by Felicia Low, Alan Beedle, Tatjana Buklijas & Peter Gluckman

The Evolution and Medicine Review June 12, 2012

A recent paper published in PLoS Genetics (Li et al., 2012) has made intriguing links between the epigenome and genomic structural stability in the human germline, which is suggested to provide at least a partial evolutionary explanation for neurocognitive disorders such as schizophrenia and developmental delay.

Although it is known that structural mutability – such as that involving insertions and deletions of chromosomal segments – occurs in hotspots in the genome, the mechanisms underlying mutability remain unclear. Such hypermutable loci, which would promote the pace of evolution, have been associated with some genomic disorders in humans (Lupski, 2009), but the phenomenon of structural mutability remains to be fully explained.

Prior work has suggested that the epigenome plays a role in structural mutability. For example histone modifications are highly involved in DNA repair and recombination, and mutations in human and mouse DNA methyltransferases lead to genomic instability; hypomethylation has also been linked to structural instability in the gibbon and in human cancer cells. Li et al. have now incorporated a bioinformatics approach into evolutionary, population genetic and disease analyses to assess how methylation levels impact on genome structural stability in the human germline, and if any associations with human disease can be revealed.

They studied four methylation maps of human sperm, and compared the data to the genomes from several non-human primates to determine human-specific structural rearrangements. As expected, these rearrangements were found to be promoted by low-copy repeats (LCRs), and perhaps surprisingly, were strongly associated with so-called ‘methylation deserts’ that comprise about 1% of the genome with the lowest germline methylation levels. The number of structural rearrangements at methylation deserts was an order of magnitude higher than the genome-wide average, and this association was validated by further comparison with publicly available structural variation data.

Additional tests were also performed to confirm that methylation levels in sperm were sufficiently representative of the whole human (male and female) germline, and that the hypomethylation-structural mutability association was specific to the germline.

Using published datasets of genomic regulatory blocks (GRBs), the team established that methylation deserts were enriched for loci implicated in developmental regulation. They then analysed CNV distribution data from previous case-control studies of schizophrenia, bipolar disorder, autism and developmentally delayed patients, and found that rare or de novo CNVs were particularly enriched in methylation deserts.

The novel finding is that structural mutability is linked to methylation deserts, thus identifying the latter as hotspots for evolutionary change, and it is suggested that these methylation deserts are genomically linked to neurocognitive disorders. Little is known about methylation deserts – the authors did find a higher than expected GC content owing to enrichment in “weak CpG islands”, that is those that tend to undergo methylation during differentiation and are therefore likely to be environmentally sensitive, but it remains to be seen how methylation may exert a protective effect, and what is the basis for its paucity in those specific regions.

Beyond that, this paper also prompts questions relating to evolutionary explanations for why structural mutability has persisted in humans and not been eliminated through natural selection. Based on the assumption that de novo mutations are likely to underlie diseases related to phenotypes undergoing positive selection, and that brain function is one such phenotype, the authors speculate that elevated risk of neurocognitive disorder conferred by mutability may be a tradeoff in the generation of evolutionary novelty necessary for human-specific brain function. That there is a preponderance of rare or de novo CNVs in methylation deserts of patients with such diseases is consistent with this, the authors suggest. The notion of tradeoffs underlying the evolutionary persistence of cognitive disorders in humans has had a long history of contemplation; molecular data presented in this paper is now likely to extend the discussion past ideas of specific causative genes (see De Bont, 2010).

The paper also adds to considerations of how epigenetic biology and evolutionary processes may interact. It joins a growing number of studies suggesting a role for epigenetics in influencing genetic change. For example, it has been shown that CpG content in inactive L1 retrotransposons is associated with both the extent and type of mutations in non-CpG DNA (Walser & Furano, 2010). Alternatively, the bias in mutation rates at sites of epigenetic modification could facilitate fixation – deamination of the hypermutable methylated CpG (wherein methylcytosine converts into thymine) effects a DNA sequence change that, if not enzymatically repaired, could become genomically fixed.

Such studies are feeding a wider discussion on the potential of epigenetics to drive evolutionary change (Bateson & Gluckman, 2011). In theory, should epigenetically mediated genetic change become fixed in the genome, phenotypic changes that ensue from such biased mutations can then be subject to Darwinian selection. The use of germline cells (sperm) by Li et al is particularly important given the relevance to the transgenerational passage of epigenetic marks, which would further support a role in evolutionary processes.

 

References

Bateson P & Gluckman P (2011) Plasticity, Robustness, Development and Evolution. Cambridge: Cambridge University Press

De Bont R (2010) Schizophrenia, evolution and the borders of biology: on Huxley et al.’s 1964 paper in Nature. History of Psychiatry 21: 144-59

Li J, Harris RA, Cheung SW, Coarfa C, Jeong M, Goodell MA, White LD, Patel A, Kang S-H, Shaw C, Chinault AC, Gambin T, Gambin A, Lupski JR & Milosavljevic A (2012) Genomic hypomethylation in the human germline associates with selective structural mutability in the human genome. PLoS Genetics 8(5): e1002692

Lupski JR (2009) Genomic disorders ten years on. Genome Medicine 1: 42

Walser JC & Furano AV (2010) The mutational spectrum of non-CpG DNA varies with CpG content. Genome Research 20: 875-82

 

Discover more from The Evolution and Medicine Review

Subscribe now to keep reading and get access to the full archive.

Continue reading