Rna

ribozyme

BOX 13-1 FIGURE 1 Group I introns can be converted into true ribozymes.

a pre-rnFiNA spliceosome U5

b group It self-splicing c group I self-spltcing a pre-rnFiNA spliceosome U5

b group It self-splicing

5'

FIGURE 13-9 Group I and group II introns. TT-ts figure compares the reaction of the Setf-Splidng group I and II mtrons and the spliceosome mediated reaction already described The chemistry in (he case of group II introns is essentially the same as in the spliceosome case, with a highly reactive Adenine within the intron initiating splicing, and leading to the formation cf a lanat product In the case of the group I rntron, the RNA folds in a way that forms a Cuaninobinding pocket, which allows the molecule to bnd a free Guanine nucleotide and use that to imiiate splicing Although these mtrons can splice ihemseVes out of RNA molecules unaided by proteins in vitro, in wo tliey typically do require protein components io stimulate tfie reaction (Source: Adapted from Cecil tR 19B6 Tlie generality of self splicirg RNA: Relationship to nuclear ffiRNA splic irig. Osti 44:207 -210, rig ?.J

group IMike self-splicing introns were the starting point for the evolution of modern pre-mRNA splicing. The catalytic functions provided by the RNA were retained, but the requirement for extensive sequence specificity within the intron itself was relieved by having the snRNAs and their associated proteins provide must of those functions in trans. In this way. introns had only to retain the minimum of sequence elements required to target splicing to the correct places. Thus, many more and varied sixes and sequences of introns were permitted.

It is interesting that the structure of the catalytic region that performs the first transesterification reaction is very similar in the group II intron and the pre-mRNA/snRNP complex (Figure 13-10), This observation fuels the broader speculation (discussed in Chapter 6) that early in tlie evolution of modern organisms, many catalytic functions in the cell were carried out by RNAs and that these functions have, on the whole, since been replaced by proteins. In the case of the spliceosome and the ribosorne, however, these activities have not been entirely replaced by proteins. Rather, the vestigial RNA-catalyzed mechanisms remain at the heart of tlie present complex machinery

pre-mRNA

SAAUC Aij

FIGURE 13-10 Proposed folding at the RNA catalytic regions for splicing of group II irrtrom and pre-mRNAs. Ihe doted regions of the RNA in the group li case replace an additional four folded domains not shovm in this depiction

Definition Base Pair Gwas Study

How Does the Spiiceosome Find the Splice Sites Reliably?

We have already seen one mechanism that guards against inappropriate splicing—the active site of the spiiceosome is only formed on RNA sequences that pass the test of being recognized by multiple elements during spiiceosome assembly. Thus, for example, the 5' splice site must be recognized initially by the Ul snRNP and then by the U6 snRNP. It is unlikely both would recognize an incorrect sequence, and so selection is stringent. Yet, the problem of appropriate splice-site recognition in the pre-mRNA remains formidable.

Consider the following. The average human gene has eight or nine exons and can be spliced in three alternative forms. But there is one human gene with 363 exons and one DrosophiJa gene that can be splicud in 38,000 alternative ways (Figure 13-11). If the snRNPs had to find the correct 5' and 3' splice sites on a complete RNA molecule and bring them together in the corrcct pairs, unaided, it seems inevitable that many errors would occur. Remember, also, that the average exon is only some 150 nucleotides long, whereas the average intrnn is approximately 3,000 nucleotides long (as we have seen, some inttpns can be as long as 800,000 nucleotides). Thus, the exons must be identified within a vast ocean of intronic sequences.

exori 4 exon 6 exon 9 exon 17

12 alternatives 48 alternatives 33 alternatives 2 alternatives genomic r~—i r—-------1 i—— i p exori 4 exon 6 exon 9 exon 17

12 alternatives 48 alternatives 33 alternatives 2 alternatives genomic r~—i r—-------1 i—— i p

FIGURE 13-11 The multiple exons of the Drosophila DSCAM gene. This gene was cloned as en axon guidance receptor responsible for directing growth cones to their proper target. The DSCAM gene (shown at the top) is 612 kb long; once transcribed and spliced, tt produces one or more versions of a 7.8 kb, 24 exon, mRNA (the figure shows the generic structure of those mRNAs). As shown, there are several mutually exclusive alternatives forexons 4. 6, 9, and 17 t hus, each mRNA wilt contain one of 12 possible aftematives for exon 4 (in orange), one of 46 for excn 6 (purple), one of 33 for exon 9 (blue), arid one cf ? for exon 17 (red), if all possible combinations of these exons are used, the DSCAM gene produces ¿8,0)6 different mRNAs and proteins (Source: Adapted from Stack D. 2000. Protein diversity from alternative splicing Cell 103: 36B Copyright © 7000 Used with permission from Elsevier )

FIGURE 13-11 The multiple exons of the Drosophila DSCAM gene. This gene was cloned as en axon guidance receptor responsible for directing growth cones to their proper target. The DSCAM gene (shown at the top) is 612 kb long; once transcribed and spliced, tt produces one or more versions of a 7.8 kb, 24 exon, mRNA (the figure shows the generic structure of those mRNAs). As shown, there are several mutually exclusive alternatives forexons 4. 6, 9, and 17 t hus, each mRNA wilt contain one of 12 possible aftematives for exon 4 (in orange), one of 46 for excn 6 (purple), one of 33 for exon 9 (blue), arid one cf ? for exon 17 (red), if all possible combinations of these exons are used, the DSCAM gene produces ¿8,0)6 different mRNAs and proteins (Source: Adapted from Stack D. 2000. Protein diversity from alternative splicing Cell 103: 36B Copyright © 7000 Used with permission from Elsevier )

Splice-site recognition is prone to two kinds of errors [Figure 13-12). First, splice siles can be skipped, with components bound at, for example, a given 5' splice site pairing with those at a 3' site beyond the correct one.

Second, other sites, close in sequence but not legitimate splice sites, couid be mistakenly recognized. This is easy to appreciate when one recalls that the splice sile consensus sequences are rather loose. And so, for example, components al a given 5' splice site might pair with components bound incorrectly at such a "pseudo" 3' splice site (see Figure 13-12b).

Two ways in which the accuracy of splice-site selection can be enhanced are as follows- First, as we saw in Chapter 12, while transcribing a gene to produce the RNA, RNA polymerase II carries with it various proteins with roles in RNA processing (see Figure 12-18).

a exon skipping exon 1

dna b pseudo splice-site selection exon 1 2

pre-mrna

"incorrect" mRNA

"pseudo" splice sile

FIGURE 13-12 Errors produced by mistakes in splice-site selection, (a) Shows the consequence of skipping an exon This happens if the spliceosome components bound at the 5' splice site of one exon interact with spliceosome components bound at the i' splice site ot not the next exon, but one beyond, (b) Illustrates the effect of spliceosome components recognizing "pseudo" splice sites—sequences that resemble (but fire not) legitimate splice sites, tn the case shown the pseudo site is within an exon and leads to regions near the 5' end of that exon being mistakenly spliced out deng with the intrc.n

These include proteins involved in splicing. When a F>' splice site is encountered in the newly synthesized RNA, those components are transferred from the polymerase C-terminal "tail" [that part of the enzyme where they hitch a ride) onto the RNA. Once in place, the 5' splice site components are poised to interact with those that bind to the next 3' splice site to be synthesized. Thus, the correct splice site ran be recognized before any competing sites further downstream have been transcribed. This co-transcriptional loading process greatly diminishes the likelihood of exon skipping.

It is worth noting that even though much of the splicing machinery assembles while the gene is being transcribed—and on individual inlrons in the order they are transcribed—this does not mean the inlrons are themselves spliced out in that order. Thus, in contrast to many other activities we have heard about—transcription, replication, and so on—there appears to be no "tracking" mechanism involved, whereby the machinery assembles at one end of the gene or message and acts as it tracks to the other end.

A second mechanism guards against the use of incorrect situs by ensuring that splice sites close to exons (and thus likely to be authentic) are recognized preferentially. So-called SR (Serine Argenine rich) proteins bind to sequences called exonic splicing enhancers (ESEs) within the exons, SR proteins bound to these sites interact with components of the splicing machinery, recruiting them to the nearby splice sites. In this way, the machinery binds more efficiently to those splice sites than to incorrect sites not close to exons. Specifically, the SR proteins recruit the U2AF proteins to the 3' splice site and Ul snKNF to the 5' site (Figure 13-13). As we saw earlier, these factors demarcate the splice sites for the rest of the machinery to assemble correct! y.

SR proteins are essential for splicing. They not only ensure the accuracy and efficiency of constitutive splicing (as we have just seen) but also regulate alternative splicing (as we will see presently). They come in many varieties, some controlled by physiological signals, others constitutively active. Some are expressed preferentially in certain cell types and control splicing in cell-type specific patterns. We will discuss some specific examples of the roles of SR proteins in the next section.

ESE ESE

inlron exon intron exon incrcn

FIGURE 13-13 SR proteins recruit spliteosome components to the 5' and 3' splice sites.

Legitimate splice sites are reccgni/ed by the splicing machinery by virtue of being dose to exons. Thus, SR proteins bine! to sequences within the exons {exonic splicing enhancers), end from there recruit U2AF and U1 snRNP to the downstream 5' and upstream 3' splice sites respectively. This initiates the assembly of the splicing machinery on the correct sites and splicing can proceed as outlined earlier In looking at this figure, note that an intron is drawn in the center, bounded on each side by an exon. This is in contrast to many of the earlier mechanistic figures in which a single central inlron is depicted lying between two introns. (Source: From Maniatis 1 and Tasic E. 2002. Alternative pre-MRNA splicing and proteome expansion in metaeoans Nature 418: 236-243. Copyright C 2002 Nature Publishing Group. Used with permission )

Was this article helpful?

0 0

Post a comment