More on methylation:
For years now I have pondered epigenetics and read what came my way.  I thought it to be important.  I still do.  But it was like standing in a valley with low clouds.  I could see the edge of the valley but not upward.  Then this article came up.  (Robert J. Schmitz et al Patterns of Population Epigenetic Diversity NATURE vol. 495 no. 7440 March 14, 2013 page 193 and Steven Eichten and Justin Borevitz Methylation’s Mark on Inheritance page 181 in the same issue) Lo the mountains appear.  They are not yet climbed.  And I knew there had to be mountains out there somewhere, but getting a glimpse is a big help.

At the bottom there are genes written in the language of DNA.  Those genes can be converted into the relevant proteins by a mechanism that involves among other things different sorts of RNA.  (RNA is ribonucleic acid.  DNA is deoxyribonucleic acid.  If that’s not much help, then “ribose” is a kind of sugar that the acid contains, DNA is found in the nucleus of a cell while RNA is both in the nucleus and the surrounding cytoplasm and an acid is a chemical class that is sort of the opposite of a base.)  Information in the DNA is encoded in a series of bases.  They are cytosine, guanine, adenine and thymine (abbreviated C, G, A and T and for our purposes in a bit H will stand for anything other than G.)  The bases are situated in pairs C with G and A with T within the double helix structure we are all familiar with.  That’s right, the information is tucked away inside, where it’s got to be harder to access.  And then the whole thing is wrapped around a core of protein of a class called “histone.”  I can’t figure out why the whole thing doesn’t snarl up like a telephone cord … oh … You never heard of a telephone cord.  It was this stuff we used to tie up dinosaurs. 

The sequence of the bases is what encodes the information for making a protein, and it does a lot of other things, too.  A sequence that encodes a protein can reasonably be called a “gene.”  Progress is being made understanding the rest. 

So far you have a source of proteins.  I think of them as tools.  But owning a machine shop or a woodshop is not sufficient for making something.  You need a plan.  You need to know when to use the tools.  Anything, not itself a gene, that controls how genes are turned on and off is epigenetic.  There are various epigenetic mechanisms but one that is important is the process of “methylation.”  A methyl group is a carbon atom with three bonds to hydrogen atoms and its fourth bond attached to something else.  If that fourth thing is another hydrogen atom, then it is “methane” not a methyl group.  When a gene has methyl groups attached to it then its function is usually suppressed to a degree particularly if there is a string of such attachments.  

That is my best thumbnail description.  If you need to know for sure, consult a proper source.  If you know me to be wrong in some detail, forgive me.  But it was together pretty much my understanding of the mechanism.  And then the clouds rose a bit as I read the article.

One question is, “All right.  So just where is this methyl group attached to the gene?  Is it the bases, the outside helix, the histone core or something else?”  As it turns out, there is more than one possibility, but basically a methyl group attaches to a base C; just where on those bases the methyl groups attach I do not know, but I am sure somebody does.

There are three classes of methyl attachment.  The methyl group can attach to CG, CHG and CHH.  These are all produced differently and inherited differently; they are different systems and they are mediated by some sorts of RNA molecules.   In other words this is, if not a purpose built set of mechanism, it is a set of mechanisms that evolved in the fulfillment of some vital function. 

Methyl groups are, at least to a degree, inherited.  This is a key interest in the Schmitz paper and an analysis was done showing how the pattern of methylation varies among different geographical areas opening up the promise that the differences will be useful in figuring out which populations are more closely related to which.  There is plenty to work with.  The number of sporadic methylations generally number in the hundreds of thousands and they change from generation to generation much more than DNA base pair sequence changes.

Since the methyl groups change so much faster than DNA it has been proposed that methylation is sort of an evolutionary scratch pad.  An organism adapts to the environment with a quick but unstable methylation fix and maintains that fix with some loss due to regular changes in the methylation pattern until an appropriate DNA mutation comes along, which is stable and takes over.  I don’t think that because I need the methylation processes for my own purposes … I mean I am inclined to think they are there for another purpose; I mean they are there fulfilling a different function. 

The function I suspect they are involved with is the depression of fertility if an organism goes to many generations without a consanguineous mating.  I shall not rehash the logical necessity for that nor the evidence for it.  Most of it is located at and in my own work on the paper M.L. Herbert & M.G. Lewis Fluctuation of fertility with number in a real insect population and a virtual population African Entomology 21(1): 119–125 (2013) with further support from the obersvations in Thomas Cornultier et al Europe-wide Dampening of Population Cycles in Keystone Herbivores SCIENCE vol. 340 no. 6128 April 1, 2013 page 63. 

I think this is logical because the targeted genes are involved in making gametes and – this is key – if you strip all methylation off a plant by raising them from seeds grown in a demethylating  environment the plant does fine.  In fact you relieve any and all inbreeding depression that a plant has been subjected (Elizabeth Pennisi Epigenetics Linked to Inbreeding Depression SCIENCE vol. 333 no. 6049 September 16, 2011 page 1563 reviewing work by a team led by Philippine Vergeer, Hugens Building, Room HG 01.132, Radboud University Nijmegen, Molecular Ecology, Heyendaalseweg 135, 6525 AJ Nijmegen, The Netherlands)  In other words methylation does a plant no good at all.  It is there sabotaging the plant, depressing its fertility under conditions of inbreeding.

Now it is hard to understand why evolution should have bolted on – so to speak – a mechanism to decrease fertility under conditions of inbreeding.  As if the population didn’t have enough problems already for the goodness’ sake.  It is generally thought that inbreeding depression occurs because genetic diversity falls and recessive deleterious genes are expressed.  That really doesn’t wash; they should simply be selected out.  But if inbreeding depression were some inescapable consequence of the way genes are managed, it is understandable that it would persist.  But it isn’t genetically inescapable at all.  It has nothing to do with genes.  It’s purely the effect of methylation in a system the individual does not need.

But the species needs it.  My own simulations of fertility depression show that inbreeding depression is an inescapable consequence of any post-zygotic (happens after the egg is fertilized) mechanism for outbreeding depression.  The need for that is outlined in the link above and that need is absolute … all right … it is as important as evolution for any sexually reproducing organism. 

I had initially thought that there was probably just one such mechanism.  But life does tend to be redundant, and as the African Entomology paper lays out, there is evidence that in mammals there are at least two such mechanisms, one pre-zygotic (happens before the egg is fertilized) and the other post-zygotic.  And now, lo and behold, in plants there are three … at least three.

So we ought to look for more in mammals.  We only found evidence for one in fruit flies (that paper again) and two in mammals.

So how many mechanisms might there be in people?  At a stretch I can think of seven.

  1. Failure of male sexual development as evidenced by falling sperm counts, increased hypospadias, diminished average organ size, less interest in marrying and having babies and increased divorce rate once marriage is undertaken.  This would be post-zygotic.
  2. Failure of insufficiently related eggs and sperm to recognize and bind to each other as evidenced by the ability of door mouse sperm to recognize and bind to each other if they are from related males but not if they are from very unrelated males.  This would be pre-zygotic.
  3. Failure or at least delay in female sexual development as evidenced by women postponing having their babies.  I don’t know whether this counts as pre-zygotic or post.
  4. Some sort of mental thing involving a tendency toward magical thinking with people assuming that the demographic disaster now upon us will not have any serious consequences as evidenced by peoples ability to ignore this issue when I bring it up and by … well just consider entertainment; there is much more of magic, extraterrestrials and simply down right impossibility than there seemed to be a generation ago and the political scene seems to have gone utterly mad.  Post-zygotic.
  5. An aversion to marrying kin, which is obvious to the most casual observer and which effectively turns outbreeding into a runaway.  I’m not clear whether this is pre or post because as in 3) above the prime cause is post-zygotic but it entails a pre-zygotic final cause.
  6. Something you might call somatic opulence or the tendency for outbred individuals to be bigger, stronger, smarter (So they say, but I recently took an IQ test and scored 167, which is about ten points higher than I used to score so the tests may be getting easier rather than harder as we are told they are.), healthier and maybe better looking.  Again the mechanism is out to sabotage any population that has embarked upon outbreeding and making the outbred ones prettier might help.  Perhaps a continuation of the same process leads to obesity, hypertension, type 2 diabetes and hypercholesterolemia, any of which can kill you, but that was the general idea in the first place. 
  7. An increase in psychosis as evidenced by the fact that they say city dwellers have more than country dwellers and even those born in the countryside, where presumably there is greater consanguinity, have a lower rate even when they move into urban areas.  This also is very speculative because the diagnosis remains problematic (David Dobbs A Very Sad Story NATURE vol. 497 no. 7447 May 2, 2013 page 36 reviewing Gary Greenberg The Book of Woes, Blue Rider 2013) so statistics will remain shaky, and besides the numbers are probably too small to have a significant impact unless you count a mild form of people just acting like jerks, which from my own experience may be increasing but I have no numbers. 


Only the first three of those could remotely be related to plants and the others are dubious enough in people.  But some day we are going to know more. 

Today, as a first step, I would like to know for each of the three mechanisms what proportion of the methylated sites for each mechanism is inherited by the next generation. 

There have been 83 visitors over the past month.

Home page