NEW YORK (GenomeWeb News) – New research is providing a glimpse into the way imprinted genes — those expressed from just one of a chromosome pair — have evolved through mammalian evolution.
In a paper appearing in today’s issue of PLoS Biology, a team of researchers from the UK and Australia, led by University of Cambridge developmental geneticist Anne Ferguson-Smith, focused on a stretch of DNA that’s imprinted in placental mammals but not in monotremes or marsupials. By comparing seven different vertebrates, the team was able to pinpoint factors facilitating imprinting and to start to predict their effects on mammalian evolution.
In most human cells, corresponding genes found on paired chromosomes are dealt with similarly. But that’s not always the case. In humans, other placental mammals, and some plants, certain genes are expressed differently depending on whether they originate from maternally or paternally inherited chromosomes — a situation called imprinting.
Although the chromosomes are homologues and the DNA looks identical, Ferguson-Smith told GenomeWeb Daily News, they function differently. In some instances imprinting is developmentally regulated or tissue specific. In mammals, the brain and placenta are particularly prone to imprinting. Germline methylation seems to be one factor that influences imprinting.
As such, genomic imprinting is a useful model system for understanding epigenetic control, Ferguson-Smith explained, making it possible to look at what’s causing different expression of identical stretches of DNA in the same cellular environment.
One such region is the Dlk1-Dio3 cluster — which contains three protein-coding genes in placental mammals or eutherians. All three protein-coding genes are expressed from the Dlk1-Dio3 region of the paternally inherited chromosome. The maternally-inherited chromosome, on the other hand, expresses functional non-protein coding RNAs including microRNAs, small nucleolar RNAs, and larger non-coding RNA.
Proper imprinting in the region seems to be critical for prenatal development in placental mammals, even though some related animals — such as monotremes and marsupials — lack Dlk1-Dio3 imprinting.
In an effort to understand the evolution of this imprinted domain, Ferguson-Smith and her team compared three types of mammals: egg-laying mammals, marsupials, and eutherians. They sequenced bacterial artificial chromosome clones representing the Dlk1-Dio3 region from the platypus (a monotreme) and the tammar wallaby (a marsupial) using Sanger sequencing.
The researchers also looked at the expression patterns of genes in the region using real-time, reverse-transcriptase PCR and compared their results to previously obtained Dlk1-Dio3 data.
Overall, the marsupial Dlk1-Dio3 region was about twice as long as that of eutherians, owing to multiple repeats. And, in contrast to mammals with imprinting in this region, the researchers found expression from protein-coding genes on both chromosomes of the Dlk1-Dio3 region in monotremes and marsupials.
There were other differences as well. For example, the researchers found DLK1 and DIO3 — “the ancestral genes” — on the outsides of the cluster in both the marsupial and monotoreme. But whereas eutherians have a third protein-coding gene called RTL1 in the region, the wallaby had only non-expressed remnants of RTL1.
In general, what’s going on in the maternal chromosome of imprinted regions doesn’t influence the paternal chromosome, Ferguson-Smith said. But there’s at least one exception: during eutherian imprinting, RTL1 on the paternal chromosome is expressed in the opposite direction as seven or so complimentary, antisense, non-coding RNA from the maternal chromosome.
“We know that these miRNAs can interact with the RTL1 message,” she said. “These miRNAs can modulate the levels of Rtl1.” The miRNAs likely have other genomic targets as well, she added.
It’s difficult to ascertain whether the non-coding RNAs cause or are a consequence of imprinting, Ferguson-Smith emphasized. Still, there is some evidence suggesting the rise of RNAs is a bi-product of imprinting.
One such clue is GTL2, a gene coding for a large, non-protein coding RNA. The gene was present in the wallaby, but expressed at very low levels compared to the strong, mono-allelic expression in eutherians. “When imprinting came along and the protein-coding genes got turned off, it probably facilitated the increased action from this GTL2 transcript,” Ferguson-Smith said.
The team also compared the wallaby and platypus DLK1-DIO3 sequences with five other vertebrate DLK1-DIO3 sequences from the UCSC genome browser: three placental mammals (human, mouse, and dog), another marsupial (the opossum), and a bird (the chicken).
Using this approach, the team identified 141 evolutionarily-conserved regions across the vertebrate sub-groups. Of these, 31 were common to all seven vertebrates. Others were shared between certain species. By looking for changes between the evolutionarily conserved regions of different species they could flag regions of potential import for imprinting and, potentially, mammalian evolution, Ferguson-Smith said.
In the future, her team plans to systematically assess the function of these conserved regions along with changes associated with imprinting in different species. “Because imprinting seems to have evolved at different times in different species, we’d really like to understand why some genes are imprinted in marsupials and some aren’t,” Ferguson-Smith said.