One of the themes of our eponymous book, and of MT the blog as well, is the pervasiveness of cooperation in the nature of life. We view this as an antidote to what we believe is the post-Darwinian hyper-obsession with life-as-competition.
Everything we're learning about the relationship between genes and the biological traits they affect makes the number of contributing effects larger, more complex, and more specific to the species, population, individual, or even cell than simple models of genetics generally specify.
The degree and subtlety of this is shown by a recent paper by TR Mercer et al., current (Aug, 2011) issue of the prominent journal Cell. The authors show that even the tiny mitochondrial genome, a DNA ring molecule contained, one copy each, in the hundreds to thousands of mitochondria within a cell, are used in complex ways.
The human mtDNA is 16569 nucleotides long, and contains DNA sequence that codes for 37 genes of various types. The investigators identified mRNA copies of these genes in different cells from 16 different human tissues, to determine which mtDNA genes were being used in each tissue and and at what relative level (how many copies of each of the coded proteins were being made).
This figure shows on the left in red-shade, the presence and concentration of 11 tested genes (rows labeled on the right of the figure), in the 16 tested tissues (columns, labeled below the figure). You can see the patchwork of variable intensity and variable tissue expression. The rows and columns have been sorted to cluster like patterns. On the left is a tree-like diagram of tissue-expression similarities, with the rows of overall greatest similarity being shown near each other. Similarly, the cluster diagram shown atop the figure is of the overall relative similarity of expression levels. Clearly different genes are expressed in partly correlated ways in different tissues.
The part of the figure on the right shows, for the same tissues, the relative total contribution of genes expressed by the mtDNA compared to all genes expressed in the same cells. The highest fraction, 30%, the tall bar on the left, represents high relative mtDNA gene levels in heart tissues, while these genes have the least relative expression, only 5%, in lung tissue, the short bar on the right.
The size and gene content of the human mtDNA has long been known, but not the relative levels of the usage of these genes. This paper is mainly a presentation of the results (and an online data base one can use to get actual data), rather than an explanation of the pattern. But if mtDNA is largely related to energy usage, one might expect higher activity in more energy demanding tissues than in more passive types of cells. Actually, this is not what the data show: the brain, an energy sink if there ever was one, is dead center in its mtDNA usage compared to the other tissues. This seems an important fact, or perhaps a warning about the results, and the authors seem rather blithely to skip over this fact in their brief description of the relative usage data.
The point here, however, is not to quibble about such details, because if there are mistakes or inaccuracies and the like, future work will figure them out. Our point is that mtDNA, the simplest part of our genomes in a sense, add yet another level of complexity to the connections between human phenotypes and their underlying genotypes. No new principles of nature are revealed, but there is another tack in the lid of the simple-causation box.
The mitochondrion arrived 1-2 billion years ago as an invader of organisms' cells, and has become a vital and permanent part of the genome, even if not directly in chromosomes in the nucleus (mitochondria are in the cytoplasm, not the nucleus). Their coded proteins interact with proteins coded by genes in the nucleus, to make the cell work, and in particular to providing the cell its chemical energy. These functions are very ancient and stable. Nonetheless, this little circle of friends, of interacting genes, is used even among the tissues of a given individual, in many different ways.
The correlations among expression levels shows an example of cooperation that has been established by evolution, that varies even among the tissues in our own bodies. If all you care about is the net function--how lungs allow certain levels of activity, or how fast hearts beat under various circumstances, then these details simply show one of the levels of underlying complexity in attaining that result, and will be of little use to you. But it is of use to know that this complexity is involved because complexity also implies some redundancy or buffering.
These investigators did not study a diversity of conditions to see how stable the expression values they obtained are, nor how similar to other species. That kind of information may be hard to obtain, but is for the future, and we are very likely to find that even within a cell there will be substantial variation.
But at least this paper beings to reveal the revelry that takes place even within this little circle of friends....