The high levels of DNA methylation at repetitive elements in plants are very similar to the pattern at repetitive elements in the chromatin of higher animals such as mammals. By contrast, the methylation in the bodies of widely expressed genes is much more like that seen in honeybees (which don’t methylate their repetitive elements). This doesn’t mean that plants are some strange epigenetic hybrid of insects and mammals. Instead, it suggests that evolution has a limited set of raw materials, but isn’t too obsessive about how it uses them.
Prediction is very difficult, especially about the future.
Niels Bohr
One of the most exciting things about epigenetics is the fact that in some ways it’s very accessible to non-specialists. We can’t all have access to the latest experimental techniques, so not all of us will unravel the chromatin changes that underlie epigenetic events. But all of us can examine the world around us and make predictions. All we need to do is look to see if a phenomenon meets the two most essential criteria in epigenetics. By doing this, we can view the natural world, including humans, in a completely new light. These two criteria are the ones we have returned to over and over again throughout this book. A phenomenon is likely to be influenced by epigenetic alterations in DNA and its accompanying proteins if one or both of the following conditions are met:
1. Two things are genetically identical, but phenotypically variable;
2. An organism continues to be influenced by an event long after this initiating event has occurred.
We always have to apply a common sense filter, of course. If someone loses their leg in a motorbike accident, the fact that they are still minus a leg twenty years later doesn’t mean that we can invoke an epigenetic mechanism. On the other hand, that person may continue to have the sensation that they have both legs. This phantom limb syndrome might well be influenced by programmed gene expression patterns in the central nervous system that are maintained in part by epigenetic modifications.
We are sometimes so overwhelmed by the technologies used in modern biology that we forget how much we can learn just by looking thoughtfully. For example, we don’t always need sophisticated laboratory equipment to determine if two phenotypically different things are genetically identical. Here are a couple of examples with which we are all familiar. Maggots turn into flies and caterpillars turn into butterflies. An individual maggot and the adult fly into which it finally develops must have the same genetic code. It’s not as if a maggot can request a new genome as it metamorphoses. So, the maggot and the fly use the same genome in completely different ways. The painted lady caterpillar has interesting spikes all over its body and is fairly dull in colour. Like a maggot, it has no wings. The painted lady butterfly is a beautiful creature, with enormous wings coloured black and vivid orange, and it has no big spikes on its body. Once again, an individual caterpillar and the butterfly into which it develops must have the exact same DNA script. But the final productions from these scripts differ enormously. We can hypothesise that this is likely to involve epigenetic events.
The stoat
Mustela ermine
is found in Europe and North America. It’s an athletic little predator in the weasel family, and in summer the fur on its back is a warm brown and its front is a creamy white. In cold climates its coat turns almost completely white all over in the winter, except for the tip of its tail, which remains black. With the arrival of spring, the stoat reverts to its summer colours. We know that there are hormonal effects that are required for this seasonal change in coat colour. It’s pretty reasonable to hypothesise that these influence the relevant expression of coat colour genes by methods which include epigenetic modifications to chromatin.
In mammals, there’s usually a clear genetic reason why males are males and females are females. A functional Y chromosome leads to the male phenotype. In lots of reptile species, including crocodiles and alligators, the two sexes are genetically identical. You can’t predict the sex of a crocodile from its chromosomes. The sex of a crocodile or an alligator depends on the temperature during critical stages in the development of the egg – the same blueprint can be used to create either a male or a female croc
1
. We know that hormonal signalling is involved in this process. There hasn’t been much investigation of whether or not epigenetic modifications play a role in establishing or stabilising the gender-specific patterns of gene expression, but it seems likely.
Understanding the mechanisms of sex determination in crocodiles and their relatives may become a rather important conservation issue in the near future. The global shift in temperatures due to climate change could have adverse consequences for these reptiles, if the populations become very skewed in favour of one sex over another. Some authors have even speculated that such an effect may have contributed to the extinction of the dinosaurs
2
.
The ideas above are quite straightforward, easily testable hypotheses. We can generate a lot more like these by simple observation. It’s a lot riskier to make broad claims about what other more general developments we might expect to see in epigenetic research. The field is still young, and moving in all sorts of unexpected directions. But let’s render ourselves hostages to fortune, and make a few predictions anyway.
We’ll start with a fairly specific one. By 2016 at least one Nobel Prize for Physiology or Medicine will have been awarded to some leading workers in this field. The question is to whom, because there are plenty of worthy candidates.
For many people in the field it’s extraordinary that this hasn’t yet been awarded to Mary Lyon for her remarkably prescient work on X inactivation. Although her key papers that laid the conceptual framework for X inactivation didn’t contain much original experimental data, this is also true of James Watson and Francis Crick’s original paper on the structure of DNA
3
. It’s always tempting to speculate the lack of a Nobel Prize might be down to gender, but that’s partly because of a myth that has grown up around Rosalind Franklin. She was the X-ray crystallographer whose data were essential for the development of the Watson-Crick model of DNA. When the Nobel Prize was awarded to Watson and Crick in 1962 it was also awarded to Rosalind Franklin’s lab head, Professor Maurice Wilkins from Kings College, London. But Rosalind Franklin didn’t miss out on the prize because she was a woman. She missed out because she had, tragically, died of ovarian cancer at the age of 37, and the Nobel Prize is never awarded posthumously.
Bruce Cattanach is a scientist we have met before in these pages. In addition to his work on parent-of-origin effects, he also performed some of the key early experimental studies on the molecular mechanisms behind X inactivation
4
. He would be considered a worthy co-recipient with Mary Lyon by most researchers. Mary Lyon and Bruce Cattanach performed much of their seminal research in the 1960s and are long-since retired. However, Robert Edwards, the pioneer of in vitro fertilisation, received the 2010 Nobel Prize in his mid-eighties, so there is still time and a little hope left for Professors Lyon and Cattanach.
The work of John Gurdon and Shinya Yamanaka on cellular reprogramming has revolutionised our understanding of how cell fate is controlled, and they must be hot favourites for a trip to Stockholm soon. A slightly less mainstream but appealing combination would be Azim Surani and Emma Whitelaw. Together their work has been seminal in demonstrating not only how the epigenome is usually reset in sexual reproduction, but also how this process is occasionally subverted to allow the inheritance of acquired characteristics. David Allis has led the field in the study of epigenetic modifications to histones, and must also be an attractive choice, possibly in combination with some of the leading lights in DNA methylation, especially Adrian Bird and Peter Jones.
Peter Jones has been a pioneer in the development of epigenetic therapies and this is another growth area for epigenetics. Histone deacetylase inhibitors and DNA methyltransferase inhibitors have been in the vanguard of these approaches. The vast majority of clinical trials with these compounds have been in cancer, but this is starting to change. An inhibitor of the sirtuin class of histone deacetylases is in early clinical trials for Huntington’s disease, the devastating inherited neurodegenerative disorder
5
. The greatest excitement, for both cancer and non-oncology conditions, is currently centred around the development of drugs that inhibit more focused epigenetic enzymes. These include enzymes that change just one modification at one specific amino acid position on histone proteins. Hundreds of millions of dollars are being invested worldwide in this sphere, either in new biotech companies, or by the pharmaceutical giants. We are likely to see new drugs from these efforts enter clinical trials for cancer in the next five years, and clinical trials for other less immediately life-threatening conditions within a decade
6
.
Our increased understanding of epigenetics, and especially of transgenerational inheritance, may also create problems in drug discovery, as well as opportunities. If we create new drugs that interfere with epigenetic processes, what if these drugs also affect the reprogramming that normally occurs during the production of germ cells? This could theoretically result in physiological changes that don’t just affect the person who was treated, but also their children or grandchildren. We maybe shouldn’t even restrict our concerns to chemicals that specifically target epigenetic enzymes. As we saw in
Chapter 8
, the environmental pollutant vinclozolin can affect rodents for many generations. If the authorities that regulate the licensing of new drugs begin to insist on transgenerational studies, this will add enormously to the cost and complexity of developing new drugs.
At first glance, this might seem perfectly reasonable; after all, we want drugs to be as safe as possible. But what happens to all the patients who desperately need new drugs to save them from life-threatening diseases, or who need better drugs so that they can live healthy and dignified lives free of pain and disability? The longer it takes to get new drugs to the market, the longer those patients suffer. It’s going to be very interesting to see how drug companies, regulators and patient advocacy groups deal with this issue over the next ten or fifteen years.
Transgenerational effects of epigenetic changes may be one of the areas with the greatest impact on human health over the coming decades, not because of drugs or pollutants but because of food and nutrition. We started this journey into the epigenetic landscape by looking at the Dutch Hunger Winter. This had consequences not just for those who lived through it but for their descendants. We are in the grip of a global obesity epidemic. Even if our societies manage to get control of this (and very few western cultures show many signs of doing so) we may already have generated a less than optimal epigenetic legacy for our children and grandchildren.
Nutrition in general is one area where we can predict epigenetics will come to the fore in the next ten years. Here are just a few examples of what we know at the moment.
Folic acid is one of the supplements recommended for pregnant women. Increasing the supply of folic acid in the very early stages of pregnancy has been a public health triumph, as it has led to a major drop in the incidence of spina bifida in newborns
7
. Folic acid is required for the production of a chemical called SAM (S-adenosyl methionine). SAM is the molecule that donates the methyl group when DNA methyltransferases modify DNA. If baby rats are fed a diet that is low in folic acid, they develop abnormal regulation of imprinted regions of the genome
8
. We are only just beginning to unravel how many of the beneficial effects of folic acid may be mediated through epigenetic mechanisms.
Histone deacetylase inhibitors in our diets may also play useful roles in preventing cancer and possibly other disorders. The data are relatively speculative at the moment. Sodium butyrate in cheese, sulphoraphane in broccoli and diallyl disulphide in garlic are all weak inhibitors of histone deacetylases. Researchers have hypothesised that the release of these compounds from food during digestion may help to modulate gene expression and cell proliferation in the gut
9
. In theory, this could lower the risk of developing cancerous changes in the colon. The bacteria in our intestines also naturally produce butyrate from the breakdown of foodstuffs
10
, especially plant-derived materials, which is another good reason to eat our greens.
There’s a speculative but fascinating case study from Iceland on how diet may epigenetically influence a disease. It concerns a rare genetic disease called hereditary cystatin C amyloid angiopathy, which causes premature death through strokes. In the Icelandic families in which some people suffer from the disease, the patients carry a particular mutation in the key gene. Because of the relatively isolated nature of Icelandic societies, and the country’s excellent record keeping, researchers were able to trace this disease back through the affected families. What they found was quite remarkable. Until about 1820, people with this mutation lived until around the age of 60 before they succumbed to the disease. Between 1820 and 1900, the life expectancy for those with the same disorder dropped to about 30 years of age, which is where it has remained. The scientists speculated in their original paper that an environmental change in the period from 1820 onwards altered the way that cells respond to and control the effects of the mutation
11
.