Since I last wrote, my first semester has ended, and I’ve mostly wrapped up my work for my second rotation (more on this in another post). Now, NYU has a campus in Abu Dhabi with its own Biology research labs, and we are allowed to rotate and even complete our PhDs in one of these labs if we choose. I managed to set up a mini-rotation (just January) with a transposon evolution and population genomics lab at NYU Abu Dhabi.
Before I go into the subject of this post, I should explain what transposons or transposable elements are. Often referred to as “jumping genes”, transposons are DNA sequences found in genomes that seek to make copies of themselves throughout the host genome even if it’s at the detriment of the genome. This imparts selective pressure on the host genome to evolve mechanisms of repressing transposon activity, as the random insertion of transposons into genes and regulatory regions can be deleterious or lethal for the host.
Transposons come in a couple of different flavors (first section in the above link), based on the mechanism of transposition. They occur across all domains of life, but there is a lot of diversity across species in terms of which specific transposons are present and in what abundances. Transposons constitute ~50% of the human genome, but only a handful of these are thought to still be active.
One of the ways in which hosts control transposon activity is through epigenetic modifications such as DNA methylation, which refers to the addition of methyl groups at specific sites along the sequence in order to repress expression. If transposons are not expressed, they cannot produce the enzymes that mediate transposition. The phenomenon of transposon control through DNA methylation and other epigenetic mechanisms has been widely demonstrated in plants, and even in mice, among others.
My little project during the mini-rotation aims to look at transposon evolution since a past hybridization event in an allotetrapolyploid (carrying two sets of each pair of chromosomes from two different ancestral species) species. It was demonstrated way back in 1998 that inter-species hybridization can result in the loss of typical methylation patterns in the original species, leading to the de-repression of transposons. This is where the title and this blogpost comes from.
In that study, the authors looked at the offspring of a mating between a tammar wallaby and a swamp wallaby, whose genera diverged 1-2 million years ago. They were aiming to identify the mechanism of generating chromosomal abnormalities/remodeling in hybrid offspring, which had previously been observed in other macropod hybrids. More specifically, they were looking for an association between chromosomal remodeling, transposon activity, and DNA methylation, presumably based on what was already known at the time about DNA methylation and transposon control. One of their earliest observations for their specific hybrid organism was that the chromosomes of tammar wallaby origin in the hybrid had abnormally extended centromeres (an abnormality).
Taking advantage of a pair of restriction enzymes that can and cannot cut at methylated sites, respectively, they showed that DNA from the hybrid offspring undergoes similar amounts of digestion by both enzymes, while DNA from the parental species undergoes much less digestion by the enzyme that cannot cut at methylated sites. This implies that there is loss of methylation in the hybrid compared to the parental species. They also saw additional bands of DNA in both kinds of digestion of the hybrid’s DNA, suggesting amplifications.
The authors sequenced some of the DNA from the additional bands, and found a novel transposable element with many smaller, derived copies. When they used fluorescently-labeled probes to look for the transposable element in the full chromosomes, they found enrichment of the sequence in the abnormally extended autosomes of the tammar-wallaby-derived autosomes.
In addition, they showed that the transposable element sequence can be found in the digested DNA fragments of the parental DNA (from both parents) only when digested by the restriction enzyme that can cut methylated DNA, but it is not among the digested fragments when only the unmethylated parts of the DNA are cut using the other restriction enzyme. This implies that the element exists in a methylated state in the parental genomes, and loss of methylation results in activation and subsequent amplication in the centromeres in the hybrid offspring.
I will put in a caveat that while these experiments demonstrate strong associations between the loss of methylation, transposon amplification, and a chromosomal defect, they do not directly show causality in a traditional way with experimental manipulations and controls. I still think the evidence is very compelling.
Transposon reactivation following hybridization, if it is a general phenomenon, has important evolutionary implications. Could the resulting changes in chromosomal architecture result in reproductive isolation of the offspring? If the offspring are still fertile among themselves, that could lead to the emergence of a new species. Note that aberrant methylation and/or transposon reactivation in hybrid offspring are not restricted to polyploid hybrids (links in next sentence).
While other studies have also reported similar findings, like in fruit fly, sunflower, and rice hybrids, less is understood about how host genomes deal with and possibly evolve to control such bursts of transposon activity. This is especially intriguing in the case of animals, as viable hybridization is a much rarer event in animals than in plants. Genomes of allotetraploid species like the common carp (Cyprinus carpio), which are known to have originated from hybridization and whole genome duplication events millions of years ago, offer avenues to learn more about host-transposon coevolution in the aftermath of hybridization.