UK United Kingdom

Explainer: what is epigenetics?

The word epigenetics means things imposed “on top of genetics”. But what sort of things? Imagine a white mouse breeds with a black mouse – say you get three white babies and three black babies. That’s…

Genetics can explain a black or white cat in a litter, but what about a stripey cat? Enter epigenetics. Taylor Bennett

The word epigenetics means things imposed “on top of genetics”. But what sort of things?

Imagine a white mouse breeds with a black mouse – say you get three white babies and three black babies. That’s genetics. We can explain that. The black babies inherited a gene encoding the ability to make the pigment melanin, the others got a defective gene, so they are white.

But what if you get some stripey mice like zebras? How do you explain that? All parts of the mouse have the gene that can make black fur but the blackness only occurs in some places. Think of dalmatian dogs, or of giraffes or leopards.

Epigenetics is about turning genes on or off. It’s also about doing this stably; a leopard doesn’t change its spots even if it sheds its fur each year. So epigenetics is about stable cellular memory that persists after cell division and, in some cases, even through sexual reproduction.

Epigenetics, then, concerns the mechanisms that make organisms or parts of organisms look different, despite the fact they have the same genes and are in the same environment.

The beginning of epigenetics

Epigenetics is also involved in regulating genes at different times.

Episodes of disease are a relevant example. Recently, there was an interesting article on this site about herpes. That disease is caused by a virus that infects nerve cells.

When the virus is inactive and its genes are not expressed, there are no symptoms. But, for reasons that are not fully understood, the virus occasionally comes alive, its genes are expressed and symptoms occur.

This example is interesting because it reflects some of the earliest work on epigenetics. Researchers worked on viruses called bacteriophage, which infect bacteria. The viral DNA integrates into the bacterial genome.

If it is active, the virus expresses its genes, which make more viruses, and the cell bursts, releasing viral progeny. But sometimes, the viral genes remain unexpressed. The bacteria can divide and multiply for many generations, all the time carrying their deadly viral cargo, but the viral genes are not expressed.

Then, eventually, the virus may awake, express its genes, make progeny, and burst the cell.

These patterns of viral genes being on or off – sometimes for eons in terms of bacterial generations – fascinated early geneticists, such as Francois Jacob, Jacques Monod, Francis Crick and Mark Ptashne, who was once the head of biochemistry at Harvard.

The viral genes were not changing – the virus progeny were the same when they burst from the cell. But the expression of the genes was changing – off then on. There was some sort of change in state that could be inherited.

If the genes were not changing by mutation, then something on top of the genetics was changing – there was a change in epigenetic state.

Similarly, the white stripes in the zebra are not thought to be due to mutations. Something has settled down on top of the gene and silenced it – in some places or at some times and not others. What could this be?

Epigenetic regulators

The answer is a repressor, such as a DNA-binding protein. At Harvard, Mark Ptashne identified the first repressors.

Today Keith Shearwin and colleagues at the University of Adelaide continue this work and have mathematical models elegantly describing how complex control circuits of repressing or activating DNA-binding proteins with different binding affinities and half-lives can explain the stable epigenetic states of bacterial viruses.

Epigenetics helps explain wondrous things such as zebra’s stripes and butterflies' wings. Lincoln Smith

When humans and plants were studied, another mechanism associated with epigenetic control was observed. This new mechanism is fascinating and widespread. And it appears to be displacing the original broader definition.

In humans and plants, the major new epigenetic mechanism concerns chemical modifications of particular genes – typically the marking of certain genes with methyl groups.

The methyl groups are either attached to the DNA itself – this mostly leads to silencing of the gene – or to proteins that coat the DNA, called histones. The patterns of marks can be stable over time and the DNA marks at least can be replicated, so epigenetic states can affect parts of the organism for life or can even cross generations.

But there is some controversy here because it is not certain that histone marks can be directly inherited when a cell divides or when a new organism is formed via sexual reproduction. It seems more likely that DNA-binding proteins or functional RNAs resident in the cell are involved in re-establishing the epigenetic marks in each new daughter cell.

A recent commentary on epigenetics in the Proceedings of the National Academy of Sciences (PNAS) triggered two spirited responses, one in PNAS and one in Current Biology.

The reason was simple: enthusiasm for the histone code - the modification of proteins coating DNA - was obscuring the usual view that feedback loops generated by DNA-binding transcription factor proteins, and their allies functional RNAs, are the primary mediators of stable patterns of gene expression.

The important point that the review in PNAS failed to acknowledge was that epigenetics isn’t just about methylation, it’s more about control proteins and RNAs laying down the methylation marks.

In fact, it’s well known that epigenetic control occurs in organisms that have either no DNA methylation or no histones. So clearly methylation could not be an essential mediator of epigenetic control.

Epigenetics and Lamarckian inheritance

So why is epigenetics so exciting and controversial?

The field attracted public attention in part because it provided a mechanism for Lamarckian inheritance. The idea that we can learn from our environment and pass characteristics to our offspring has long been popular but the understanding of modern Darwinian mechanisms left little room for such ideas.

But in some special cases, acquired characteristics, such as viral infections or epigenetic marks imposed by DNA-binding proteins that respond to the environment, could be passed from one cell to its offspring. The total contributions of such mechanisms to human biology are not known. And while most researchers would consider them small, there are others who are invigorated by the possibilities.

Epigenetics and the inheritance of stable states is important in normal development, in disease, in ageing and in explaining wondrous things such as zebra’s stripes and butterflies' wings. Research in the broader world of epigenetics will provide fascinating and important insights for many years to come.

Join the conversation

15 Comments sorted by

  1. John Quintner

    logged in via Facebook

    Thanks you for a very timely contribution. A good example of how epigenetic effects on health have been played out is provided by the event in the Second World War known as the Dutch Hunger Winter. It lasted from early November 1994 to the late spring of 1945. A famine of major proportions in the Western Netherlands resulted from the blockade by the Germans. Over 20,000 people died. The dreadful privations created a remarkable scientific population study, which is ongoing, of the health of the survivors and of their offspring extending into future generations. An excellent discussion is contained within Nessa Carey's book "The Epigenetics Revolution (2011)".

    1. Stuart Purvis-Smith

      Clinical Cytogeneticist (retired)

      In reply to John Quintner

      John, thanks for that insight but I think that you may wish to check the dates of the famine. Did you mean 1944 to 1945?

      Another example of an epigenetic effect on human disease is the fact that chromosomes, or at least some parts of them have a "parent of origin" memory which is imprinted at conception. An abnormality such as a deletion of material in these "imprinted " parts can result in different congenital syndromes, depending on whether it is the maternal or paternal chromosome involved.

    2. Stuart Purvis-Smith

      Clinical Cytogeneticist (retired)

      In reply to Stuart Purvis-Smith

      I should have added that the two syndromes showing a "parent of origin" effect with which I am familiar are the Prader-Willi and Angelmann syndromes, both linked to chromosome 15.

  2. Peter Ormonde
    Peter Ormonde is a Friend of The Conversation.


    Heresy alert! Lamarck wasn't so totally wrong then? How excellent. I shall start chopping away at mouse tails immediately.

    Thanks for this wonderfully lucid explanation of such a complex and critically significant issue. We are not so much "genetically determined" - running on the iron rail tracks of inherited destiny - but are given options and opportunities (potentials) by our parents. A much more nuanced and realistic understanding I think.

    Great science.

    1. Stuart Purvis-Smith

      Clinical Cytogeneticist (retired)

      In reply to Peter Ormonde

      Indeed Peter, genetics today is much more nuanced and might I say, much more exciting. I go back to a time two score years ago when genetic determinism was much more in vogue and being at the top of the phylogenetic tree we "superior" humans were thought to have up to 100,000 genes. About 98% of our DNA was thought to be non-coding and referred to as "junk" DNA. Today the estimate of human gene number is about 20 - 25 thousand which puts us on a par with a lot of those small slimey creatures beneath us on the totem pole - it turns out however that most of our "junk" DNA is not junk but contains all sorts of controlling elements which allow for our nuanced developmental complexity.

      All very exciting stuff!

    2. Peter Ormonde
      Peter Ormonde is a Friend of The Conversation.


      In reply to Stuart Purvis-Smith

      I recall 25 years back being taught that 90 - 95% of our individual characterististics were "genetically determined". Everything from hair colour to who we would marry and our eventual occupations apparently.

      Didn't seem right at the time - being a mature aged student I had learned more respect for environmental factors and individual choice in life - but for some curious (undoubtably genetically determined) reason I always remember the wrong things... at least those things that are subsequently…

      Read more
  3. Max Walsh

    Education Consultant/Adviser

    Great Science indeed.
    These are the type of articles for which I "subscribe" to The Conversation and why I open it each day and scan the contents.
    Now I understand just a little bit more about Epigentics.

    1. Max Walsh

      Education Consultant/Adviser

      In reply to Max Walsh

      No editing facility.
      Even Education Advisers make typos. I meant, of course, "Epigenetics".

  4. Odile Pouliquen-Young

    Environmental Sustainability Manager at Curtin University

    What is the relationship between epigenetics and transposable elements?

    1. Peter Ormonde
      Peter Ormonde is a Friend of The Conversation.


      In reply to Odile Pouliquen-Young

      Not sure there is one directly - other than Transposable Elements (TEs) can become inserted into the chromosome and then are capable of being triggered by the specific conditions that lead to their expression.

      Here's an interesting article on TEs and how they work - very interesting bits of gear in plants. Some grass genomes for example comprise about 80% TE material that slip in and out like a revolving door.

      Epigenetics - as far as I understand it - is best conceived as a set of switches and triggers that lead to genes being expressed - particularly by environmental factors, chemicals, methylised proteins and the like.

      That help?

    2. Merlin Crossley

      Dean of Science and Professor of Molecular Biology at UNSW Australia

      In reply to Odile Pouliquen-Young

      Hi Odile and Peter,
      Peter is correct. There is no direct relationship between Transposable Elements (TE) and epigenetics. But it turns out that people interested in one tend to be interested in the other. If TEs are actively transcribed they can replicate themselves to new locations. If they land in an important gene they may interfere with the functioning of the cell or the whole organism. So most organisms use epigenetic silencing to control TEs. Lots of different epigenetic mechanisms are used…

      Read more
  5. Andrew Webster


    Firstly I have to say that i don;t know very much about this but have found the the relationship between epiegentics and neural development and the risk of developing mental health issues later on in life to be very interesting.

    How does the social environment ‘get into the mind’? Epigenetics at the intersection of social and psychiatric epidemiology

    Soc Sci Med. 2012 January ; 74(1)

    Epigenetic Modulation of Mood Disorders

    J Genet Syndr Gene Ther. ; 4(120):

    (Sorry but I haven;t figured out how to create hyperlinks)