2.4 Alternative Splicing
Even though the molecular mechanisms of intrinsic pre-mRNA splicing have
been well-characterised, it is not well understood how splicing events are directed
to determine isoform specificity in selecting between different alternative splicing
outcomes [27]. It is appreciated that splicing outcomes are influenced by the
dynamic interplay of splicing regulatory factors, histone marks and chromatinbinding proteins and that lncRNAs could also have key roles in selectively directing
splicing events towards the production of specific isoforms [28]. Among the
lncRNAs that participate in the formation of nuclear subcompartments, some
may also regulate splicing factors. MALAT1 (metastasis-associated lung adenocarcinoma transcript 1) forms nuclear speckles in which it preferentially colocalises
with serine/arginine (SR) splicing regulatory factors in their phosphorylated forms,
most prominently with SRSF1. It is thought that MALAT1 influences alternative
splicing by sequestering splicing factors in nuclear speckles [29] (Fig. 1d). A second
lncRNA with a well-established influence on alternative splicing regulation is MIAT
(myocardial infarction-associated transcript) [30], sometimes alternatively named
Gomafu. MIAT directly binds splicing factor 1 (SF1) via a repeat SF1-interacting
motif that is homologous to the optimal SF1-binding intronic branch point sequence
and thus has a higher affinity for SF1 than endogenous, suboptimal SF1-binding
branch points [31]. It was demonstrated that repeat oligos mimicking the MIAT
SF1-binding region can delay the onset of splicing of a pre-mRNA substrate
containing a suboptimal SF1-interacting branch point, but had no effect on the
splicing rate of a substrate containing a strong SF1-binding branch point [31].
MIAT can also bind to other splicing factors including Celf-3 and QKI, further
corroborating the notion that MIAT, like MALAT1, can also sequester splicing
factors to alter the kinetics of alternative splicing events [32, 33].
2.5 Crosstalk with miRNAs: Competing Endogenous ncRNAs
It is thought that RNA molecules can influence each other’s levels by competing for
a limited pool of miRNAs, an idea that has given rise to the concept of a “regulatory
language” of miRNA-RNA crosstalk [34]. The competing endogenous RNA
(ceRNA) hypothesis postulates that protein-coding mRNA transcripts and ncRNAs
can act by the reverse logic of miRNA-mediated mRNA suppression to crossregulate the expression of other RNA molecules with which they share miRNA
recognition elements (MREs) by competing for occupancy of mutually recognised
miRNAs (Fig. 1e). The elucidation ceRNA interaction networks would thus depend
on identifying crosstalking ceRNAs and characterising the precise number and
location of MREs that they contain, which would be the “letters” of the RNA
cross-regulatory code. Moreover, three conditions that influence the significance
of ceRNA regulatory events were outlined: (1) the expression level changes of any
408
D. Hanly et al.
Even though the molecular mechanisms of intrinsic pre-mRNA splicing have
been well-characterised, it is not well understood how splicing events are directed
to determine isoform specificity in selecting between different alternative splicing
outcomes [27]. It is appreciated that splicing outcomes are influenced by the
dynamic interplay of splicing regulatory factors, histone marks and chromatinbinding proteins and that lncRNAs could also have key roles in selectively directing
splicing events towards the production of specific isoforms [28]. Among the
lncRNAs that participate in the formation of nuclear subcompartments, some
may also regulate splicing factors. MALAT1 (metastasis-associated lung adenocarcinoma transcript 1) forms nuclear speckles in which it preferentially colocalises
with serine/arginine (SR) splicing regulatory factors in their phosphorylated forms,
most prominently with SRSF1. It is thought that MALAT1 influences alternative
splicing by sequestering splicing factors in nuclear speckles [29] (Fig. 1d). A second
lncRNA with a well-established influence on alternative splicing regulation is MIAT
(myocardial infarction-associated transcript) [30], sometimes alternatively named
Gomafu. MIAT directly binds splicing factor 1 (SF1) via a repeat SF1-interacting
motif that is homologous to the optimal SF1-binding intronic branch point sequence
and thus has a higher affinity for SF1 than endogenous, suboptimal SF1-binding
branch points [31]. It was demonstrated that repeat oligos mimicking the MIAT
SF1-binding region can delay the onset of splicing of a pre-mRNA substrate
containing a suboptimal SF1-interacting branch point, but had no effect on the
splicing rate of a substrate containing a strong SF1-binding branch point [31].
MIAT can also bind to other splicing factors including Celf-3 and QKI, further
corroborating the notion that MIAT, like MALAT1, can also sequester splicing
factors to alter the kinetics of alternative splicing events [32, 33].
2.5 Crosstalk with miRNAs: Competing Endogenous ncRNAs
It is thought that RNA molecules can influence each other’s levels by competing for
a limited pool of miRNAs, an idea that has given rise to the concept of a “regulatory
language” of miRNA-RNA crosstalk [34]. The competing endogenous RNA
(ceRNA) hypothesis postulates that protein-coding mRNA transcripts and ncRNAs
can act by the reverse logic of miRNA-mediated mRNA suppression to crossregulate the expression of other RNA molecules with which they share miRNA
recognition elements (MREs) by competing for occupancy of mutually recognised
miRNAs (Fig. 1e). The elucidation ceRNA interaction networks would thus depend
on identifying crosstalking ceRNAs and characterising the precise number and
location of MREs that they contain, which would be the “letters” of the RNA
cross-regulatory code. Moreover, three conditions that influence the significance
of ceRNA regulatory events were outlined: (1) the expression level changes of any
408
D. Hanly et al.