Involvement of the U1 snRNP Particle in Splicing

Involvement of the U1 snRNP Particle in Splicing

Initially, progress in the study of the biochemical mechanisms of splicing was slow. One of the first clues about the mechanism of splicing came from the sequencing of many splice sites. It was noticed that the sequence flanking the 5’ splice site was closely complementary to the sequence of RNA found at the 5’ end of an RNA found in the U1 class of small nuclear ribonucleoprotein particles called snRNPs. In addition to the U1 particles, U2, U4, U6, and others exist, each containing 90 to 150 nucleotides and about 10 different proteins. The complementarity be-tween U1 RNA and the splice site of pre-mRNA suggested that base pairing occurred between the two during splicing. Stronger evidence for the proposal was the discovery by Steitz and Flint that splicing in nuclei could be blocked by antibodies against U1 particles.

The experiment showing inactivation of splicing by anti-U1 antibod-ies is not definitive since their specificity may not be high and because other antibodies could also be present. One ingenious method for specifically inactivating U1 particles was to remove nucleotides from their 5’ ends with RNAse H. This enzyme digests RNA from RNA-DNA

Figure 5.16 How compensating mutations in two interacting structures canrestore activity.

duplexes. DNA oligonucleotides complementary to the 5’ end of U1 RNA were hybridized under gentle conditions to U1 particles in cell extracts and then RNAseH was added. These treatments left the extracts unable to catalyze intron removal whereas extracts which had received oligonu-cleotides of different sequence were not inhibited.

One of the nicest demonstrations of biologically significant interac-tions between two macromolecules are those in which compensating mutations can be utilized. First, a mutation is isolated that interferes with a particular interaction between A and B (Fig. 5.16). That is, A’ no longer interacts with B in vivo. Then a compensating mutation is isolated in B that potentiates the in vivo interaction between A’ and B’. The isolation of the mutation and the compensating mutations in components of the splicing apparatus could not be done using tradi-tional genetics tools since the splicing reactions occur in animals for which only rudimentary genetics exists. Genetic engineering methods had to be utilized.

Figure 5.17 The consensus of the sequence at the 5’end of an exon-intronjunction. The sequence is highly complementary to the 5’ end of U1 RNA. Alterations in the splice sequence reduce splicing, and compensating alterations as shown in the U1 sequence restore splicing.

Two steps are necessary to perform the experiment. The first is inducing the cells to synthesize messenger with an altered splice site as well as synthesize U1 RNA with an altered sequence, and the second is assaying for splicing of the special messenger in the presence of the normal cellular levels of messenger and pre-messenger RNAs. Weiner introduced into the cells a segment of adenovirus sequence coding for the E1a protein with wild-type or variant 5’ splice sites as well as the gene for an altered U1. Only when DNA was introduced that encoded a variant U1 gene that compensated for the splice site mutation and restored Watson-Crick base pairing across the region was the variant splice site utilized (Fig. 5.17).

The use of an adenovirus sequence for the experiment permitted altering the splicing of a messenger which was not important to the cells. Similarly, introducing a new gene for U1 RNA avoided disruption of the ongoing cellular splicing processes. Finally, examining splicing in the E1a gene from the virus increased the sensitivity since this gene contains three different 5’ splice sites, each utilizing the same 3’ splice site. Damaging the activity of one 5’ splice site diverted splicing to the other splice sites, whereas had there been only one 5’ splice site, altering it might only have slowed kinetics of the splicing process without chang-ing the actual amounts of RNA eventually spliced.

Figure 5.18 The pattern of splicing in adenovirus. The 5’splice sites separatingregions A, B, and C are all joined to the 3’ splice site separating regions D and E. Thus the three RNAs shown can be produced. The amount of the species ABE can be determined by hybridizing to radioactive RNA complementary to ABC and digesting with T1 and pancreatic RNAses. The amount of product of length AB is determined by electrophoresis.

An RNAse protection assay was used to monitor the splicing. RNA was isolated from the cells and hybridized with radioactive RNA com-plementary to the adenovirus mRNA. The regions of the RNA which are not base paired are sensitive to T1 and pancreatic RNAses and are digested away (Fig. 5.18). This leaves a radioactive RNA molecule whose size indicates the splice site utilized. A variant sequence at the middle E1a splice site eliminated splicing from this location, but the introduc-tion of the variant U1 gene to the cells containing the compensating mutation restored use of this splice site.