Where does mRNA formation occur?

Answer

Pre-mRNA Splicing Many, if not most, protein coding genes in eukaryotes are interrupted by regions which do not appear in the mRNA - we call these introns. The process of intron removal is called splicing.

The Mechanism of Splicing The splicing mechanism must be able to recognise exon/intron boundaries so that cuts and joins can be made in the correct place.

The only feature which is 100% conserved at intron/exon junctions is that introns begin with GU and end in AG. There are, however, nucleotides that are found more frequently at particular positions (percentages are shown in the diagram above). Vertebrates also usually have a pyrimidine rich sequence (12Py) close to the 3' end of the intron. Deletion analysis has shown that although intron size varies widely, only 30-40 nucleotides at each end of an intron are required for its efficient removal.

Pre-mRNA splicing can be studied in vitro using nuclear extracts:

Using such systems, the factors which influence the splicing reaction can be investigated.

Analysis of splicing intermediates showed that pre-mRNA splicing proceeds via a lariat intermediate, just as it does in group II self-splicing.

First, cleavage occurs at the 5' junction - sometimes called the splice donor site. The phosphate at the 5'end of the intron then becomes linked to the 2' OH of an adenine approximately 25 nucleotides upstream of the 3' end of the intron, which is sometimes called the acceptor site. This A residue is called the branch point.

The next step is that cleavage occurs at the 3' splice junction and the 5' phosphate of the downstream exon is joined to the 3' OH of the upstream exon.

SNURPs Pre-mRNA splicing requires complexes of RNA and protein called snRNPs or SNURPs.

The reason that they are called small nuclear ribonucleoproteins is that the RNA component of SNURPs is one of a small number of RNA molecules called small nuclear RNAs. There are at least 7 of these numbered U1-U7, those involved in pre-mRNA splicing are U1, U2, U4, U5 and U6. Each of these RNAs is complexed with 6-10 proteins to make a SNURP. SNURPs are named after the snRNA which they contain. Some of the SNURP proteins are common to all SNURPs while some are specific to particular SNURPs. During splicing, SNURPs are found in a large complex with pre-mRNA called a spliceosome.

The U1 SNURP binds to the 5' junction because its snRNA (U1 snRNA) contains a sequence which is complementary to the consensus sequence at the 5' junction.

The addition of a oligonucleotide corresponding to the 5' consensus sequence blocks splicing, as does mutation of the consensus sequence - presumably by preventing binding of U1 to the 5' junction sequence.

The U2 SNURP is also thought to bind to the intron by base-pairing of its RNA - this time to a sequence at the branch point.

Spliceosome Assembly A protein which is not part of any SNURP, called U2AF (for U2 associated factor) binds close to the 3' splice site and seems to help locate the U2 SNURP at the branch point. When the U1 and U2 SNURPs have bound to the pre-mRNA a pre-spliceosome is formed. Interaction between U1 and U2 causes RNA to fold so that the 5' junction and the branch site are in close proximity.

At this point the other three SNURPS U4, U5 and U6, join the complex as a unit. Prior to joining the spliceosome U4 and U6 are linked closely together in the U5-U4-U6 complex. This is due to base pairing between the U4 and U6 snRNAs.

After joining the spliceosome U4 and U6 dissociate from each other but remain joined to the spliceosome. U6 now interacts with U2 - it does this by base pairing in two regions called helix I and helix II. At this point U1 and U4 probably leave the spliceosome, allowing U6 to interact with sequences close to the 5' splice site and U5 interacts with nucleotides at the end of the first exon, just before the 5' splice junction. This is now the active spliceosome.

5' cleavage and lariat formation occur in the first splicing reaction. U5 remains attached to the end of the first exon and now interacts also with the beginning of the second exon, thus bringing the two exons together. The exons are joined together in the second reaction. The ligated exons are released from the spliceosome but the lariat intron remains bound. The spliceosome then disassembles, releasing the lariat which is linearised and degraded. The SNURPs are recycled.

The splicing mechanism is likely to be universal but there are some differences between mammals and yeast. Yeast consensus sequences at the 5' junction and the branch site are much more conserved than in higher eukaryotes. If these sequences are mutated in yeast splicing is completely inhibited but in mammals, alternative sites called cryptic sites are used. Mammalian splicing is therefore more flexible and the requirement for particular sequences is less strict. Yeast also does not have the pyrimidine tract found upstream of the 3' junction in mammalian introns.

PRP Proteins The advantage of using yeast to study splicing is that mutations which prevent splicing can be isolated - this allows identification of genes (and therefore proteins) which are involved in splicing - these are called PRP genes and proteins. PRP stands for precursor RNA processing. At least 30 different PRP genes have been identified. Some PRP proteins are integral components of SNURPs, for example, PRP3, PRP4 and PRP6 are part of the U4-U6 SNURP and PRP8 is found in the U5 SNURP. Others are extrinsic factors which may associate only briefly with the spliceosome or pre-spliceosome eg PRP2. The points in the splicing pathway at which some of these act have been identified.

PRP5 is required for the binding of U2 to the branch point. PRP4 is required for the addition of the U5-U4-U6 SNURP to the splicesome. PRP2 is required for the first splicing reaction. PRPs 16, 17, 18 and 29 are required for the second reaction. PRP22 is required for the release of the ligated exons from the spliceosome.

Most PRP genes have been cloned and sequenced and a number have similarity to known RNA helicases - these are enzymes which unwind double stranded RNA. This may reflect the importance of RNA-RNA base pairing at various points in the pathway.

Most PRP proteins are thought to be necessary for the assembly or disassembly of the spliceosome - none have as yet, been identified which carry out the transesterification reactions.

Because of the similarity between group II self splicing and pre-mRNA splicing, one possibility is that there are no proteins which catalyse the transesterifications - instead the RNA components of one or more SNURPs would have catalytic activity. The function of the multitude of proteins involved in pre-mRNA splicing would be to fold the RNA into the correct conformation for splicing to take place. The U6 snRNA has been proposed as the most likely candidate for the catalytic RNA (if it exists). This is because its sequence is highly conserved between species. Mutation of two highly conserved regions of the U6 snRNA inhibits splicing. This suggests that U6 snRNA may be catalytic but before we can be sure about this purified U6 snRNA without any protein will have to be shown to possess some catalytic activity.

Alternative Splicing Alternative splicing of exons may be used as a means of regulating gene expression. There are several different ways in which a pre-mRNA could be alternaticely spliced:

1. Alternative 5' splice sites and a common 3' splice site

2. Splice or no splice

3. Exon skipping

4. Common 5' splice site, alternative 3' splice sites.

When splicing a pre-mRNA with multiple introns, how does the splicing machinery join adjacent 5' and 3' sites and avoid missing out exons?

One possible answer to this might be the proximity effect. In many pre-mRNAs 5' splice sites closer to the 3' splice site are chosen in preference to sites further away. This would ensure that exons are not missed out. It is not clear how this proximity effect is achieved - it could be by a scanning mechanism where the splicing machinery scans the pre-mRNA to find the closest splice sites. Alternatively, it has been suggested that the machinery is somehow able to recognise exons - it would then choose sites at the end of one exon and the beginning of the next.

The exon skipping type of alternative splicing may therefore be regulated by influencing the splicing machinery's choice of proximal (close) versus distal (distant) splice sites. A protein which is capable of influencing this decision has been identified, called SF2 or ASF. Increased levels of SF2 promote proximal splice site choice and prevent exon skipping. It is not known whether SF2 plays any role in alternative splicing in vivo.

SF2 is a member of a group of proteins called SR proteins. These proteins have an N-terminal RNA recognition sequence and a C-terminal region which is rich in serine and arginine - this is where they get their name from (S=serine, R=arginine). They all appear to be involved in splicing but their roles are not clear at the moment.

One of the best characterised examples of alternative splicing occurs during sex determination in Drosophila. Sex in Drosophila is determined by the activity of a gene called Sex lethal (Sxl). When active, female development occurs and when inactive, male development occurs.

The Sxl gene codes for an RNA binding protein and it exerts its effect by regulating the alternative splicing of the transcript of the tra gene. In females a downstream 3' splice site is preferentially used in favour of the upstream 3' splice site which may be used in either sex. So, in females, Sxl seems to promote the use of the downstream site by binding preferentially to the upstream site and preventing binding of the U2AF splicing factor. In the absence of Sxl U2AF preferentially binds to the non-specific upstream site and promotes its use.

Another example of alternative splicing is the Drosophila P element. In this case the third intron is spliced out only in germ line cells. The U1 SNURP seems to be involved in the repression of splicing in somatic cells. Just upstream of the normal 5' splice site of the third intron there are two pseudo splice sites called F1 and F2. In somatic cells U1 preferentially binds to the F1 site and this prevents U1 binding to the authentic site. This actually prevents splicing, presumably because the splicing machinery sees the splice site as being in exon and therefore ignores it. In germ line cells a factor prevents U1 from binding to F1, thus allowing it to bind to the correct 5' splice site and splicing takes place.