Abstract PD8-07: A pro-metastatic alternative splicing program that drives breast cancer progression

Abstract

Abstract Background: Breast cancer progression accompanies broad reprogramming of the alternative splicing landscape with functional implications for tumor growth and metastasis. However, the underlying regulatory pathways that are hijacked by cancer cells to drive pathological splicing programs remain largely unknown. Here, we have described a novel regulatory pathway, mediated by the splicing factor SNRPA1, that promotes breast cancer progression. Results: We used high-coverage RNA sequencing to compare alternative splicing patterns in highly metastatic breast cancer cells (MDA-LM2) to their less metastatic parental line (MDA-231). We identified a novel structural splicing enhancer (SSE) element that is enriched among the exons up-regulated in highly metastatic cells and showed that it is bound by the RNA-binding protein SNRPA1, which in turn acts as a splicing factor. We used loss-of-function and gain-of-function both in vitro and in vivo xenograft mouse models to establish SNRPA1 as a promoter of breast cancer cell invasion (~2-fold change in two independent cell lines, P=0.002) and metastasis (~5-10 fold change in lung colonization in two independent cell lines, P<0.01). We also identified specific splicing events downstream of SNRPA1 that are required for its pro-metastatic activity. Single and multivariate analyses of publicly available datasets established that the higher expression of SNRPA1 is significantly associated with metastatic progression (METABRIC: n=1456, P<1e-5; EXPO: n=243, P=0.002), as well as reduced metastasis-free (kmplot: n=1746, HR=1.4, P<1e-3) and disease-free survival (METABRIC: n=1979, HR=1.7, P<1e-6; multivariate: HR=1.3, P=0.01). Finally, we measured the expression of SNRPA1 and its target exon in 96 samples from normal tissue biopsy to stage IV metastatic disease and observed a concordant increase in the activity of this novel pro-metastatic splicing program (ANOVA: P<1e-10; U test stage IV to N: P<1e-3). Methods: We used a framework called ‘mixture of isoforms’ (MISO) to quantify changes in alternative splicing. We then applied iTEISER, a computational tool we have developed, to find cis-regulatory elements that drive alternative splicing patterns. We used bait RNA to co-precipitate and mass-spectrometry to identify the RNA-binding proteins that interact with SSEs. Lentiviral delivery of shRNAs (two independent hairpins) or transgenes to modulate SNRPA1 expression and RNA-seq was used to measure their regulatory consequences on splicing. For the resulting lines (both MDA-231 and HCC-1806 backgrounds), in vitro cancer cell invasion and in vivo lung colonization assays were also performed (n=5 mice in each arm). Log-rank test (univariate) and Cox Proportional Hazard Models (multivariate) were used to perform survival analyses in METABRIC and the kmplot aggregate dataset. Mann-Whitney or ANOVA was used to compare expression of SNRPA1 across samples stratified based on sub-type and tumor grade/stage in public datasets as well as our own measurements in clinical samples (n=96; 5 healthy, 23 stage I, 30 stage II, 29 stage III, and 9 stage IV). Conclusion: In sum, we have discovered a previously unknown regulatory pathway of RNA splicing that acts as a promoter of metastasis in breast, which may inform new therapeutic targets among the SNRPA1 regulon for adjuvant therapy. Citation Format: Hani Goodarzi, Lisa Fish, Bruce Culbertson. A pro-metastatic alternative splicing program that drives breast cancer progression [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr PD8-07.