Emerging Biotechnologies to Exploit Alternative Splicing in the Treatment of Disease

---

By Brendan Bell, PhD & Roscoe Klinck, PhD

Alternative splicing
One of the surprising discoveries following the sequencing of the human genome was the relatively modest number of protein-coding genes, currently estimated at approximately 25,000. Before the sequencing was completed, predictions in the range of 60,000 to 150,000 genes were common. Given the perceived genetic complexity of humans relative to simpler organisms such as yeast (ca. 6,000 genes) or the C. elegans nematode (ca. 20,000 genes), the expectation was for a considerably higher total gene count. What has since become clear, and which perhaps explains the low total gene count, is the prevalent contribution of alternative splicing (AS) to the information encoding capacity of the human genome.

AS is a post transcriptional process whereby the introns and exons of a single gene transcript are differentially spliced to yield multiple mature mRNAs. In eukaryotes, AS plays a central role in both protein diversity and post-transcriptional gene regulation. AS within mRNA coding regions can lead to multiple, functionally diverse, protein isoforms from a single gene transcript. AS can also act to introduce or remove regulatory elements to affect the translation, localization or degradation properties of an mRNA. The majority of human genes are thought to undergo AS, current estimates range from a conservative 60%, to as high as 88%. Thus, the vast array of proteins present in a human cell results not only from the number of human genes, but from the multiple products generated from those genes by AS (Figure 1).

What are the biological functions of alternative splicing? Although we are only beginning to unravel the full spectrum of the biological functions of AS, the DSCAM gene studied in the drosophila fruit fly provides an impressive example of the breadth and functional impact of AS in living organisms. The single DSCAM gene generates a potential repertoire of 38,016 transmembrane proteins via AS. The vastly diverse set of DSCAM isoforms is crucial in the ability of neurons to distinguish between one another in the intricate process of wiring of the developing nervous system1.

Alternative splicing and disease
Given the important and pervasive role of AS in gene expression, it is not surprising that changes in AS can be linked to human disease. Beta-thalassemia represents an early example linking AS to disease, where a mutation creating a aberrant splice site causes severe anemia, and potentially death, of affected patients2. A second example of the importance of AS in human disease is the Bcl-x gene. Bcl-x is involved in apoptosis, or programmed cell death, which is an essential cellular pathway that is inactivated in cancer cells. This gene has several splice variants including well-characterized short, Bcl-xS, and long, Bcl-xL isoforms. The two isoforms have opposite functions, the short form activates apoptosis, while the long isoform prevents it. The AS event regulating the expression of these two isoforms has been shown to be dysfunctional in cancer cells, such that the long isoform predominates, thus allowing cancer cells to proliferate. More generally, a study carried out in 1992 suggested that 15% of disease-causing mutations affect splicing directly as opposed to creating disruptive changes in protein sequence3. This study considered only mutations at splice sites, since then several examples of proximal mutations affecting AS have emerged. Furthermore, mutations in splicing factors lead to changes in AS patterns have been frequently associated with human diseases such as cancer4. The growing number of links between AS and human disease fuel the search for functions of the thousands of AS events for which no biological role is known, and importantly, for novel avenues to target AS therapeutically.

Splice switching oligonucleotides
The realization that AS plays an important role in proteome diversity and human disease presents two research avenues for the improvement of diagnosis, prognosis and treatment in medicine. The first is the functional characterization of the huge numbers of alternative splice variants for which no function has been assigned, all of which may potentially represent novel therapeutic targets. The second related avenue is the exploitation of key alternative splicing events to develop therapeutic agents for numerous diseases in which AS is implicated. An elegant strategy to manipulate, or reprogram, AS using modified antisense oligonucleotides has been developed (see Figure 2). These oligonucleotides contain chemical modifications that render them resistant to nucleases, thus prolonging their effective lifetimes in cells. An important advantage of these modified oligonucleotides is that, unlike unmodified antisense oligonucleotides or small interfering RNAs (siRNAs), they reprogram AS, rather than simply inactivate gene expression.

The first proof-of-principle that these splice switching oligonucleotides (SSOs) have therapeutic potential came from studies demonstrating that the treatment of cultured tumour cells with SSOs directed against the Bcl-x gene either sensitized the cells to chemotherapeutic agents5 or induced programmed cell death6. The technology can also be used to identify novel therapeutic targets. The Bell laboratory at the Université de Sherbrooke recently showed that the induction of a newly identified splice variant termed TAF6delta by SSOs efficiently kills tumour cells in culture7. An exciting aspect of this finding is that the tumour cells died even in the absence of the key tumour suppressor gene p53 that is often inactivated in human tumours. Several oligonucleotide chemistries, including 2’-O-methyl, 2’-O-methoxyethoxy, peptide nucleic acid, locked nucleic acid and morpholino have successfully been used to alter AS, suggesting a vast potential for the improvement of the potency and pharmacological properties of SSOs. Several clinical trials using SSOs are currently underway, one example being as a therapy for Duchenne muscular dystrophy. Given the prevalence of AS in human disease, one can optimistically predict that therapeutic applications of SSOs will continue to expand.

Large scale characterization of AS
The ubiquity of AS and its relevance to disease have prompted the development of large-scale techniques for the discovery, validation and functional analysis of AS in human cells. The Laboratory of Functional Genomics at the Université de Sherbrooke, headed by Dr. Sherif Abou Elela, has focussed specifically on this area of research. With funding from Genome Canada and Genome Quebec, the laboratory has developed an automated platform for AS characterization, termed the LISA (layered and integrated system for splicing isoform annotation). The LISA is a computational and experimental platform for the annotation and functional analysis of AS. The LISA comprises a four-step pipeline that takes selected genes from the process of AS annotation, through to functional characterization of isoforms expressed in relevant human cell lines.

Computational design and analysis modules are integrated with a robotic platform for the high throughput RT-PCR based experimental annotation of AS from selected RNA sources. A transcript map containing all publicly available mRNAs for each gene is generated and a set of PCR primers and experiments are designed such that all putative AS events are covered by at least two independent PCR reactions. PCR reactions are analyzed by automated capillary electrophoresis and the resulting amplicon size and quantitation data is transferred back to the LISA database. The experimental data is analyzed both to confirm proposed AS events, and to identify potential novel events.

The screening and annotation process yields AS events which show statistically significant differences between normal and tumour tissues. The next step is to screen related cell lines to identify this AS event, and to reprogram it using an SSO approach. The laboratory has developed a bifunctional SSO, termed TOSS (targeted oligonucleotide silencing of splicing) whereby the inhibition of splicing is enhanced by the recruitment of splicing factors to the one end of the SSO8. Cells are transfected with TOSS agents and the effect of this isoform specific inhibition is monitored at the RNA and protein levels. At the same time, analysis for cellular phenotypes is performed. The phenotypic assays are performed on the robotic platform using an automated confocal microscope to image the cells. Both single cells and populations of cells can be monitored following the reprogramming of splicing. The results are integrated into the LISA database for analysis. Typical assays include cell viability and growth, chemosensitivity, apoptosis and DNA damage checkpoint activation. Thus the system can link the function of a given splice variant to cell growth, cell cycle arrest or apoptosis. In addition, changes in cell adhesion, migration and invasion, all hallmarks of tumour progression and metastasis are routinely monitored using the LISA.

The LISA is currently being applied to explore the role of AS in breast and ovarian cancer. It has been shown that 7% of ovarian and breast cancer related genes harbour differentially expressed AS events in a panel of normal and cancerous ovarian tissues9. Similar results have been obtained in breast tissues, suggesting that several or all cancers are linked to AS deficiencies. Preliminary phenotyping experiments in breast and ovary derived cell lines have shown promise for the functional elucidation of several key cancer specific AS events, opening the way for large scale biomarker and drug target discovery using the LISA platform. The LISA technology has been made available to the academic and industrial communities as a Genome Quebec sponsored platform, and as such has proven to be a highly sensitive complement to microarray techniques.

Conclusion
It has recently become clear that alternative splicing plays a key role in gene expression with an impact on virtually all of the biological processes occurring within the human cell. As the application of large scale techniques for the discovery and functional analysis of AS becomes more widespread, the evidence for links between AS and human disease continues to grow exponentially. This in turn has ushered in a new era in drug target discovery. The fact that all AS events can, in principle, be targeted by various forms of SSO technology holds immense potential for new treatments for human disease. The combination of AS-focussed high-throughput functional screening and SSO technology for functional drug target discovery, promises to take an important place in the future of biotechnology.

References
1.     Hattori, D. et al. Dscam diversity is essential for neuronal wiring and self-recognition. Nature 449, 223-7 (2007).
2.     Busslinger, M., Moschonas, N. & Flavell, R. A. Beta + thalassemia: aberrant splicing results from a single point mutation in an intron. Cell 27, 289-98 (1981).
3.     Krawczak, M., Reiss, J. & Cooper, D. N. The mutational spectrum of single base-pair substitutions in mRNA splice junctions of human genes: causes and consequences. Hum Genet 90, 41-54 (1992).
4.     Garcia-Blanco, M. A., Baraniak, A. P. & Lasda, E. L. Alternative splicing in disease and therapy. Nat Biotechnol 22, 535-46 (2004).
5.     Taylor, J. K., Zhang, Q. Q., Wyatt, J. R. & Dean, N. M. Induction of endogenous Bcl-xS through the control of Bcl-x pre-mRNA splicing by antisense oligonucleotides. Nat Biotechnol 17, 1097-100 (1999).
6.     Mercatante, D. R., Bortner, C. D., Cidlowski, J. A. & Kole, R. Modification of alternative splicing of Bcl-x pre-mRNA in prostate and breast cancer cells. analysis of apoptosis and cell death. J Biol Chem 276, 16411-7 (2001).
7.     Wilhelm, E., Pellay, J. F., Bencecke, A. & Bell, B. TAF6delta orchestrates transcription programs and apoptosis in the absence of p53. Submitted for publication.
8.     Villemaire, J., Dion, I., Elela, S. A. & Chabot, B. Reprogramming alternative pre-messenger RNA splicing through the use of protein-binding antisense oligonucleotides. J Biol Chem 278, 50031-9 (2003).
9.     Klinck, R. et al. Multiple Alternative Splicing Markers for Ovarian Cancer. Cancer Res (in press).

Roscoe Klinck, PhD is the director, RNomics Platform, Laboratoire de Génomique Fonctionnelle de l¹Université de Sherbrooke Centre de développement des biotechnologies (CDB) de Sherbrooke, in Sherbrooke, QC.

Brendan Bell, PhD, Canada Research Chair Genomic Regulation Département de microbiologie et d’infectiologie, Faculté de médecine et des sciences de la santé, Université de Sherbrooke, in Sherbrooke, QC.