Inside the cell, DNA exists in association with various proteins.  This DNA-protein complex is known as chromatin.

Based on the structure of DNA, each base pair is about 0.34 nm in length. A kilobase (that is, 10³ base pairs) of DNA is therefore about 0.34 µm in length.   A bacterium, like E. coli, has ~ 3 x 10⁶ base pairs of DNA – that is a DNA molecule almost a millimeter in length. 

Regions of chromatin that are packaged so as to be inaccessible to regulatory proteins are known as heterochromatin.

Chromatin that is accessible and transcriptionally active is known as euchromatin.

Some regions of chromatin are folded away more or less permanently; they are unfolded only during DNA replication.  These are known as constitutive heterochromatin.  Other regions can be unpacked in different cell types and are known as facultative heterochromatin.

A particularly dramatic example of this process occurs in female mammals.  One of the two X chromosomes is packed into a heterochromatic state, known as a Barr body.

Most of the genes on this heterochromatic X chromosome are not expressed; so even though females (XX) have twice as many of the genes located on the X chromosome as males (XY), both make about the same amount of gene product.

Transcription factors also influence the proteins that act on chromatin to make the DNA within it accessible or inaccessible.   This regulation of chromatin packing can act on blocks of DNA, regulating all of the genes within the affected region. 

In the end, the gene regulatory elements within the DNA compete with one another for the binding of DNA-dependent, RNA polymerases.  Genes that assemble the most efficient promoters are the most effective at loading RNA polymerase molecules. These genes will produce more copies of the RNA that they encode.

Networks:  A cell's behavior will be determined in large part by which of its genes are actively expressed. 

How can a cell determine that the "right" set of genes is activated for a particular situation? Clearly feed back systems are needed.

Genes that need to be expressed together to perform a function can be co-regulated by using similar sets of regulatory sequences.

← Multicomponent loops: Transcription factors act on regulatory sequences in a close loop arrangement.

Regulatory chains:→ In this arrangement, a transcription factor binds to regulatory sequences in another gene and induces expression of a second transcription factor, which in turn binds to regulatory sequences in a third gene, etc. 

The chain ends with the production of some non-transcription factor products (if not, it is probably a multicomponent loop). 

← Single and multiple input modules: A transcription factor binds to sequences in a number of genes, regulating their coordinated expression.  In most cases, sets of target genes are regulated by sets of transcription factors that bind in concert. 

Regulation of regulatory networks:→  Transcription factors are proteins, and so their stability and activities can be regulated by interactions with other proteins, allosteric factors, and post-translational modifications.  It is through such interactions that signals from outside the cell can alter the patterns of gene expression.

At the same time, internal signals, generated by metabolic processes, can also feedback to the gene regulatory system in order to maintain the homeostatic state. 

In fact, such feedback is critical to maintaining regulatory networks – inhibiting them from spiraling out of control – which would lead to death.  

Genome and gene duplication:

An obvious question is, how can biological systems increase their complexity?  Where do new genes and new activities come from? 

One mechanism is by duplicating pre-existing genes [link].   Typically selection will act to maintain the function of one copy of the gene (which was important in the organism before the duplication and remains so afterward).  The second copy is then freer to change.  Mutations may lead to the loss of a functional gene product, or a gene product with altered properties. 

There a number of different types of duplication events, which range from the duplication of a small region of a specific chromosome to the duplication of the entire genome (we will not go into the molecular details involved).   The advent of genomic sequencing and the comparison of genomes from different organisms suggests that many types of organisms have undergone both processes. 

The most dramatic type of duplication event is the duplication of an entire genome.  After such an event (which is rare), some genes will be lost.  This will be the result of random mutations that inactivate the gene.  Since there is another copy of the gene, the loss of the second copy may have little effect.

On the other hand, when a gene is duplicated, it is likely that the total amount of gene product produced will increase - in some cases, increased gene expression can have harmful effects (think mammalian X chromosome and females). When this is the case, inactivation of one copy of the gene may, in fact, be beneficial. 

Mutations can also lead to changes in activity and specificity - a mutation can change a protein so that it can bind new substrates and catalyze new reactions, a process known as neofunctionalization.  

One issue with "blind" duplication events is that the development of new functions occurs only after duplication.  The duplicated gene may be lost through mutation (or genetic drift).  

An alternative process involves genes that develop potentially valuable "side" activities. Before a duplication event, the evolution of these side functions is constrained by the fact that the gene already has important functions to carry out.  If the gene is duplicated, however, then the new function can be modified through the standard process of mutation and selection, while selection (conservative) acts to maintain the original function.  

As we look at the genes of various organisms, we find evidence for both gene and genome duplication.  For example, compared to their invertebrate ancestors, the vertebrate lineage appears to have gone through two rounds of whole genome duplication. 

1. How could you tell whether the same X chromosome was inactivated in each cell of a female person?
2. How would you design a multicomponent loop to produce a steady level of product?
3. How would the presence of a repressor factor influence the various modules described above?
4. How can transcription factors be regulated?
5. How does regulating the intracellular localization of a transcription factor alter gene expression?  
6. What kind of mutation would permanently inactivate a gene?

• Why, do you think, that one of the two female X chromosomes is packed into heterochromatin?  What is the point of that?  
• What type of event would lead to total genome duplication?
• Why are some genes lost after genome duplication?