Friday, November 2, 2012

Leafing through Nature or, How does development work?

We have some mature oak trees in our yard.  Last week, the winds and rain associated with the hurricane blew down googles of leaves, blanketing our yard (and we've already raked 3 times!).  But looking at the leaf-litter, I noticed the two leaves in the picture below.  Both were from the same tree.  Yep, that's a 1-foot ruler.


This isn't actually the tree (can't get a good camera angle on it; plus, we haven't had any snow yet), but the point can be made from this pic I grabbed from the web.  The arrows show, figuratively, that our two leaves, one normal size, one quite huge, fell from different parts of the tree (I had no way to know the exact relationships in the case of the real leaves).

Now, this tree started as a single cell, the embryo in some lucky acorn at least 50 years ago (this tree was already big when we moved here 30 years ago).  How can this same genome generate such massively different leaves?  What this means is that the leaf, the phenotype, cannot be predicted from the genome.

Note in the first picture that the pattern of the leaves is the same, basically the same toothy oaky structure of side growths, with a major vein going into each 'tooth' of the leaf.  But the details vary, which means that except for the basic structure, they too cannot be predicted from the genome.  None of the 23andLess promises, nor those of the Director of NIH, can change this.  Could the difference be due to environment? 

Genes and environments
Now we can at least ask about genes interacting with environments, such as temperature, water supply, sunlight and so on, that can differ in different parts of the tree, so some leaves get better living or growing conditions than others.  Of course, these cannot be predicted from the acorn, so personalized treenomic medicine isn't possible.  But worse, of course, at the acorn stage we don't have any way to know the future of dry and wet seasons, cloudiness, and the like.

And, sorry to say it, but in fact in this case it was not true that the large leaves were all on one side of the tree, and the small leaves from elsewhere on it.  Before they fell, the leaf sizes were distributed across the tree.  So nothing so simple, from what we know to measure at least, could account for this.

So in a way this makes such predictions essentially impossible, even in principle.  The solo genome of the acorn is not a predictor of more than the general features of the tree:  its overall branching and root structure, general shapes of its leaves, types of flowers and flowering time, nature of its bark--that sort of generic oakiness.  We could tell it will not grow to be a hemlock or a maple!

Explaining the similarities
But there are ways to account for the major aspects of similarity between these two very different leaves.  Those are the general principles of organization, gene action, and development, that we refer to as 'cooperation' among interacting factors--genes, signal molecules, cell surface receptors, cells, and so on.  In the book MT, we describe various processes that are driven by genes but not contained within any gene.  These processes of interaction involve timing effects, branching, and differentiation among cells.  That for example is how the veins branch and growth occurs along them in a leaf.

Now, there are still vastly different cell-distances in the two leaves shown above.  So how can the same genome, even with these process characteristics, generate them both?  One likely answer is that each leaf develops its basic structure--and the basic cell commitment to specific gene expression, when the baby leaf is very small.  Once these things are pre-set, rates of growth can, in principle at least, lead to different numbers of cells in each stage, between each branching event, etc.  Environmental factors can control these rates.

The idea: differentiate very early when the embryo is very small, and let the cells retain their gene-expression characteristics with them as they individually grow.

Still, it's very interesting how this can happen, the flexibility so to speak of what a single genome can do.  Or was it a single genome?

But do the leaves have the same genome, after all?
We say all the leaves in this one tree descend from a single acorn, a single cell with its copies of the Oak genome.  But in fact some mutations occur in every cell division. If they kill the cell, then it has no further bearing on what happens from then on.  But millions if not billions of cell divisions take place from the time the acorn begins to grow and the countless distant meristems that result and from which leaves and flowers grow.  That can be compared to the number of different parent-offspring transmissions that have occurred in the history of the human species.  It means that no two cells actually have the identical genome.  The meristem cells, long separated from common ancestry, are different in a general way as are any two humans.

Most of the genomes are very similar and most mutations have no real effect, but some may, or must.  So it could be that some leaves, or some branches, really are genetically different from each other in ways relevant to things like leaf size.  Maybe that's the reason (with or without additional environmental effects) for this huge difference. 

Personalized tree-nomic medicine, not!
 Could this be predicted in any specific way from the sequence of the genome that preceded the entire leaf?  It's doubtful, because it would be hard to get such a cell, and of course we'd have to do it for thousands of leaves on the same tree to understand the pattern, if any.  Personalized prediction of such small, later-age effects, can surely not be made from the acorn. Yet that is exactly what is promised for humans, relative to late-onset disease, being predicted from a person's 'genome'.

This is not wild speculation.  Your body differs locally in terms of where hairs are located, freckles, age spots, and many other details, for similar reasons--differences arising among your cells during your life.  Mostly they make no difference but if they lead a cell to start dividing too fast, they do matter, and it's called cancer.

So much to contemplate, even while you're just raking your lawn.....


  1. There are indeed significant genetic differences between different parts of the same tree as well as between trees that grew clonally by root budding. So in fact the leaf differences could be genetic. See:

    1. Thanks for pointing this out, which I hadn't seen. Somatic mutation is something I have written about, and I think is a greatly overlooked area.

      A tree, in this sense, is a population. Each branch may be analogous to a subpopulation, and if they have differential success, and shed differential numbers of acorns, or the acorns have relevantly different genotypes, then there can be evolution more complex than the usual model.

      Not only that, but the mutation may help the leaf but, by providing more shade from that side of the tree (say), disadvantage its own acorns' chances of surviving.

  2. I like your point re many genomes. I suppose there's not only sequence variation, but epigenetic also, etc. And thanks for writing about plants :)

    1. It would have to involve all these things, to the extent they happen in plants, and to the extent they are 'remembered'from cell to cell, or that their gene expression pattern leads to any information being differently transmitted to other nearby leaves via released signaling factors or via the vessels.