Studying the interaction of genes and environment just got even more complicated, with two papers that introduce a new dimension—time. Work published in the June 25 issue of JAMA establishes that DNA methylation, an epigenetic controller of gene expression, varies in individuals as they age, and provides evidence that the variation is under genetic control. That study comes from Andrew Feinberg of Johns Hopkins University, Baltimore, Maryland, and Vilmundur Gudnason of the Icelandic Heart Association (Hjartavernd) in Reykjavik, Iceland. A second paper, also with an Icelandic connection, reveals that people in that country with a hereditary amyloid angiopathy have experienced a dramatic decline in lifespan over the past 200 years. That work, published in PLoS Genetics by Astridur Palsdottir and colleagues at the University of Iceland and scientists from deCODE Genetics, Reykjavik, suggests that recent lifestyle changes have unmasked the deadly effects of a previously tolerable genetic mutation. Both studies reveal the complexity of gene-environment interactions, and illustrate how changes in either or both over time can play a role in disease. Both studies bear on age-related neurodegenerative diseases like Alzheimer’s, where genes, environment, and time act in concert to bring on the pathology.

The genome, as defined by the information encoded in DNA, does not normally change over time, or from tissue to tissue, but modifications to DNA and associated proteins do. Methylation of DNA is one heritable modification that controls gene expression, and this epigenetic regulator allows fine-tuning of genetic programs during development. In adulthood, methylation remains dynamic, playing a role in regulating gene expression related to memory (see ARF related news story).

Despite all this work, it was not clear if there are widespread changes in methylation over the lifetime of an individual. To address the question, Feinberg and team, led by first author Hans Bjornsson, measured global DNA methylation in the white blood cells of 237 people at two points more than a decade apart. One group consisted of 111 Icelanders whose DNA was sampled twice 11 years apart. The other group consisted of families from Utah, where the analysis spanned an average of 16 years.

A quantitative measure of DNA methylation across the genome revealed that a third of the Icelandic subjects showed a change in methylation over time that was greater than 10 percent. An approximately equal number of people showed an increase or decrease, so that the average change over the whole group was zero. In the Utah families, variation in the changes in methylation was also seen, and the extent of the variation clustered in families. Half of the most extreme decreases (methylation down 20 percent) occurred within two families out of 21 tested. These results support the idea that differences in the stability of overall methylation levels are due at least in part to inherited factors. When the researchers looked at the methylation status of a subset of 807 genes in some of the Utah subjects, they found some common genes that were uniformly highly affected among different families, including a significant number of genes for immunological mediators.

“These data support the idea of age-related loss of normal epigenetic patterns as a mechanism for late-onset of common human diseases,” the authors write. More work will be needed to determine whether and how losses and gains of methylation might contribute to disease. This idea that there are people who are more or less epigenetically stable might explain some of the differences in individual susceptibility to late-onset diseases like AD.

The work also has implications for population-based studies of human disease, the authors write, where epigenomic changes over time might influence the disease phenotype generated by a given genotype. Thus, including epigenetic measures in such studies may lead to a better understanding into environmental or genetic factors involved in disease.

The impact of environmental exposure to the expression of genotype is dramatically illustrated in the second study, which also involved Icelandic subjects. In that work, Palsdottir, Kari Stefansson from deCODE Genetics, and their colleagues report that carriers of a mutation that causes amyloid angiopathy have experienced a 35-year reduction in lifespan during last 200 years, which the researchers attribute to changes in lifestyle, particularly diet.

The mutation in question occurs in the amyloidogenic protein cystatin C, and causes hereditary cystatin C amyloid angiopathy (HCCAA), an autosomal dominant disease with high penetrance. The disease is unique to Iceland, and carriers die from brain hemorrhages quite early, at an average age of only 30 years old. The study traces the mutation back in 15 families, and the data suggests it originated about 18 generations ago in the mid-1500s. People with the mutation who were born before 1825 lived more than 60 years, but after that, there was a rapid drop in lifespan to today’s levels. At the same time, an effect of maternal inheritance also appeared—people who inherited the gene from their mothers have a shorter lifespan by an average of 9.4 years.

The change in phenotype associated with the mutation is unlikely to be due to an additional, unidentified mutation, the authors write, because the effect was seen in several different extended families. Instead, it is likely to result from some environmental factor. They favor diet, because an increase in the consumption of carbohydrates and salt that accompanied economic development occurred at the same time as lifespan was decreasing. Consistent with this, there was a geographic effect on lifespan—the decrease hit a remote northwest coastal region about 20 years later than south and west regions, coinciding with economic development occurring later in the outlying area.

“A mutation with such radically different phenotypic effects in reaction to normal variation in human lifestyle not only opens the possibility of preventive strategies for HCCAA, but it may also provide novel insights into the complex relationship between genotype and human disease,” the authors write. The situation is somewhat like phenylketonuria, where a mutation in phenylalanine metabolism means that exposure to dietary phenylalanine causes brain damage, but the mutation has no phenotype when phenylalanine intake is restricted. In the case of the cystatin mutation, the exact environmental trigger responsible for worsening the disease remains to be determined.

Understanding the interplay could also be relevant to AD, where cerebral amyloid angiopathy occurs frequently, and where cystatin C in its normal form seems to counteract amyloid formation (see ARF related news story).—Pat McCaffrey


  1. The Bjornsson et al. study provides further evidence that DNA methylation differences between individuals increase with age. However, the study not only confirms this principle, but shows that genetic factors play a role in inter-individual methylation differences. It highlights the complexities when studying DNA methylation in aging. While it is thought that "environmental" factors such as alcohol, diet, perhaps medications, etc., play a role in modifying DNA methylation patterns in the genome, genetic factors could play a role as well. Recently, we identified in a postmortem brain study 2/50 gene loci that showed significant alterations in Alzheimer's subjects, as compared to controls (Siegmund et al., 2007). Interestingly, the changes in the Alzheimer's cohort, in terms of DNA methylation, appeared to reflect an acceleration of normal aging. Therefore, one could assume that the findings of Bjornsson et al. will be of great interest for aging-related disorders, including Alzheimer disease.


    . DNA methylation in the human cerebral cortex is dynamically regulated throughout the life span and involves differentiated neurons. PLoS One. 2007;2(9):e895. PubMed.

  2. Palsdottir et al. show in a fascinating analysis a major decrease in the age of death in carriers of hereditary cystatin C cerebral angiopathy (a L68Q mutation in the cystatin C gene) since the eighteenth century. The comparison with spouse lifespan is particularly striking because life expectancy of those surviving to adults was increasing at the same time as life expectancy of the L68Q carriers (“age of lethality penetrance”) was decreasing. In considering the possible environmental factors during these 200 years, the authors note the striking shift in diet composition, including a twofold greater carbohydrate intake (Fig. 7). It is also likely that the total caloric intake increased since the 1800s. Iceland suffered a major food shortage after the Viking age due to the increasingly cold climate: the population declined by about 35 percent and adult height shrank by two inches. As Einarsson (1573-1659) described it: "Formerly the earth produced all sorts of fruit, plants and roots. But now almost nothing grows.... Frost and cold torment people. The good years are rare.” The eighteenth century Icelanders were plausibly still under severe caloric restriction, compounded by micronutrient deficiency. Even in the later nineteenth century with improving climate, Iceland was one of the poorest countries in Europe.

    If this view is valid, then we may consider that caloric restriction was protective for cystatin C L68Q penetrance at an early age. In fact, caloric restriction is protective in various mouse models of brain amyloidosis, familial dominant Alzheimer mutant genes, and of aortic atherosclerosis (Finch, 2007, Chapter 3.2.2; Patel et al. 2005; Wang et al. 2005). There is thus good rationale to examine cystatin C L68Q and other angiopathic mutations for responses to caloric restriction in mouse models as a new approach to prevention.

    See also:

    Einarsson O, quoted in Decline of the Vikings in Iceland.

    Finch CE. 2007. The Biology of Human Longevity. Inflammation, Nutrition, and Aging in the Evolution of Lifespans. Academic Press: San Diego.


    . Caloric restriction attenuates Abeta-deposition in Alzheimer transgenic models. Neurobiol Aging. 2005 Jul;26(7):995-1000. PubMed.

    . Caloric restriction attenuates beta-amyloid neuropathology in a mouse model of Alzheimer's disease. FASEB J. 2005 Apr;19(6):659-61. PubMed.

    View all comments by Caleb (Tuck) Finch
  3. Palsdottir et al. conducted extensive linkage disequilibrium and genealogical studies of patients with HCCAA (also called hereditary cerebral hemorrhage with amyloidosis, Icelandic type—HCHWA-I) and found a decrease in age at onset of the disease, and age at death, of mutation carriers during the nineteenth century. This decrease in age at death, from 65 years in carriers born in 1825 to the present-day average of about 30 years, occurred while an increase in lifespan was documented in the general population in Iceland. This decrease in lifespan paralleled a major change in diet, most significantly an increase in sugar and salt intake in Iceland.

    This study has important significance for our understanding of factors that affect amyloid deposition as well as cerebral hemorrhages. Studies, mainly in animal models of amyloidosis, should be conducted to determine the role of carbohydrates and/or salt in either cerebral amyloid angiopathy (CAA) or cerebral hemorrhage. Carbohydrates have been related to both. Multiple studies have suggested a link between type 2 diabetes and stroke and that glucose lowering in high-risk patients would lower the risk of the disease. In addition, type 2 diabetes mellitus has been associated with a higher incidence of Alzheimer disease (AD). Excess consumption of sugar-sweetened beverages plays an important role in the epidemic of obesity, a major risk factor for type 2 diabetes mellitus. A study has shown that APP/PS1 transgenic mice that were provided with 10 percent sucrose-sweetened water had exacerbation of memory impairment and an increase in insoluble Aβ levels and deposition in the brain compared with control mice with no sucrose added to the water.

    The Leu68Gln variant of cystatin C forms amyloid deposition in cerebral and spinal arteries and arterioles, leading to recurrent hemorrhagic strokes causing serious brain damage and eventually fatal stroke. In vessels affected by CAA, local muscle and elastic elements are lost and replaced by amyloid fibrils, thereby weakening the overall structure of the vessel. Consequently, it was suggested that CAA predisposes towards cerebral infarction and cerebral hemorrhage. However, CAA is usually asymptomatic and only a subpopulation is at high risk of hemorrhage. Several studies implicated non-fibrillar cystatin C in amyloid β-CAA-related hemorrhage. Therefore, it would be of great interest to study whether a change in carbohydrate consumption affects amyloid deposition and/or the occurrence of cerebral hemorrhages with significance not only for cystatin C-related cerebral amyloidosis, but also for AD and stroke.

    View all comments by Efrat Levy
  4. These are indeed highly interesting papers.

    To add to the story of epigenetic influences in the aging process, a new and fascinating study was published in PLoS ONE. The group around Axel Schumacher et al. at the Technical University Munich/Germany could show that people with late-onset Alzheimer disease have indeed an increased “epigenetic drift” in genes that may be responsible for some of the observed phenotypes. Additionally, the group found that some genes that participate in amyloid-β processing and methylation homeostasis show a significant interindividual epigenetic variability, which may contribute to disease predisposition. The observed epigenetic pattern would complement and support the aforementioned data, showing that the changes in the Alzheimer brain appeared to reflect an acceleration of normal aging. This could indicate that everybody has a certain likelihood of developing the disease.


    . Age-specific epigenetic drift in late-onset Alzheimer's disease. PLoS One. 2008;3(7):e2698. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Heavy Methyl—DNA, Protein Modification Affect Memory, APP, and Tau
  2. Genetic Risk Explained: Cystatin C Staves Off Plaque Formation in Mice

Further Reading


  1. . DNA methylation in the human cerebral cortex is dynamically regulated throughout the life span and involves differentiated neurons. PLoS One. 2007;2(9):e895. PubMed.

Primary Papers

  1. . Intra-individual change over time in DNA methylation with familial clustering. JAMA. 2008 Jun 25;299(24):2877-83. PubMed.
  2. . A drastic reduction in the life span of cystatin C L68Q carriers due to life-style changes during the last two centuries. PLoS Genet. 2008 Jun;4(6):e1000099. PubMed.