In the months before the publication of the first draft of the human genome at the turn of the millennium, those in the know started a sweepstake. They were gambling on how many genes were contained within our genome in a little competition that came to be known as Gene Sweep. The cost of entry started out at $1 back in 2000 and went up to $5 in February 2001. By 2002 the cost was $20, the increase in price reflecting the increase in knowledge available on which to base your guess. There were over 400 entrants from the genome community with the smart money being in the region of 45,000-75,000 genes, the average was 61,710 genes.
In summer 2003 the deadline for announcing a winner arrived. Lee Rowen, who ran a sequencing project at the Institute for Systems Biology in Seattle, picked up the $1,500 dollar jackpot with her guess of 25,947 genes. She won not so much because she was correct but because she was closest by dint of the fact that she happened to have the lowest guess. The reality was that everyone was wrong, and I do mean everyone. At that time the official number of genes stood at 24,847, this has since been revised down to a figure around 22,000. This is a fraction of what the greatest minds in genomics thought we might have.
Why, then, were so many clever people so wrong? In fairness it was a reasonable assumption given the data we had at the time. By 2001 we had the genomes published of some fairly basic organisms like Caenorhabditis elegans (nematode worm; 20,000 genes) and Drosophila melanogaster (fruit fly; 15,600 genes) and we knew that we were more complicated than they were; not just on a macro level but on a cellular level, too. The false assumption in the reasoning, though, was that one gene would encode one protein which would have one function. This couldn’t be further from the truth.
We now know that one gene can easily encode for half a dozen different versions of the same gene, known as transcripts. We know that each protein can have a multitude of functions dependent upon the cellular context at the time. On top of that there are elaborate genetic regulatory mechanisms that can switch genes on or off, or switch between different transcripts. There is a whole extra layer of regulation that can target the intermediate products of a gene for destruction even if a gene is switched on. We may only have 22,000 genes, then, the same as the lowly nematode; but on top of that we have layer upon layer of regulatory elements that give us an order of magnitude more complexity to play with.
The reason I mention all this is due to an open access paper recently published in Cell Metabolism. In it the authors discuss how the effects of stress on fathers can cause epigenetic changes in their sons that cause an increase in blood sugar levels. Epigenetics was another one of those layers of complex regulatory mechanisms that were unforeseen by those involved in Gene Sweep. It was known at the time that environmental factors were capable of modifying the chemical signals of the genome. These chemical signals had the power to switch genes on or off by making the DNA more or less physically accessible to the molecular machinery that expresses it.
What wasn’t known was that some of these changes can be inherited, indeed, the perceived wisdom was that all epigenetic changes are reset back to default during gametogenesis (production of eggs and sperm). There would be a good reason for this: resetting back to default would allow offspring the flexibility to deal with their environments instead of being hindered by their parent’s epigenetic adaptations. In recent years, however, we have learnt that about 5% of the epigenome is resistant to being reset and therefore provides a portal through which traits can be inherited from the previous generation.
I have mentioned epigenetics in previous posts and commentors have effectively accused me of being an epigenetics denier. To clarify my position, I by no means claim that there is no such thing as epigenetics, I’m a geneticist after all and have long found the whole topic fascinating. What I do reject is anyone who claims that they can magically exert control over the epigenome using some gimmicky nonsense – I’m looking at you Deepak Chopra.
There are at least two or three examples of epigenetic effects that are extremely well documented and understood. There are other cases where there does seem to be effects at work, normally induced by extreme food deprivation, massive stress or, as ever, smoking.
The paper at hand uses a model of stress in mice. It’s not all that pleasant but basically they restrained a male mouse every day to stress it out. The mouse was then allowed to mate and they found that the male offspring but not the female offspring were hyperglycaemic. Further work revealed the mechanism which is, it has to be said, quite complicated.
The hyperglycaemia, an excess of glucose in the bloodstream often associated with diabetes, was caused by an increase of a process called gluconeogenesis in the liver. This mechanism was in overdrive due to increased expression of a gene called PEPCK. PEPCK is, in part, regulated by a micro-RNA. miRNAs are one of the layers of regulation we didn’t know about 20 years ago. They are short lengths of RNA that can latch on to a gene product and thereby mark it for destruction. The miRNA that regulates PEPCK is itself regulated by another gene called Sfmbt2. The researchers found that there was en epigenetic marker on the promotor of that gene that set this whole cascade off. I hope by now you are starting to get a flavour for how complex the modern world of genetics is and why the Gene Sweep guesses were so wide of the mark.
To summarise then: extreme stress resulted in an epigenetic modification that led to increased amounts of sugar being produced by the stressed out mouse. This makes sense, the body is making fuel available for a flight or fight response. The problem is that this modification is not being reset in the next generation where food is plentiful and there is already enough sugar in the system. With sugar production in overdrive the result is hyperglycaemia and the risk of diabetes.
We are only in the very early years of getting a handle on epigenetics. It will undoubtedly reveal some very interesting effects and, who knows, maybe we will find a way of altering these in people and improving public health. Bare in mind, though, that prevention is always better than cure and therefore we need to stop these harmful effects being passed on in the first place. How do we do that? I’m afraid it’s the same old boring advice as always: don’t smoke, eat a healthy diet and try to avoid stress where possible. It sucks, I know, but those are the facts we can be scientifically sure of. Anyone who says otherwise, especially if they have written a book on the topic, is likely leading you merry dance.