|Home | About | Journals | Submit | Contact Us | Français|
Close to 90% of human genes are transcribed into pre-mRNA that undergoes alternative splicing, producing multiple mRNAs and proteins from single genes. This process is largely responsible for human proteome diversity, and about half of genetic disease-causing mutations affect splicing. Splice-switching oligonucleotides (SSOs) comprise an emerging class of antisense therapeutics that modify gene expression by directing pre-mRNA splice site usage. Bauman et al. investigated an SSO that upregulated the expression of an anti-cancer splice variant while simultaneously eliminating an overexpressed cancer-causing splice variant. This was accomplished by targeting pre-mRNA of the apoptotic regulator Bcl-x, which is alternatively spliced to express anti- and pro-apoptotic splice variants Bcl-xL and Bcl-xS, respectively. High expression of Bcl-xL is a hallmark of many cancers and is considered a general mechanism used by cancer cells to evade apoptosis. Redirection of Bcl-x pre-mRNA splicing from Bcl-xL to -xS by SSO induced apoptotic and chemosensitizing effects in various cancer cell lines. Importantly, the paper shows that delivery of Bcl-x SSO using a lipid nanoparticle redirected Bcl-x splicing and reduced tumor burden in melanoma lung metastases. This was the first demonstration of SSO efficacy in tumors in vivo. SSOs are not limited to be solely potential anti-cancer drugs. SSOs were first applied to repair aberrant splicing in thalassemia, a genetic disease, they have been used to create novel proteins (e.g., Δ7TNFR1), and they have recently progressed to clinical trials for patients with Duchenne muscular dystrophy.
Pre-mRNA splicing, the co-transcriptional process of intron removal and exon joining seen in higher eukaryotes, was first thought to be an exception to the rule that mRNAs are carbon copies of their respective genes. Sequencing of the human genome revealed that not only splicing but also alternative splicing, which enables a single pre-mRNA transcript to give rise to multiple distinct splicing isoforms and therefore multiple variant proteins, are the norm. Alternative splicing explains how the relatively limited human genome (~30,000 genes) gives rise to more than 100,000 proteins. Recent analyses indicate that nearly 95% of all multi-exon gene transcripts undergo alternative splicing, 86% with a minor isoform frequency of 15% or more.1 In addition to regulating gene expression, pre-mRNA splicing and alternative splicing are being recognized for their roles in human disease. Indeed, up to half of disease-causing genetic mutations affect pre-mRNA splicing.2 The near-ubiquity of alternative splicing and the prevalence of splicing aberrations in human disease open up a new field that can result in the harvest of new gene- and sequence-specific medicines that achieve clinical benefits by modulating splicing. The SSO described in the Bauman et al. paper has the potential to be one such medication.
Splice-switching oligonucleotides (SSOs) comprise an emerging class of antisense drug candidates that modify pre-mRNA splicing. They are designed to block sequences in pre-mRNA essential for splicing, redirecting the splicing machinery to a new pathway. That this mechanism is feasible in vitro, in vivo in animals, and in patients has been demonstrated in numerous publications from this laboratory and by others.3–7
The SSO mechanism of action is distinct from conventional antisense oligonucleotides (ASOs) and siRNAs that inhibit gene expression by degrading the target mRNA through RNase H and the RISC complex, respectively. Instead, SSOs sterically block sequences in pre-mRNA without leading to their degradation, thereby shifting the pattern of splicing. This action requires that the SSO form very stable duplexes with its pre-mRNA target to enable successful redirection of the splicing factors to alternative, desired sequences. These requirements are met by incorporating various chemical modifications to the oligonucleotide backbone that improve binding affinity and do not support RNase H cleavage. Synthetic SSO chemistries include 2′-O-methyl, 2′-O-methoxyethyl (2′MOE), phosphorodiamidate morpholino, peptide nucleic acid and locked nucleic acid.8,9
Defective human cells can be pushed onto a suicide pathway through a programmed cell death mechanism called apoptosis. Apoptosis is essential for normal development and maintenance of tissue homeostasis; however, dysregulation of apoptotic signaling contributes to numerous pathological conditions, including cancer. Indeed, one of the hallmarks of cancer cells is their ability to evade and corrupt apoptosis, prompting the search for drugs that can restore or potentiate apoptotic signaling in tumor cells.10
Apoptosis is regulated by interactions among members of the Bcl-2 family of proteins, some of which (e.g., Bax and Bak) are pro-apoptotic and induce programmed cell death. Some are anti-apoptotic (e.g., Bcl-2 and Bcl-xL), and prevent the cell from committing suicide unnecessarily. A fine balance is maintained until an apoptotic event, such as DNA damage or growth factor deprivation, prompt Bax and Bak to trigger apoptosis.11,12 However, if anti-apoptotic Bcl-2 and/or Bcl-xL proteins are overexpressed they can overcome Bax and Bak, enabling the cell to survive with damaged or rearranged DNA and in the absence of growth factors. The cell is converted into a cancer cell that is free to proliferate, escaping normal controls. Because many chemotherapeutics induce apoptosis through Bax and Bak-like proteins, cancer cells with excess of Bcl-2 and Bcl-xL become resistant to chemotherapy.13 It follows that a drug that can reactivate the apoptotic pathway could be a good anti-cancer agent in its own right and/or could re-sensitize the cancer cell to chemotherapy.
Bcl-x undergoes splicing at two alternative 5′ splice sites of exon 2 (Fig. 1), yielding two protein with opposing functions, Bcl-xL and -xS.15 Use of the downstream splice site produces Bcl-xL, which exerts an anti-apoptotic effect by antagonizing and inhibiting the pro-apoptotic Bax and Bak proteins. It also induces growth of blood vessels that vascularize the tumor.16 Not surprisingly, Bcl-xL overexpression has been detected in a number of malignancies. A recent study of 3,131 cancer samples across more than two dozen cancer types found that the bcl-x gene was amplified in almost all of them, suggesting that upregulation of this gene may be a common mechanism by which cancers increase survival.17 Previously, this laboratory targeted Bcl-x in prostate and breast cancer cells,18,19 whereas the work discussed here focused on Bcl-x in metastatic melanoma.14
Use of the upstream 5′ alternative splice site of Bcl-x exon 2 produces an internally deleted protein, Bcl-xS (Fig. 1).15 Bcl-xS is anti-apoptotic because it directly binds and inhibits the pro-apoptotic Bcl-xL and Bcl-2 proteins.20 The authors show that targeting the downstream (Bcl-xL) splice site with a 2′MOE SSO redirects the splicing machinery to the upstream (Bcl-xS) splice site, effectively converting an anti-apoptotic molecule into a pro-apoptotic molecule. As a result, the tumor load in treated mice is reduced. Previously it was shown that newly generated Bcl-xS induced by SSO also sensitizes cancer cells to treatment with UV- and γ-irradiation and chemotherapeutic drugs, including etoposide, 5-fluorouracil, cisplatin, 5-fluorodeoxyuridine and doxorubicin.18,21
Two significant characteristics of splicing modulation that distinguish this approach from downregulation of anti-apoptotic genes by ASOs and siRNA are worth highlighting.22 First, in every instance of SSO-redirected splicing of Bcl-x pre-mRNA from Bcl-xL to Bcl-xS an anti-apoptotic Bcl-xL molecule is eliminated and the pro-apoptotic Bcl-xS molecule appears. The latter, as mentioned above, will inhibit existing Bcl-xL and Bcl-2 proteins, further reducing resistance of cancer cells to chemotherapy and apoptosis. The SSO thus produces a double or triple bang for the buck.
Second, the higher the level of expression of Bcl-x pre-mRNA in a cell, the more SSO-induced, anti-apoptotic Bcl-xS molecules that will be produced. This suggests that cells from aggressive cancers with higher levels of Bcl-x expression will be more susceptible to SSO-induced apoptosis than healthy, untransformed cells. This counterintuitive mechanism should thus reduce the undesirable side effects that are the bane of classic chemotherapeutics. Cancers with high Bcl-xL expression and those that depend on Bcl-xL expression for survival therefore represent good candidates for treatment with SSOs targeting Bcl-x.19 In contrast, RNase H-sensitive ASOs and siRNA become less effective as their target gene expression increases because as more target mRNA is expressed incomplete inhibition of target gene expression becomes more consequential.
For the above mechanisms to be fully exploited, obstacles associated with the delivery of macromolecules to tumors must be overcome. To this end, Bauman et al. delivered the Bcl-x SSO using lipid nanoparticles previously shown to be effective and well-tolerated in siRNA delivery to subcutaneous and lung tumor xenografts.23–25 This treatment resulted in significant redirection of Bcl-x splicing in lung metastases and significant reduction in tumor burden.14 Such effects were not seen in animals treated with vehicle only, SSO only or nanoparticles formulated with a control SSO. The paper presents the first demonstration of sequence-specific SSO efficacy in tumors in vivo. Table 1 shows other potential targets for SSOs as anti-cancer drugs, indicating that this approach is not limited to the Bcl-x gene alone.