In article <Pine.SGI.3.95.970808001626.26526C-100000 at umbc9.umbc.edu>, david
ford <dford3 at gl.umbc.edu> wrote:
>> "There is no obvious genetic barrier to such changes." In the
> transformation of a land animal into a whale, approximately how many new
> genes would you guesstimate were required? Some areas where new genes
> perhaps would be required are in the appearance of the tail fluke, for
> any new enzymes or proteins (e.g., perhaps for the whale's skin), for
> the new structure that would allow the baby whale to drink milk
> underwater, for the cap that is around the nipple, for the organ making
> spermaceti, for the muscles and flaps that allow the blowhole to be
> closed off when the whale dives, for the baleen filtration system in
> baleen whales, for making the mother's milk the composition that it is,
> and for the melon.
>> Geneticists, biologists, please help us out. Jim and I know precious
> little about genetics. With genes coding for structures, is it
> necessary that every nucleotide be present, or else the structure coded
> for develops malformed? For genes coding for proteins, must all the
> nucleotides be exactly right? How many triplet codons typically make up
> a structural gene, and how many typically make up a gene coding for a
> protein?
You do not need "new" genes to convert e.g. a leg into a fin. Developmental
genes work in a hierarchical fashion, so that a fertilized egg (that is, an
embryo) is first divided into broad segments by genes that are expressed
(make proteins) within those segments. The expression of those genes
overlaps and the patterns of overlap serve to turn on other genes in finer
segmental patterns, then those segments aquire different fates as yet other
genes are turned on in those various segments. Finally certain genes cause
differential patterns and rates of cell division that give rise to limbs,
head structures, etc. To turn a leg into a fin would not so much involve the
creation (heh heh!) of entirely new genes, as a difference in timing of the
turning on of various genes, resulting in different patterns of growth. In
fact, in a structure like the human hand, it is cell death in the skin
between the fingers that gives separate fingers rather than a fin like
structure with webbing. Knock out the genes that control this process, or
change their timing significantly, and you'll get something more like a fin
than a hand. Change the timing of expression of genes that control
elongation of the arm, and that fin may remain closer to the body.
It is the genes expressed earlier that control the timing and pattern of
expression and it is changes in these genes that probably have a greater
effect in changing morphology (though changes in downstream effector genes
undoubtedly play a role as well). These timing/pattern control genes work
by producing proteins that bind to regulatory regions (promoters) of
downstream genes, either turning them on or keeping them off. Slight
changes in these proteins (transcription factors) can change the strength of
binding, and the ability to regulate downstream genes. So gene changes
required for changing morphologies are theoretically not that great, but
what is important is keeping *all* the genes coordinated so you end up with
a intact animal, and not just a blob of tissue.
See The making of a fly : the genetics of animal design /, Peter A.
Lawrence. Oxford [England] ; Cambridge, Mass., USA :
Blackwell Science, 1992 (1995 printing) for information on what is known
about these processes in the best understood example, the fruit fly
Drosophila melanogaster.
As for lenght of genes, typically they are on the order of a few thousand
nucleotides, plus anywhere from one to ten-thousand more forming the
regulatory region where transcription factors bind to control expression.
