13 Jul 2005

Twin discordance—the manifestation of vastly different behaviors and susceptibilities to disease despite genetic identity—is a bit of a mystery. Some might attribute it to exposure to quite different environments after birth. But is this nurture argument sufficient to explain why one identical twin falls victim to something like Alzheimer disease, for example, while the other does not?

An international study led by Manel Esteller at the Spanish National Cancer Center in Madrid paints a slightly more detailed picture of twin discordance. Reporting in last week’s PNAS online, first author Mario Fraga and colleagues show that as twins age they accumulate vast numbers of epigenetic differences. In short, nurture may be altering nature.

Epigenetic differences, such as DNA methylation or histone acetylation, can have a profound impact on gene expression. To see if this might explain twin discordance, Fraga and colleagues, including researchers in Spain, Sweden, England, and the US, examined DNA and histone modification patterns in tissue samples taken from 80 identical (monozygotic) twins (30 female, 50 male). The authors found that in very young twins (3 years old), there were no statistical differences in overall cytosine methylation or in acetylation of histones H3 and H4. Samples from older twins told a different story, however. While the majority of twin pairs exhibited few differences, in 35 percent of twin pairs, all three parameters varied considerably. In one pair of 50-year-old twins, DNA methylation was 3.5 percent in one twin and 4.5 percent in the other, while histone H3 and H4 acetylations were about 60 vs. 48 percent and 24 vs. 18 percent, respectively.

The data suggest that in at least some cases, epigenetic characteristics that were similar close to birth have diverged as the twins aged. What’s more, this divergence may have an impact on health. Fraga and colleagues found that those twins who had discernibly different health and medical histories were those who had the greatest epigenetic differences. Twins who spent less of their lifetimes together also turned out to have greater epigenetic discordance than those who were closely knit.

For the most part, Fraga and colleagues focused on changes in lymphocytes. But they also examined epigenetic profiles of epithelial, fat, and skeletal muscle cells, finding similar patterns of divergence as for the blood cells. Functionally, the differences do seem important. For example, when the authors examined loci where methylation differences were observed, they found that about 34 percent of the loci matched expressed sequence tags, while 13 percent matched single copy genes. Zeroing in on a few specific sequences, Fraga and colleagues found that a pattern of hypomethylation in one twin and hypermethylation in the other was common.

So might these differences affect gene expression? To answer that, the authors used gene array analysis to compare mRNA levels between twins. They found that, as with methylation and acetylation patterns, portraits of gene expression showed the greatest differences in older twins and that transcript levels correlated with epigenetic changes. Twins with the most severe hypomethylation and hyperacetylation, for example, had the highest numbers of overexpressed genes, up to almost 4,000.

What seems clear from this study is that epigenetic changes can accumulate as people age. What is less clear is the cause of these changes. The authors suggest that they could be due to “epigenetic drift,” defects in the transmission of epigenetic changes during cell division. But they also suggest that external factors such as smoking habits, physical activity, or diet can also be influential, a suggestion backed up by the fact that twins with the most divergent natural health histories also had the greatest epigenetic differences. In the wider context of disease, “our comparison of MZ [monozygotic] twins suggests that external and/or internal factors can have an impact in the phenotype by altering the pattern of epigenetic modifications and thus modulating genetic information,” conclude the authors.—Tom Fagan

Comments

No Available Comments

Make a Comment

To make a comment you must login or register.

References

No Available References

Further Reading

No Available Further Reading