7/24/01

Lecture #11 - RNA processing

In eukaryotic organisms, the primary mRNA transcript is often altered through a process called mRNA (or pre-mRNA) splicing.
The splicing process involves removing introns from the pre-mRNA and ligating the exons together.
Other post-transcriptional modifications include the addition of a 5' guanosine CAP and a 3' poly-adenosine tail to the pre-mRNA.
These modifications concern transcripts transcribed by RNA Pol II.
All post-transcriptional modifications occur in the nucleus, prior to transport to the cytoplasm.

5' guanosine CAP addition

The 5' end of RNA Pol II mRNA's are "capped" with a guanosine after transcription.
After transcription the pre-mRNA looks like this (p=phosphate, N=nucleotide)
5' -- pppNpNpN.....--3'
The CAP is added via an unusual 5'&emdash; 5' bond which is formed by an enzyme guanyl transferase that adds a pppG (GTP) onto the 5' end of the RNA. The pre-mRNA now looks like this:
Guanosine CAPs are also methylated at the N7 position, so they are often called 7-methyl G CAPs.
Other methylations occur on the sugar rings of the first two bases and on the N6 position of the first base, but these are more variable and less important.

Poly-A tail

If the pre-mRNA contains the specific sequence 5'--AAUAAA --3', it is targeted for cleavage in preparation for poly-A addition.
The poly-A tail can contain as many as 200 A's.
Note &endash; transcription termination in RNA Pol II is not well understood. Pol II continues to transcribe after generation for the AAUAAA, and what causes it to eventually fall off the template is unknown.

A large complex carries out cleavage and poly-A addition. The protein components required are:

1) endonucleases &endash; these carry out the actual mRNA cleavage.; there are two of them, CF1 and CF2. They cleave the mRNA to prepare a 3' end for poly-A addition.

2) stimulatory factor (CTSF) &endash; this binds to a G/U rich sequence downstream (3') to the cleavage site. It is involved in assembly of the complex and activates other components.

3) specificity factor (CPSF) &endash; this recognizes the 5'--AAUAAA --3' sequence in the mRNA. It requires CTSF to function.

4) poly-A polymerase (PAP) &endash; This synthesizes the poly-A tail in a two-step process. The first ten adenosines added are dependent on the 5'--AAUAAA --3' sequence and the specificity factor. In the second stage, after adenosine number ten, these requirements are no longer present.

5) poly-A binding protein (PABP) &endash; This is a non-sequence specific RNA binding protein. It binds stoichiometrically to runs of A's and coats the poly-A tail.

pre-mRNA splicing

The machinery that carries out the splicing is a large complex called the splicesome which contains both RNA and protein. It is a very large complex; it runs at about 50-60S, so it is comparable in size to a ribosome.
Splicesomes contain protein and snRNA components, both of which are important for its activity.
The individual subunits of splicesomes are called snRNP's, or small nuclear ribonucleoproteins. Each snRNP contains proteins and snRNA's (small nuclear RNA's) of about 100-300 bases.
There are 4 snRNP's involved in pre-mRNA splicing: U1, U2, U5 and U4/U6.
There are 5 snRNA's contained within these snRNP's &endash; U1, U2, U4, U5 and U6.
Mutations in the snRNA's block splicing, so RNA activity is important.

Splicing sequences

There are three kinds of sequences that are important for splicing:

1) 5' and 3' splice sites. There are conserved sequences at the 5' (GU) and 3' (AG) ends of the intron. The 5' GU is also called the splice donor site while the 3' AG is also called the splice acceptor site. The strength of the 5' splice site is also be determined by the last nucleotides in the exon.

2) Branchpoint sequences. These are located approximately between 20 and 40 nucleotides from the 3' end of the intron. The branchpoint sequences are not as conserved as the 5' and 3' splice site sequences. Only the second to last nucleotide in the seven nucleotide consensus sequences is conserved - it is always an A. This A is important in the creation of unusual 5' - 2' bond which results in the formation of a lariat structure.

3) A pryrimidine rich sequence between the branch site and the 3' AG site. (see steps in splicing section)

Mammalian gene structure and alternative splicing

Many eukaryotes have more than one exon and intron per gene. Multiple introns can be removed, and they can be removed in different ways to form different protein products; this is called alternative splicing.
Alternatively spliced products are formed in a regulated way &endash; there is no random "exon skipping."
Mammals especially exhibit a lot of alternative splicing. The human genome has only 35,000 genes. This initially was a surprise because many plants (such as Arabidopsis) have 25,000 genes. Alternative splicing allows the creation of many more than 35,000 proteins from 35,000 genes.

Gene structures of various organisms

Mammals &endash; tend to have very large genes which frequently contain many exons and introns. The introns may be very large and are often larger than the exons. A typical gene is 16kb with 7-8 kb of exons.
Yeast &endash; has small genes with only 5% of the genes containing introns. Most genes that do contain introns contain only one. The splicing machinery is conserved from yeast to mammals.
Plants &endash; are somewhere in between yeast and mammals. They have intermediate size genes which often have multiple introns, but the introns tend to be smaller than the exons.

What order are introns removed?

In genes with multiple introns, the question arose as to whether they are spliced out in an ordered manner. splicing could theoretically proceed 5' --> 3', 3'--> 5', or randomly.

Experimental data indicates that none of these possibilities are actually true. introns are removed in an ordered manner that is dependent on the secondary and tertiary structure of the RNA. Prediction of the order is very difficult. An experiment involving a Northern blot of steady-state levels of RNA in an ovomucoid gene demonstrates these findings (see Fig. 22.5).

Transesterification

After the 5' end of the intron is cut, the 2' hydroxyl of the conserved A in the branchpoint sequence makes a 2' --> 5' bond with the 5' end of the intron. This breaking of this 5' phosphodiester bond is somewhat unusual because it does not require ATP. It makes use of transesterifcation reaction instead. In a transesterification reaction, the breaking of one bond is coupled with the creation of another &endash; the energy is recycled in a way. The 2' hydroxyl of the A attacks the phosphodiester bond between exon and intron.. Although the transesterification reactions do not require the hydrolysis of ATP, but the process of splicing as a whole does (i.e. other steps of splicing are ATP requiring).

U1 snRNA

The U1 snRNA is very important in splicing. Its 5' end is especially critical because the 5' end of U1 RNA has a 3-'CAUUCAU-5' sequence which recognizes the complementary 5' splice sequence in an intron. If you mutate the intron sequence, you destroy splicing. If you make a compensatory mutation in the 5' end of the U1 snRNA, splicing is restored. This indicates that pre-mRNA/ U1 snRNA base pairing is necessary for splicing.

Steps in splicing

1) U1 snRNP binds to the 5' splice set.

U2AF binds to the pyrimidine rich tract.

ASF / SF2 bind. These are general splicing factors and SR proteins. (SR = serine and arginine rich.)

2) U2 binds the branch site. Note that U1 and U2AF are required for U2 binding.

U2 base-pairs with a branchpoint consensus in yeast.

The binding of U2 requires ATP hydrolysis.

3) U5-U4/U6 trimer binds

U5 binds the exon at the 5' splice site while U6 binds U2.

All of this requires bending of the RNA which is important to get the RNA in the correct orientation to carry out the transesterification reactions.

4) U1 is released and leaves the complex

U5 moves from the exon portion of the 5' splice site and replaces U1 at the intron portion of the 5' splice site.

U6 also binds the 5' splice site.

5) U4 is released (an energy dependent process)

U6 base pairs with U2. This forms the catalytic center.

The first transesterification reaction takes place &endash; the 5' splice site is cleaved and the lariat forms.

6) U5 contacts both exons and the second transesterification reaction occurs.

The 3' splice site is cleaved and the exons are ligated as the intron is released in lariat form. This step requires ATP. The lariat is later degraded.

U6 is important for activation of the catalytic center. It arrives with the U6 snRNA bound to the U4 snRNA. When bound in this form it is inactive.

When ATP is hydrolyzed to release U4, U6 then base pairs with U2 and becomes active.

Recall that U2 was initially bound to the branchpoint. At this stage it is bound to the branchpoint and U6. So U6 can base pair with either U4 (inactive) or U2 (active).

Overall, 3 ATP's are hydrolyzed for one intron removal.

How does the splicing machinery recognize the introns?

1) intron definition - In this method, the machinery scans through the pre-mRNA to find 3' and 5' splice sites of a single intron. This scanning process bridges the intron, hence the name intron definition.

2) exon definition - This involves scanning and recognizing the 5' end of one intron and the branchpoint of the next intron upstream. This scan therefore spans an exon and is called exon definition. This is used more often in mammals because the exons are so much smaller than the introns that it improves scanning efficiency. Exon definition probably does not occur in yeast since more than one intron is required for the process and most yeast genes that have introns have only one.

Recall that plants are the intermediate case &endash; some experiments addressing this issue in plants make use of the AP3 gene from Arabidopsis (in the end no time to talk about this!).

RNA editing

RNA editing is a somewhat bizarre phenomenon where the RNA sequence is altered post-transcriptionally and pre-translationally.
Thus it no longer reflects the code in the DNA. This is troubling because it violates the central dogma of molecular biology.
It can occur in mammals in a relatively simple way: The apolipoprotein has 29 exons. in its DNA, codon 2153 codes for glutamine. this normal coding is expressed in liver cells to give a full-length protein. In intestine cells, codon 2153 is altered at the RNA level to UAA which is a stop codon. This results in production of a truncated protein. Alternate explanations such as alternative splicing and duplicate genes were ruled out.
In mammals, the editing appears to be occurring by regulated deamination of adenosine to guanosine or cytosine to uracil.
The secondary structure of the mRNA activates the deaminase &endash; an enzyme with two subunits. One subunit contains the deaminase activity and one is a specificity component made of an RNA binding protein. This restricts deaminase activity to the correct positions.
In addition to the apolipoprotein genes, glutamate and serotonin receptors are also edited.
Editing in the mitochondrial genes of trypanosomes (a protozoan) is a bit more extreme.
In the cytochrome oxidase subunit 2 gene, 4 U's are inserted into the RNA which actually change the reading frame.
In the coxIII gene, hundreds of U's are added and some pieces of RNA are deleted. less than half of the RNA is actually encoded in the DNA. This must be a regulated process because it always occurs in the same way.
The U addition / deletion operates by means of guide RNA's which are encoded in the genome. These guide RNA's contain extra A's and base pair with a pre-edited RNA. the extra A's themselves do not base pair, but direct insertion of U's in the RNA to be edited. For this to happen, the phosphodiester bonds of RNA and actually broken and remade during this process. To do this requires
1) an endonuclease cuts the RNA

2) terminal uridytyl transferase (TUTase) adds the U's.

3) RNA ligase reseals the DNA.