Investigating the protein and RNA interactions of the Polypyrimidine Tract Binding protein: High resolution structures with implications in the regulation of alternative splicing
Author(s)
Joshi, Amar
Type
Thesis
Abstract
Alternative splicing produces multiple mRNAs from a single gene. It is a powerful
mechanism used to amplify complexity of the metazoan proteome. The Polypyrimidine
Tract Binding protein (PTB) is a key regulator of alternative splicing. In addition, it has
roles in the localization and stability of mRNA and is involved in IRES mediated
translation initiation. It has four RNA recognition motif domains (RRMs) which consist
of two α‐helices which pack against a β‐sheet. The RRMs act in concert to direct splicing
as each RRM binds pyrimidine rich RNA (UCUU or CUCUCU) across the β‐sheet.
Solution structures have been solved of each RRM in complex with short hexameric
oligonucleotides. However, high‐resolution structural studies of larger protein
fragments with longer RNAs have been very difficult. Small angle X‐ray scattering has
been used to probe this, but sample aggregation obstructed collection of high quality
data. To address this we have investigated how longer RNAs bind PTB fragments. We
have determined optimal RNAs which should help to determine structures of PTB
fragments in complex with longer RNA molecules.
PTB is necessary for skipping Tpm1 exon 3 in smooth muscle cells. However, interaction
with an additional protein, Raver1, is required to accomplish this. Raver1 contains four
PTB‐Raver1 interaction motifs (PRIs) with a consensus of [S/G][I/L]LGxxP. The PRIs
bind to the α‐helical side of PTB RRM2. We have created chimeric fusion constructs
where Raver1 PRIs are encoded as N‐terminal extensions to PTB RRM2. High‐resolution
crystal structures have been solved of 2 PRIs bound to RRM2, one high affinity and one
low affinity. The differences between the conformations of each PRI bound to the RRM
explain differences in affinity. We have used structure‐based mutagenesis to probe these
differences and have revealed what is necessary for a high affinity interaction. From this
we have proposed a new consensus for high affinity PRIs [S/G][I/L]LGx[AVP]P
Additionally, we have solved the structure of RRMs 3 and 4 of a neuronal homologue of
PTB, nPTB. As with the prototypic protein, the helical sides of RRMs 3 and 4 interact
forming a di‐domain with RNA binding surfaces on opposing sides.
mechanism used to amplify complexity of the metazoan proteome. The Polypyrimidine
Tract Binding protein (PTB) is a key regulator of alternative splicing. In addition, it has
roles in the localization and stability of mRNA and is involved in IRES mediated
translation initiation. It has four RNA recognition motif domains (RRMs) which consist
of two α‐helices which pack against a β‐sheet. The RRMs act in concert to direct splicing
as each RRM binds pyrimidine rich RNA (UCUU or CUCUCU) across the β‐sheet.
Solution structures have been solved of each RRM in complex with short hexameric
oligonucleotides. However, high‐resolution structural studies of larger protein
fragments with longer RNAs have been very difficult. Small angle X‐ray scattering has
been used to probe this, but sample aggregation obstructed collection of high quality
data. To address this we have investigated how longer RNAs bind PTB fragments. We
have determined optimal RNAs which should help to determine structures of PTB
fragments in complex with longer RNA molecules.
PTB is necessary for skipping Tpm1 exon 3 in smooth muscle cells. However, interaction
with an additional protein, Raver1, is required to accomplish this. Raver1 contains four
PTB‐Raver1 interaction motifs (PRIs) with a consensus of [S/G][I/L]LGxxP. The PRIs
bind to the α‐helical side of PTB RRM2. We have created chimeric fusion constructs
where Raver1 PRIs are encoded as N‐terminal extensions to PTB RRM2. High‐resolution
crystal structures have been solved of 2 PRIs bound to RRM2, one high affinity and one
low affinity. The differences between the conformations of each PRI bound to the RRM
explain differences in affinity. We have used structure‐based mutagenesis to probe these
differences and have revealed what is necessary for a high affinity interaction. From this
we have proposed a new consensus for high affinity PRIs [S/G][I/L]LGx[AVP]P
Additionally, we have solved the structure of RRMs 3 and 4 of a neuronal homologue of
PTB, nPTB. As with the prototypic protein, the helical sides of RRMs 3 and 4 interact
forming a di‐domain with RNA binding surfaces on opposing sides.
Date Issued
2012-01
Date Awarded
2012-05
Advisor
Curry, Stephen
Matthews, Steve
Publisher Department
Cell and Molecular Biology
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)