Feb 4, 2005; Biology 2005 Albert Harris   1) All our cells contain all our genes. (except for a few special cases, which will be listed and discussed below) 2) Cell differentiation is the result of "turning on" (making the proteins coded for) particular subsets of our total <50,000 genes. 3) For each of our bodies <250 differentiated cell types, a certain list of genes are "turned on", and all the rest are "turned off" (don't transcribe RNA) Extracting m-RNAs from each differentiated cell type, and letting them reanneal onto "microarrays" should eventually allow people to make lists of all the genes that are expressed for each differentiated cell type; that isn't expected to be a "breakthrough" 4) In plant cells, differentiation tends to be much more reversible than in animal cells. (so you can plant cuttings in the ground, and they make roots) The causes of this resistance to "de-differentiation" are not understood very well; Nor do people even study them! If you could cause animal cells to dedifferentiate, then you wouldn't need to get "stem cells" from human embryos. Another phenomenon that almost nobody studies is why cells never differentiate into a combination of two different cell types! Cancer cells often do something like this; But no one asks why 5) The transcription of each particular gene is "turned on" by the binding of special proteins (called "transcription factors" to DNA wherever it has certain base sequences. Transcription factor proteins always have a concentration of arginines and lysines (with very basic amines on side chains) at the part that touches the DNA, and they also have some mechanism for holding this region out on some kind of extension in the three-dimensional structure of the protein. Later, we will learn the names of lots of transcription factors. I promise you will get totally sick of transcription factors. Last spring, Jason Lieb kindly invited me to have supper with a couple of scientists whose research concentrates on the binding of transcription factors to DNA, and it turns out nobody can yet predict what amino acid sequence a binding site needs to have in order to bind to a given DNA base sequence! Computers can't either. It worries me that somebody could just make random co-polymers heavy in lysine and arginine, and put these in cells, and maybe turn on random combinations of genes. Who knows! Or if you made artificial m-RNA containing almost all guanines and adenines, then you get very many lysines and arginines! (please don't memorize the genetic code; but you might try the "genetic code JAVA program" on my web site.) 6) Transcription factors bind to special DNA regions just "upstream" of the parts of genes that code for the proteins' amino acid sequences. These are called "promoters" or "promoter regions"

 

back to syllabus

back to index page