Monday, January 15, 2007
Tandem Repeats, Growth and Form, Cellular Zip Codes, and Molecular Tags
This also plays into observations made by D'Arcy Thompson, whose book On Growth and Form showed in many cases widely diverse species where features are separable by fairly simple mathematical deformation equations. I wonder if many of those might be specified within the genome by tandem repeats. The cis-regulatory development "kernels" perhaps provide the gross template, while the tandem repeats specify parameters to deformation equations to apply to the kernels.
The evolution of tandem repeats is very highly regular -- they seem to be tightly controlled by the cell to be almost entirely copy-number variations. This lends a lot of weight to the idea that they are indeed tuning knobs on the genome.
On a slightly different note, Rinn et all recently gave evidence that cellular knowledge of position within the body may rest with specific altered gene expression (Hat Tip to Creation Safaris). That is, cells know what kind of tissue and where they are in the body based on gene expression. I believe they were suggesting an epigenetic mechanism for this sort of differentiation (methylation possibly? Or simply a signalling switch left open/closed? chromatin modification? It would be even more interesting if it were a mutation.). The differential expressions they were able to detect were primarily for proximal/distal, anterior/posterior, and dermal/nondermal. You can see the gene expression maps here.
The interesting thing is that (a) there are positional determinants, but (b) the categories they were able to detect are fairly broad, while in ontogeny cells are very finely position-specific. This could mean either that their sample set was not large enough to determine the fine-grained expression pattern differences, they were looking in the wrong place for it, or perhaps that adults cells don't have nearly the same need for position-based knowledge as do embyos.
Anyway, we keep on finding more tagging mechanisms in biochemistry. Someday perhaps I should make a list! That would be fun.