RNA splicing dysregulation and the hallmarks of cancer

Abstract

Dysregulated RNA splicing is a molecular feature that characterizes almost all tumour types. Cancer-associated splicing alterations arise from both recurrent mutations and altered expression of trans-acting factors governing splicing catalysis and regulation. Cancer-associated splicing dysregulation can promote tumorigenesis via diverse mechanisms, contributing to increased cell proliferation, decreased apoptosis, enhanced migration and metastatic potential, resistance to chemotherapy and evasion of immune surveillance. Recent studies have identified specific cancer-associated isoforms that play critical roles in cancer cell transformation and growth and demonstrated the therapeutic benefits of correcting or otherwise antagonizing such cancer-associated mRNA isoforms. Clinical-grade small molecules that modulate or inhibit RNA splicing have similarly been developed as promising anticancer therapeutics. Here, we review splicing alterations characteristic of cancer cell transcriptomes, dysregulated splicing's contributions to tumour initiation and progression, and existing and emerging approaches for targeting splicing for cancer therapy. Finally, we discuss the outstanding questions and challenges that must be addressed to translate these findings into the clinic.

Figure 1.. Principles of constitutive and alternative splicing.

(a) Stepwise assembly of spliceosomal complexes on a pre-mRNA molecule and catalysis of the splicing reaction to generate mature spliced mRNA. During the first step of the splicing reaction, the ATP-independent binding of U1 snRNP to the 5’SS initiates the assembly of the E complex, while SF1 and U2AF2 bind, respectively, to the BPS and Py-tract. In the second step, the ATP-dependent interaction of U2 snRNP with the BPS, stabilized by U2AF2–U2AF1 and SF3a–SF3b complexes, leads to A complex formation and SF1 displacement from the BPS. The recruitment of the U4/U6/U5 tri-snRNP complex marks the formation of the catalytically inactive B complex. The active B* complex is formed following major conformational changes, including release of U1 and U4, and the first catalytic step generates the C complex and results in lariat formation. The C complex performs the second catalytic step, which results in joining of the two exons. The spliceosome then disassembles releasing the mRNA and the lariat bound by U2/U5/U6.Spliceosomal core factors that exhibit alterations in human tumors are colored next to each complex. (b) Alternative splicing patterns are classified into cassette alternative exon splicing, alternative 5’ and 3’ splice site usage, mutually exclusive exons, and intron retention. These splicing patterns lead to distinct spliced mRNA isoforms that can be translated into protein isoforms with distinct sequences and functions. (c) Trans-acting regulatory splicing factors act as splicing activator (A) or repressor (R) and promote or inhibit spliceosome assembly by binding enhancers (ESE/ESS) or silencers (ISE/ISS) cis-acting regulatory sequences. 5’/3’ ASS: 5’/3’ alternative splice site; BPS: branch point site; Py-tract: polypyrimidine tract; snRNP: small nuclear ribonucleoprotein.

Figure 2.. Recurrent splicing factor alterations in cancer.

(a) Examples of SFs frequently upregulated, downregulated, or mutated in human primary tumors shown per tumor type. (b) Recurrent hotspot mutations in components from the spliceosomal A complex detected in human malignancies (BRCA-breast cancer, CLL-chronic lymphocytic leukemia, CMML-chronic myelomonocytic leukemia, MDS-myelodysplastic syndromes, PDAC-pancreatic adenocarcinoma, UVM-uveal melanoma). Positions of recurrent mutations are indicated along with the protein structures and domains (Zn- zinc finger domain, UHM-U2AF homology motif domain, RS-arginine/serine-rich domain, RRM-RNA-recognition motif, HD-heat domain). ZRSR2 mutations primarily affect U12-type introns, but as ZRSR2 has been biochemically implicated in U2-type splicing as well, it is illustrated in association with a U2-type intron above.

Figure 3.. Splicing hallmarks of cancer.

Examples of spliced isoforms implicated in the regulation of critical cellular processes defined as the Cancer Hallmarks by Weinberg and Hanahan^,. Note that the cancer hallmark ‘Polymorphic microbiomes’ is not included here.

Figure 4.. Splicing-driven alterations in drug responses.

Examples of alternatively spliced isoforms associated with altered response or resistance to targeted therapies, including isoforms that confer resistance to therapies targeting HER2 (a) or the androgen receptor (b), as well as to chimeric antigen receptor (CAR) T cells (c,d). AR, androgen receptor; Ex, exon; FL, full length.

Figure 5.. Therapeutic approaches to target splicing in cancer.

Current strategies either target (a) the splicing machinery itself or (b) the aberrantly spliced isoforms expressed in tumor cells. (a) Approaches targeting the spliceosome and splicing factors (SFs) include broad-spectrum inhibition or modulation as well as SF-specific inhibition both directly or through inhibition of upstream regulators of post-translational modifications (e.g., targeting methylation (Me), phosphorylation (P), or ubiquitination (Ub) processes). (b) Modulation of specific isoforms can be achieved using small molecules, splice-switching antisense oligonucleotides (ASOs), DNA- or RNA-targeting Cas with CRISPR-based approaches, or engineered small nuclear RNAs (snRNAs).