Epigenetics and the developmental origins of disease

October 12, 2012 at 9:36 am | Posted in Feature Articles | Leave a comment

Key lessons

1. During developmental periods, epigenetic structures are put in place which set the parameters within which tissues function.

2. The functional changes set by epigenetic structures do not have to be phenotypically (physically) visible to be significant, yet can have long-term consequences for health.

3. Functional changes cannot reliably be detected by short-term, acute toxicity testing: the effects will only be observable in certain environmental contexts and are likely to be masked by the systemic effects of standard regulatory toxicological testing.

The developmental origins of disease. It is a rare research initiative which does not announce itself with bold claims. And the Developmental Origins of Adult Disease (DoHAD) hypothesis is no rarity, claiming in a recent white paper to “provide insight into new strategies for research and disease prevention, while being robust enough to require a public health and policy response” (Barouki et al. 2012). Should we be as excited as the authors?

The significance of the DoHAD hypothesis hinges on the concept of “developmental plasticity”. Developmental plasticity is most prominent during periods when cells are differentiating and forming specific tissues, as principally happens during pregnancy (to both mother and child), early childhood, puberty and menopause. The DoHAD hypothesis is that environmental exposures during developmental periods can cause subtle alterations in physical function which, although physically almost invisible, may increase the risk of disease and dysfunction later in life.

Some of the most striking and earliest-discovered evidence of developmental plasticity has come from famine. Dutch individuals conceived during the wartime famine of 1944-45 were found to be more likely to develop metabolic syndrome during adulthood (see e.g. Rooij et al. 2007); subsequent research has confirmed these findings in Chinese victims of famine (Yanping et al. 2011). In both cases, the offspring are more likely to suffer hypertension, impaired glucose tolerance and excess weight gain.

Dr Robert Barouki, lead author of the DoHAD white paper, explains: “Where [DoHAD] really started was with DES [diethylstilboestrol] and with famine, where exposure was during the foetal period but only clinically observable much later.”

DES is significant in the DoHAD hypothesis for a further reason: it shows that exposure during development to environmental stressors does not have to have visible effects at birth in order to manifest later in life. In the case of DES, the drug had no observable adverse effect on the mother, and there were no physical abnormalities in the children at birth. [Editor’s note: Here we understand “environment” in the broadest sense of encompassing exogenous stressors such as diet, chemical exposure, infection etc. How “environment” is defined is often a source of ambiguity in the literature.]

How might exogenous stressors have almost invisible physiological effects, yet affect long-term health? In this regard, of particular interest to the DoHAD hypothesis is epigenetics (see H&E #22).

Epigenetics can be understood as a kind of in-between mechanism by which organisms can survive in an unpredictable and varying environment. The most rapid adaptations to the environment come from the immediate demands of homeostasis, and are the processes by which the body maintains a consistent internal environment in response to outside changes such as needing more energy, or being too cold or too hot. These are mainly governed by hormone and nerve signals.

At the other end of the adaptive spectrum is evolutionary change, the long-term species adaptions which are driven by random genetic change and the resultant survival (or not) of members of the species which are affected by these changes.

Epigenetics sits somewhere in the middle, switching on and off gene transcription. Transcription is the process by which an RNA copy of a gene is made; the RNA copy then instructs the machinery of the cell to manufacture a specific protein. Adding a methyl group to, or removing one from, a gene is one way of governing whether or not the gene can be transcribed into RNA, and therefore whether or not the gene codes a protein, thereby controlling whether or not a gene is able to contribute to physical processes in the body.

Switching suites of genes on and off will make subtle changes to the overall balance of physiological processes happening in the body, and for example might predispose an organism to conserving energy or spending it, or even (in the case of agouti mice) determine the colour of hair.

Developmental plasticity likely confers some evolutionary advantages, as it allows an organism to adapt to environmental challenges at a pace faster than genetic variation, but in a more long-term, resilient way than is achievable by the changes which maintain homeostasis in response to immediate demands on the body: essentially, epigenetic changes set up the parameters within which the homeostatic changes take place.

What we have learned from famine, for example, is that children conceived during periods of food scarcity are born with methyl groups stripped from several genes involved in growth and metabolic control, with the result that they are predisposed to conserving energy.

Barouki and his co-authors observe that in some circumstances this is good preparation for an organism which will live in a low-food environment. Intuitively it seems more likely that this particular adaptation is of less benefit to humans (food availability cycles being much shorter than a human life) and instead inherited from shorter-lived ancestors.

Regardless, the important point is that if a person can be metabolically primed for a low-food environment, then it makes it possible for there to be mismatches between what has been pre-programmed during development and what is encountered in the real world. This is why someone conceived during a famine is more likely to become overweight if they encounter an environment which is food-rich. This is especially likely in a species as long-lived as a modern European, for whom food availability is largely plentiful and famines are freak events.

What is also notable is these changes in programming are largely functional. They include alterations in gene expression, protein concentrations, cell metabolism and differentiation, and in cell numbers and location. Functional changes are not necessarily identifiable as pathological states themselves, but because the changes in gene expression may lead to increased dysfunction and disease in later life, they can be understood as markers of increased risk of non-communicable disease.

Functional changes need not even be apparent at birth, but may require a particular environmental or physiological trigger in order to become manifest; subtle changes to breast tissue, for example, may only significantly increase the risk of tumour growth in childless middle age; or (as we see from famine) alterations to metabolic set points may only be important in certain dietary contexts (which, incidentally, is why fast food, modern, high-stress lifestyles and chemical exposures may be a perfect storm for obesity).

The important thing is these functional changes are not necessarily easily identifiable, and may only be detectable during detailed examination of the organs, or in functional rather than physical measures of the organs.Says Barouki: “The effect of thalidomide is quite visible and it is a direct foetal toxicant; the new thing about DoHAD is that the exposure is during foetal development but is not visible clinically or toxicologically, but is going to be visible much later.”

This presents a major toxicological challenge. Histopathological assays after 90 days of exposure to relatively high doses of a chemical are shown to be ill-equipped for detecting functional changes which increase disease susceptibility; instead, long-term studies focusing on developmental exposures become more significant. High-dose studies are also of less use in the DoHAD context, as the effects of any epigenetic changes could easily be masked by the systemic effects of chemicals administered at high doses.

Anticipating long periods of latency and the subtle effects of exposure to chemicals at doses sufficient to induce epigenetic effects, which may be substantially different from the effects seen at doses which cause acute toxicity and have pronounced physiological effects, requires the development of toxicity tests which allow long-term effects to be revealed.

Although guideline toxicological studies have begun looking at developmental toxicity, with an early protocol first developed in 1995, regulatory studies for the most part look only at gross changes in organ weight and structure, they usually only last for 90 days rather than the full 2.5 year life-span of a rat, and there is little emphasis on functional changes.

The simple presence of epigenetic change is insufficient to predict disease (as the simple presence of a hormonal perturbation is insufficient to predict disease). “We are adding another level to what makes you susceptible to getting a clinical disease,” says Barouki. “Nutritional, infectious and chemical exposure mean you are born with certain susceptibilities… You can’t say that because everyone was exposed to BPA in the womb that everyone will be obese, but it increases susceptibility.”

The DoHAD White Paper says the crucial research challenge is to identify the epigenetic modifications that are predictive of long-term effects. What is not clear, however, is if this should be interpreted as a call to wait until the modifications have been identified before the possibility of epigenetic harm can be interpreted into chemicals policy.

This is because the project of gaining a complete picture of identifying the epigenetic alterations which affect health is a different project to protecting health while that project is incomplete. As things stand, we have strong reasons for believing the epigenetic changes are significant determinants of health; however, the set of changes which cause significant harm is unknown. Furthermore, with toxicological testing being conducted as it is, we are not likely to discover the effects of epigenetic change any time soon.

With epigenetics, we are therefore left in a similar bind as we are with endocrine disruptors. Some are going to be more harmful than others, and some we will not need to worry about. Right now, however, for the most part all we can do is identify compounds which make these epigenetic changes. Taking a lead from endocrine disruption, the precautionary thing to do would be to limit exposure to substances which have the potential to cause harm via an epigenetic mechanism, until those substances have been proven safe by test methods appropriate to demonstrating that safety.

Leave a Comment »

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.
Entries and comments feeds.

%d bloggers like this: