|MadSci Network: Genetics|
As all cells contain the same DNA, but are differentiated, different proteins must be expressed in the different types of cells. Does this have anything to do with the parts of DNA which are not expressed...
I truncated your question here, because the question as stated above gets at the heart of Molecular Genetics. The expression of different proteins in different cell types has everything to do with the expression of different parts of the DNA. Essentially, the reason a liver cell doesn't express muscle-specific proteins is not just because those genes are turned off, but also because many of those muscle-specific genes cannot be turned on in liver cells. The accessibility of a gene to the protein-making (transcriptional) machinery of the cell is controlled by several factors, including DNA methylation and various degrees of DNA packaging. DNA methylation is simply the addition of methyl groups to cytosines along the chromosome; thereby masking these bases so that several of the proteins required for activating a gene cannot recognize the DNA and so, do not activate the gene. DNA methylation has been shown to regulate genes required for both differentiation and development.
There are a couple of degrees of DNA packaging that determine the ability of genes to be expressed. The lowest degree of packaging is binding of nucleosomes to the DNA. A major mechanism of gene regulation in all eukaryotic cells appears to be through the loosening and rearrangement of nucleosomes around a given gene. This is accomplished through transcription factors (HAT's) that add acetyl groups to the nucleosomes, loosening their local attachment to the DNA. Conversely, there are transcriptional repressors (HDAC's) that deacetylate the nucleosomes, tightening their local association with the DNA. A higher degree of DNA packaging involves the condensation of chromosomes into chromatin. During mitosis, all of the chromosomal DNA is condensed into chromatin, containing more protein than anything else, so that the chromosomes can be easily segregated. After mitosis, the chromatin decondense to allow usage of the DNA: however, this decondensation is never complete, and much of the genetic material remains condensed and inaccesible. This non-decondensed DNA is called heterochromatin to distinguish it from the euchromatin in which genes are accesible. So genes contained with heterochromatin are not expressed by the cell, but the pattern of heterochromatin changes from cell type to cell type, so that a gene that might be necessary for one cell cannot be used by another cell. By turning off portions of the genome, chromatin structure can essentially give different cell types different genomes from which to operate.
...called "junk DNA"? If not, what do you think junk DNA is used for? Thank you!
Well, this is a different question altogether. First of all, the term "junk DNA" was coined when molecular geneticists, who had spent half a century studying bacterial DNA, first examined eukaryotic chromosomes. They found out that unlike bacteria, eukaryotic "genes" (as in stretches of DNA that code for proteins) made up only a small percentage of the total DNA in the chromosomes. They figured, incorrectly, that the rest of the DNA must be "junk". In fact, one could argue that if these early studies had been carried out in eukaryotes, the term "junk DNA" would have never made it into the literature. So then what are the components of a eukaryotic genome? Each eukaryotic gene is broken up into fragments (exons) that are separated by non-coding DNA (introns). After transcription, these intervening sequences are spliced out and the fragments joined to reform the open reading frame of the gene. These intervening sequences take up much more space than the gene fragments themselves, so that a gene that might be a couple hundred basepairs long in bacteria would be several thousand bases long in eukaryotes. On top of this, the expression of genes in eukaryotes is controlled by DNA elements that lie upstream (and sometimes downstream as well as within) of the gene as far as tens of thousands of basepairs. The spacing of these control regions is very important to the proper expression of genes, since clustering genes together often results in one control region influencing more than one gene.
In humans, most genes, including their control regions, are upto 20 kb (kilobases: thousand basepairs) long. With approximately 40,000 genes in the human genome, this gives about 0.8 billion basepairs worth of eukaryotic genes: over a quarter of the around 3 billion basepairs of the human genome. On top of this, a good percentage of each chromosome is devoted to the centromere and telomeres that are required for mitosis and replication, as well as multiple scaffold attachment sites that are necessary for complete condensation of the chromatin. So, besides genes, much of the chromosomal DNA is reserved for the mechanical work of maintaining, replicating, and segregating chromosomes in growing cells. Lastly, over 10% of the human genome is composed of millions of copies of about 3 or 4 transposable elements: short stretches of DNA that can, under the right circumstances, copy themselves to other places in the genome. In humans, these transposable elements do not contain genes; only the recognition sequence to be transposed around the DNA. Outside of all this, the leftover DNA appears to be simply a spacer to separate strong genes and limit heterochromatin. Most important is that all of these spacer, between genes as well as the introns within genes, allow the recombination that occurs during meiosis to shuffle our genomes (and sometimes our exons) without actually changing the sequences of the genes themselves.
Try the links in the MadSci Library for more information on Genetics.