Angewandte
Chemie
DOI: 10.1002/anie.201202932
mRNA splicing
An RNA Splicing Enhancer that Does Not Act by Looping**
Helen Lewis, Andrew J. Perrett, Glenn A. Burley,* and Ian C. Eperon*
Mammalian pre-mRNA splicing exhibits an abundance of
alternative sites and permissible combinations. This expands
the coding possibilities of most genes by several-fold. The
splice site signals are poorly conserved and often weak, but
their use is augmented by interactions with proteins bound to
additional sequences in the introns or exons. These sequences
are known as splicing enhancers, and they can be found at
distances up to several hundred nucleotides (nt) from the
target splice sites.[1–4]
Most well-characterized exonic splicing enhancers (ESEs)
are bound by SR proteins. These contain RNA-binding
domains and a C-terminal domain rich in arginine-serine
dipeptides (RS domain). They stabilize the binding of com-
ponents that recognize the three canonical splicing signals:
U1 snRNPs,[5–7] which base-pair to 5’ splice sites, U2AF
protein,[8–10] which binds to 3’ splice sites, and U2 snRNPs,[11,12]
which base-pair to branch points. The accepted model for the
action of ESEs is that the RS domain encounters the target
protein or RNA duplex at a 5’ or 3’ splice site by 3D diffusion
and forms a protein-bridged loop in the intervening RNA
(Figure 1a). However, although this model is around 20 years
old, it has not been possible to test it definitively. It is
supported by two lines of evidence: 1) the rate of splicing (r)
of a model substrate with an RS domain tethered to an ESE
appeared to be related to the number of nt (n) between the
splice site and the ESE, as predicted for sites interacting by
3D diffusion (r/ nꢀ3/2);[3] 2) an ESE-tethered RS domain
could be cross-linked by UV light to RNA near a splice site,
demonstrating close proximity.[13] Neither of these results is
conclusive. Our analysis of the rate data[3] suggests that r/
nꢀ5/2 or r/ eꢀkn, where k is an arbitrary constant, neither of
which supports the diffusion model for free RNA. Moreover,
if entire SR proteins can bind the ESE then the effects of the
Figure 1. Diagrams of possible mechanisms of action of exonic
splicing enhancers (ESEs). a) Direct interactions by looping. An SR
protein (blue circle with an arm representing the arginine/serine-rich
domain) bound to an ESE (green) could interact directly with proteins
bound to a splice site (orange) through 3D diffusion. b) Indirect
effects transmitted by RNA. Proteins bound to the RNA might
propagate from the ESE to the target protein bound at the splice site.
c) Looping should not be prohibited by an intervening flexible linker,
for example, PEG. d) Effects transmitted along the RNA could be
prevented by a non-natural linker such as PEG.
length n are much reduced.[14] The cross-linking results do not
exclude the possibility that a number of SR proteins were
bound along the RNA, in a process initiated by the ESE[15]
(for example, as in Figure 1b).
The looping hypothesis necessarily entails the existence of
a flexible chain connecting an ESE and the target site, but the
chain need not be RNA. A definitive test would be to insert
a flexible non-RNA linker between two sites, which would
permit direct encounters by looping (Figure 1c) but block
indirect actions transmitted along the RNA (Figure 1d). This
test was used previously to test whether the contact of splice
sites across an intron involved looping. In this case, the
introduction of a poly(ethylenegylcol) (PEG) linker into the
intron did not inhibit splicing.[16] To apply this test to the
action of an ESE, we would need to insert a PEG linker
between the ESE and a splice site. The method used
previously incorporated DNA sequences flanking the PEG
linker, which could compromise the interpretation if the
actions of an ESE were inhibited by the DNA, since the cell
contains many DNA-binding proteins. To overcome this
limitation, we used click chemistry[17,18] to incorporate a PEG
linker into RNA.
[*] H. Lewis, A. J. Perrett
Department of Chemistry, University of Leicester
University Road, Leicester LE1 7RH (UK)
Dr. G. A. Burley
Department of Pure & Applied Chemistry, University of Strathclyde
295 Cathedral Street, Glasgow G1 1XL (UK)
E-mail: glenn.burley@strath.ac.uk
Prof. Dr. I. C. Eperon
Department of Biochemistry, University of Leicester
Lancaster Road, Leicester LE1 9HN (UK)
E-mail: eci@le.ac.uk
staff/eperon
The test substrate that we used is a pre-mRNA with two
possible 5’ splice sites (site 1 and 2; Figure 2). ESEs favor the
nearest 5’ splice site.[19–22] We incorporated a GGA-rich
ESE[23,24] at the 5’ end of an adenovirus-based pre-mRNA
with two alternative 5’ splice sites,[25] and an intervening non-
RNA linker was introduced using click chemistry (4,
Figure 2).
[**] This work was supported by the Leverhulme Trust (F00212Y). G.A.B.
thanks the EPSRC for an Advanced Fellowship (EP/E055095/1).
Supporting information for this article (experimental details) is
Angew. Chem. Int. Ed. 2012, 51, 1 – 5
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!