpretation: in that this is part of a host mechanism to inhibit genomic parasites, or conversely (as Shapiro does) “another mechanism by which RTE insertions can influence the expression of nearby coding sequences and act to construct distributed suites of co-ordinately regulated loci”.
Peaston AE, Evsikov AV, Graber JH, de Vries WN, Holbrook AE, Solter D, & Knowles BB (2004). Retrotransposons regulate host genes in mouse oocytes and preimplantation embryos. Developmental cell, 7 (4), 597-606 PMID: 15469847
Shapiro JA (2005). Retrotransposons and regulatory suites. BioEssays : news and reviews in molecular, cellular and developmental biology, 27 (2), 122-5 PMID: 15666350