Transcription and RNA Processing

Transcriptional regulation


As usual, images from wikicommons.
The first question is, how does the machine that makes RNA recognize a “gene?” In the past, I’ve called a “gene” the stretch of DNA encoding a protein
and its regulatory sequences. We can separate it into the “structural gene,” which contains the actual code of the protein as it would be found in the mRNA, and the regulatory sequences. A very basic regulatory region is called the “promoter.” It is a region that serves as the initial recognition site that says: “This is a gene, transcribe here.”
In addition to the hydrogen bonds that make up the base pairs, there are additional sites for hydrogen bonds and other contacts that can be made without “unzipping” the DNA, primarily along the major groove. In particular, the major grove is just about the right size for an alpha helix of protein to fit in. There, R-chains on the outside of the helix can make specific contacts with the bases and “read” the sequence (that is, bind in a sequence-specific manner). The first step in transcription is recognition of the promotor, almost always upstream from the structural gene. There are a couple of types of DNA binding proteins. Here are two examples:
LambdaRep
Notice two things: the protein binds as a dimer. In fact, each is identical and the overall DNA sequence is a “palindrome.” Second, notice that one of the alpha-helices in each subunit is reaching down into the major groove, where it makes sequence specific contacts. This particular protein actually blocks access of the RNA polymerase in E. coli. But, similar proteins act to promote binding.
bZIPThis is a b-zip protein binding domain, of a class that acts to promote transcription in our cells. Notice the same two things. Even though the overall structure of the proteins is different (note that much of the structure is left out), the binding has some similarities.
The big picture is that the protein is reading the DNA composition and will bind to a sequence it recognizes. The DNA sequence that is recognized is called the promoter and the protein that binds is called, in general terms, a transcription factor. Things that bind DNA may stimulate or prevent transcription.
Here is a cool video on the process:


RNA Processing


processing1A
Most of what I’ll be talking about here occurs specifically in
eukaryotes.
As usual, most images are from Wikicommons.
Following transcription, the “pre-mRNA” must be processed on its way out of the nucleus to the cytoplasm.
First, notice that the sequence that made up the promotor does not end up in the pre-mRNA. This is a general theme: With each step in the DNA→RNA→Protein pathway, the information needed to specify regulation of each step generally is lost as we go to the next step. This makes sense, since it is no longer needed.

Poly-A tail and GTP cap.


When the RNA polymerase reaches the end of the sequence that’s supposed to be transcribed, it hits a signal called the “polyadenylation signal.” The sequence in the DNA (coding strand) is 5'-AATAAA. Of course, the polymerase is reading the template strand, so you could say the “signal” really is 3' TTATTT. The sequence varies a fair bit. This signals the RNA polymerase to leave the DNA. The sequence now in the pre mRNA (AAUAAA, or something similar) recruits a protein complex that will cleave the mRNA near the 3' end and an enzyme called PAP, for Poly Adenyl Polymerase, uses ATP to add a long series of A’s to the end of the mRNA. These A’s, and there may be hundreds of them, are NOT encoded anywhere.
The utility of the tail seems to be in recruiting several poly-A binding proteins that protect the mRNA from nucleases that degrade it, facilitate the next steps in processing, the transport of the mRNA out of the nucleus to the cytoplasm, and regulation of translation.
Some prokaryotes do a version of poly-A tail also.
The other initial change is the addition of the GTP cap. Actually, it’s a modified G with a methyl group on position 7 of the base, cleverly called 7-methyl GTP. Wiki has a
short discussion of the mechanism of transfer. There is a specific enzyme that does the capping and the cap looks something like this:
GTPCap

Note the “inverted” 5'-5' link. The cap structure also protects the end of the mRNA from degradation.
RNA splicing:
The pre-mRNA has regions called “introns” (for “intervening sequences” with a cool “tron” ending), and Exons. The exons encode the protein and are what end up in the final mRNA. The introns are spliced out. We will talk about the mechanism and the implications later.
The final processed mRNA might look something like this:

processed Transcript
It comprises the coding sequence, the cap, the 5' and 3' “untranslated region” (UTR) and the poly-A tail. I think all the names are pretty self-explanatory. The 5' UTR, in particular, will contain sequences that contribute to regulating translation. Only the green stretch above will make it into the protein.

RNA Splicing part 1


Take a look at
this video which is presented by Cold Spring Harbor Lab, where I used to work years ago and where some of the work I discussed today was done.
Key words:
Spliceosome: (there’s a good descriptive name) It’s the assembly of proteins and RNA that carry out splicing.
snRNPs (pronounced “snurps”). These are “small nuclear ribonuclear proteins.” They are the components of the spliceosome.
snRNAs: small nuclear RNAs are the main catalytic components. It is the RNA that carries out the reaction. They have names like “U1,” and “U2.” etc.


Self Splicing:
The story about how RNAs got spliced became more obvious when self-splicing RNAs were found. I mentioned these today.
The general reaction looks like this:
Splicing
There are two, successive reactions that occur in a concerted way sometimes using energy from an additional GTP that is brought in by the complex. The first takes a branch point sequence near the 3' end of the intron. The 2' OH of that branch point attacks the 5' exon/intron boundary. This splices the intron into a lariat and frees the 3' end of the exon, which then attack the 5' end of the next exon, resulting in the spliced exons and a lariat structure of the intron.
In the self-splicing form, “internal guide sequences” form base-paired structures that hold the players together and facilitate the attacks..
You can see below how the stem-loop structures formed using base pairs, then can fold into three-dimensional structures, facilitating catalysis.

Self3

You can see how the base pairing can be used to bring the active sites together in more detail here. By the way, these last two images are used without permission. The one below is from the Molecule of the Month blog/discussion and the one above is from Trends in Ecology and Evolution.
Closeup

Normal mRNA splicing (non self-splicing)


The story is not all that different in the splicing of mRNA. There are guide sequences and a branch-point lariat. The chemistry varies a little as to who attacks whom when and where. But, the main difference is that the sequence of the intron is not that important for the structure to work. Instead of the intron folding into a complex structure, the separate RNAs in the snRNPs form those structures and then bind the specific sequences in the intron and at the intron-exon borders.

Two benefits of splicing


While no one believes that RNA splicing evolved
because of these benefits, these are real benefits organisms now enjoy because of it.
  1. Alternative splicing: Not all the exons are included when splicing of specific mRNAs takes place. Different versions of the mRNA may be formed. In the example I gave earlier, exons 15-19 could all be joined up in sequence, or, exon 15 could be spliced directly to 19, leaving out 16, 17 and 18. This results in different versions of the protein, with different functions. The domain structure of proteins makes this possible.
  2. Exon shuffling: Also because of the domain structure of proteins, it is possible to add exons to genes via recombination and create different versions of the protein. Any chunk of DNA that has portions of introns on its ends that then is spliced into another intron will do no damage to an existing gene (It can be spliced out at the mRNA level). But, it also provides the possibility, through alternative splicing, to evolve proteins with new functions.