Detailed mapping of the sites of interaction of polypyrimidine tract binding protein with its RNA targets: viral and cellular IRESs and pre-mRNAs.

Lead Research Organisation: University of Cambridge
Department Name: Biochemistry

Abstract

The strong physical, behavioural, and mental similarity of identical twins shows that it is our genetic material that largely specifies what we are. This genetic material is our DNA, which can be regarded as an exceedingly long 'tape' (e.g. an enormous video tape) with some 3 billion bits of information, coding for the numerous different types of protein which carry out most of our bodily functions. The DNA can be thought of as the 'master tape', an archive of thousands of normal length videos joined together. The process of decoding this information in the DNA involves first making a working copy of one of these videotapes in the archive. In fact the information in this initial copy is segmented, with the meaningful segments interspersed with meaningless sections. Consequently, the meaningful segments in this initial copy need to be cut out and spliced together, much as cine films were created by splicing in the pre-electronic era. In many cases, this splicing can occur in slightly different ways, with some meaningful segments excluded in some cells or some circumstances, but included in other cells/circumstances. Thus the initial copy of the archived 'video' can give rise to more than one variant of the final playable video. A protein known as polypyrimidine tract binding protein (PTB) exerts a strong influence on the alternative splicing, causing the inclusion of some meaningful segments and the exclusion of others. After the video has been spliced together it is decoded by a small particle known as a ribosome, which essentially carries out an analogous function to that of a video recorder (VCR) as it decodes the information as pictures. The 'biological videos' are similar to real videos in so far as the part of the tape with the pictorial information is preceded by a short leader length that has no information. In most cases our biological VCR (the ribosome) finds the point where the true information starts by searching or scanning in fast-forward mode through the whole meaningless leader. However, with some such videos the biological VCR is able to go straight to the starting point where the meaningful pictorial information begins without searching through the meaningless leader section, and in many such cases PTB is necessary for the direct location of the start position by the biological VCR. The aim of this project is to find out how PTB (a) causes the biological VCR to locate these starting sites without a need for searching through the tape from the very start, and (b) can influence the exact pattern in which the different meaningful segments are spliced together to give the finished video tape.

Technical Summary

A cDNA clone suitable for bacterial expression of His-tagged PTB will be mutated to give at least 8 variants, each with a single surface-exposed cysteine residue close to one of the four RBDs (a minimum of two variants with a single cysteine proximal to each RBD). The recombinant proteins will then be conjugated with Fe(II)-BABE (iron-p-bromoacetamidobenzyl EDTA). After verifying that these Fe(II)-BABE conjugates retain in vitro biological activity towards their RNA targets, they will be used for tethered hydroxyl radical footprinting. This will involve incubating the conjugated PTB mutants individually with 32P-labelled target RNA, initiating the Fenton reaction by addition of ascorbic acid/hydrogen peroxide, and then determining the sites at which the target RNA has been cleaved by the resulting hydroxyl radicals. By this approach, we will be able to define which RNA segments are contacted by each of the 4 RBDs, i.e. we will be able to 'dock' the PTB on to the RNA sequence. The next stage will be to test whether all these contacts are necessary for biological activity. To this end, mutations will be introduced into the RNA binding surface of each RBD. The in vitro biological activity of these mutants will then be assessed, and hydroxyl radical footprinting used to verify that the interaction of the mutated RBD with the RNA has been disrupted without perturbing the interactions of the other 3 RBDs. These approaches will be applied to the interaction of PTB with the following targets: 1) the encephalomyocarditis virus IRES (internal ribosome entry site); 2) the IRESs of poliovirus (an enterovirus) and human rhinovirus; 3) the exons (plus flanking introns) of alpha-actinin and alpha-tropomyosin pre-mRNAs that are subject to alternative splicing regulated by PTB; 4) a selection of cellular mRNA IRESs that are known to be stimulated by PTB: Apaf-1 and BAG-1 mRNA IRESs.
 
Description Polypyrimidine tract binding protein (PTB) is an RNA-binding protein which plays important roles in regulating alternative pre-mRNA splicing, and is also required for the translation of several picornavirus RNAs, which is promoted by IRESs (internal ribosome entry sites). In common with many other RNA-binding proteins, PTB has multiple RNA-binding domains (RBDs), in this case 4. Although the binding sites of PTB on its target RNAs have previously been mapped in outline, the question of which RBD binds to which site on the target RNA had not previously been addressed.
In this project we have used tethered hydroxyl radical probing to answer this question with respect to the binding of PTB to 3 different types of picornavirus IRES: encephalomyocarditis virus (EMCV), poliovirus (PV) and Aichivirus (AV). The results show that PTB binds to the EMCV and AV IRESs at widely dispersed sites, which implies that it stabilises the three-dimensional structure of the IRES, restraining the flexibility of the IRES in favour of the optimum conformation. By contrast, the binding of PTB to the PV IRES is very localised, and overlaps the binding site of the essential translation initiation factor eIF4G. PTB binding to the IRES was shown to subtly reposition the binding of eIF4G, which provides a mechanistic explanation of how and why PTB strongly stimulates translation of PV RNA.
After mapping which of the 4 RBDs of PTB binds to which site on the IRES RNAs, the second objective was to determine whether all 4 interactions are essential for the biological activity of PTB. To this end, the RNA-binding activity of each of the 4 RBDs was inactivated by introducing a maximum of 3 point mutations into each RNA-binding surface. Tethered hydroxyl radical probing confirmed the inactivation of each mutated RBD, and, importantly, showed that such inactivation had no significant effect on the orientation and binding of the other 3 (unmutated) RBDs. PV RNA translation was shown to require interaction with RBDs-1, 2 and 4 (but not RBD-3), while stimulation of EMCV RNA translation requires RBD-2 (but not RBD-1) and either RBD-3 or RBD-4. Thus the importance of each RBD for the biological activity of PTB is highly dependent on the particular target RNA in question.
(Further information can be found in the Final Report submitted in March, 2010.)
Exploitation Route As detailed under the Research Tools and Methods section, the numerous PTB derivatives prepared in this project, have been used by Prof C.W.J. Smith of this department to elucidate how PTB regulates the alternative splicing of FAS pre-mRNA, in work that was supported by BB/H004203/1 with results published in Mickleburgh I. et al. 2014. Nucleic Acids Res 42: 8605-8620.
Moreover, the methods which were developed in the course of this current grant (BB/E004857/1) could equally well be applied to study how other RNA-binding proteins with multiple RNA-binding domains (for example, the large family of KH-domain proteins) interact with their target RNAs. As yet, the methods have not been extended beyond the study of PTB, but it would be relatively straightforward to apply them to other RNA-binding proteins.
Sectors Healthcare,Pharmaceuticals and Medical Biotechnology
 
Description Although the findings of work supported by this grant have been influential on subsequent academic research (as described under Key Findings), they have not yet had any known influence outside academia.
 
Title PTB mutants 
Description A panel of almost 50 mutated forms of PTB (polypyrimidine tract binding protein) was developed. Some of these were designed for tethered hydroxyl radical probing assays to determine which of the 4 RNA-binding domains binds to which site on the target RNA. Others were designed to inactivate the RNA-binding activity of each RNA-binding domain. By combining the two types of mutation and using tethered hydroxyl radical probing it was possible (i) to verify that the RNA-binding potential of the mutated RNA-binding domain had indeed been inactivated; and (ii) to verify that the binding of the other 3 non-mutated RNA-binding domains to the target RNA had not been affected. These mutants have been subsequently been used for research into the role of PTB in alternative splicing supported by BBSRC project grant BB/H004203/1 to Prof. C.W.J. Smith (Dept of Biochemistry, University of Cambridge). Although it has been made clear that these reagents are freely available to others, as yet there have been no requests for them from outside this department. 
Type Of Material Technology assay or reagent 
Provided To Others? No  
Impact As described above, these reagents have been used in this department by Prof C.W.J.Smith, supported by BB/H004203/1, to examine how PTB regulates alternative splicing of FAS pre-mRNA (Nucleic Acids Res. 2014. 42: 8605-8620).