Targeting splicing in spinal muscular atrophy


Spinal muscular atrophy (SMA) is a devastating autosomal recessive motor neuron disease caused by insufficient expression levels of the survival of motor neuron (SMN) protein. With a carrier frequency of 1 in 50 and an incidence of 1 in 10,000 births, SMA is the most common inherited disease that is lethal to infants. Importantly, SMA has a strikingly broad range of phenotypic severity ranging from profound weakness at birth associated with death before 2 years to mild proximal weakness beginning in adulthood without a shortened life span.

In humans, the SMN protein is encoded by two genes located on chromosome 5, the telomeric SMN1 gene and the centromeric SMN2 gene.1 Because of deletion or missense mutations, all SMA patients lack a functional copy of the SMN1 gene but retain at least one copy of the SMN2 gene.2 Transcripts encoded by the SMN1 gene contain a normal complement of eight exons, but most of the transcripts encoded by the SMN2 gene are alternatively spliced to exclude exon 7.3, 4 SMN delta 7 transcripts code for a truncated SMN protein that fails to oligomerize or associate with its binding partners and is therefore rapidly degraded. A small proportion of transcripts arising from the SMN2 gene are full length and code for full-length SMN protein, but this reduced level of SMN protein expression is insufficient to prevent SMA in most patients. The number of copies of the SMN2 gene varies in the population but is usually between one and four copies. In SMA patients, increased SMN2 copy number is inversely associated with disease severity. Some individuals with four or five copies of the SMN2 gene may be phenotypically normal with a total loss of SMN1.5 Because the SMN2 gene substantially modifies SMA disease severity and is retained in all patients, it is a promising target for therapeutic development.6 Currently investigated treatment strategies that take advantage of the SMN2 gene to increase SMN protein levels are to activate SMN2 gene transcription or to stabilize SMN2-derived transcripts or protein. Another strategy, which Yuo and colleagues7 explore in this issue of the Annals, is to modify the splicing of SMN2-derived transcripts to increase inclusion of exon 7.

Splicing is the process of removal of intronic sequences from pre-messenger RNA (pre-mRNA) to generate mature mRNA molecules. This process is governed by cis-acting sequence motifs that interact with trans-acting splicing factors. The genomic sequences of the SMN1 and SMN2 genes differ by only five nucleotides. The critical sequence variant in SMN2 that alters splicing of SMN2-derived transcripts is a translationally silent C→T base-pair change that alters a splicing sequence motif (Fig). This motif present in exon 7 acts as an exonic splice enhancer (ESE) and recruits the SR protein SF2/ASF, which acts to direct exon 7 inclusion.8 When this motif is altered by the C→T transition, this ESE is disrupted, SF2/SAF fails to bind, and exon 7 is excluded. In addition, this same sequence motif may act as an exon splice silencer element recruiting the ribonucleoprotein, hnRNPA1, which acts to inhibit exon 7 inclusion (see Fig). A number of other cis- and trans-acting elements have also been shown to influence this process (see Fig). The understanding of these splicing mechanisms provides the opportunity to identify treatments that increase exon 7 inclusion. Previously, a drug screen showed that the chemotherapeutic aclarubicin increased exon 7 inclusion in SMA patient–derived cultured cells, but this drug was not considered a good candidate for human use because of its toxicity profile.9 Also under development are biologics such as chimeric molecules that contain antisense moieties fused to SR proteins,8 and trans-splicing exogenous RNA that correct endogenous mRNA sequences.10

Illustration 1.

Fig. Survival of motor neuron (SMN) exon 7 splicing is determined by the balance of multiple cis- and trans-acting elements. All mature transcripts derived from the SMN1 gene are full length (Fl-SMN), but most derived from the SMN2 gene lack exon 7 (SMNΔ7). In SMN1-derived pre-mRNA, a cytosine at position 6 of exon 7 is part of an exonic splicing enhancer (ESE) sequence motif that recruits the splicing factor SF2/AF2 promoting exon 7 inclusion. Htra2-β1 binds another ESE and, together with heterogenous nuclear ribonucleoprotein-G (hnRNP-G) and SRp30c, also directs exon 7 inclusion. The intronic splicing silencers (ISS), element 1 ISS and ISS-N1, are thought to negatively influence positive elements in exon 7. In SMN2-derived transcripts, the cytosine at position 6 of exon 7 is replaced by a uracil, which interrupts the ESE and creates an exonic splicing silencer (ESS) that recruits the splicing factor hnRNP A1 promoting exon 7 skipping. Yuo and colleagues7 in this issue of Annals suggest that SRp20 also acts to increase exon 7 inclusion by a mechanism that is yet to be determined. (Adapted from Sumner CJ. J Child Neurol 2007, by permission.)

In this issue of Annals, Yuo and colleagues7 report that SMN exon 7 inclusion and SMN protein levels are increased in SMA patient–derived lymphoid cells by inhibiting the Na+/H+ exchanger (NHE) with the amiloride derivative ethylisopropylamiloride (EIPA). Proton fluxes across biological membranes regulate a number of physiological processes, including communication within cells, cell migration, and the rate at which cells grow, divide, and differentiate. At the plasma membrane, these fluxes are regulated by several families of ion exchangers, including the NHE family of proteins and various bicarbonate transporters and exchangers. EIPA is thought to both competitively and noncompetitively inhibit the NHE, partly through binding at the extracellular cation binding site. It has previously been shown that reducing the pH of the extracellular medium is sufficient to stimulate alternative splicing in cultured cells.11–13 In this report, the authors advance this concept further by showing that treatment of cells with an NHE inhibitor reduces intracellular pH, and that this is associated with a shift in splicing of SMN2-derived transcripts to include exon 7. These effects are also associated with increased protein expression levels of one of the six splicing regulatory proteins examined, SRp20. These results suggest that cells respond to changes in intracellular pH by altering the expression levels of specific splicing factors, which subsequently act to regulate the splicing patterns of pre-mRNA molecules. Furthermore, although not previously implicated in SMN splicing, SRp20 may play a role in directing exon 7 inclusion in SMN2-derived transcripts.

NHE inhibitors have been most extensively studied as a potential treatment for heart disease. During ischemia, NHE1 is activated to exchange intracellular H+ for extracellular Na+ in response to the buildup of protons caused by anaerobic glycolysis. This increase in intracellular Na+ facilitates the subsequent increase in Ca2+ via the Na+/Ca2+ exchanger. These series of events ultimately triggers cell death cascades. Several studies using NHE inhibitors in animal models of heart disease have demonstrated a cardioprotective effect of NHE inhibition.14–16 Unfortunately, despite these early promising results, clinical trials have had limited success, partly because of deleterious side effects of the drugs currently available.17, 18 Further research is required to improve our understanding of the mechanism of action of these drugs and to develop better tolerated and more potent and/or specific inhibitors.

In summary, Yuo and colleagues7 have demonstrated that NHE inhibitor EIPA can decrease intracellular pH and increase exon 7 inclusion in SMN2-derived transcripts in cultured cells possibly via activation of expression of the splicing factor, SRp20. Although the current NHE inhibitors are likely not good candidates for treatment trials in SMA patients, Yuo and colleagues7 have uncovered a novel direction for therapeutic development in SMA. Important outstanding questions include whether the effects of EIPA on SMN splicing are directly caused by inhibition of the Na+/H+ transporter. The investigators found that only EIPA of several NHE inhibitors increased exon 7 inclusion. This might suggest that the SMN splicing effects are due to other activities of the drug that are independent of intracellular pH. Another important question is whether and how SRp20 regulates SMN2-derived transcript splicing. Interestingly, one previous study showed that SRp20 function is antagonized by ASF/SF2, an SR protein that is well known to increase SMN exon 7 inclusion.19 Growing evidence supports an intricate balance among several regulator elements in orchestrating SMN2 gene splicing. Continued dissection of these regulators and how they interact with one another is likely to yield new targets for therapeutic development in SMA.