Scoring of senescence signalling in multiple human tumour gene expression datasets, identification of a correlation between senescence score and drug toxicity in the NCI60 panel and a pro-inflammatory signature correlating with survival advantage in peritoneal mesothelioma
1 Centre for Oncology and Applied Pharmacology, University of Glasgow, Cancer Research UK Beatson Laboratories, Garscube Estate, Switchback Road, Bearsden, Glasgow G61 1BD, UK
2 Molecular Pharmacology Unit, Department of Experimental Oncology and Laboratories, Fondazione IRCCS Istituto Nazionale dei Tumori, Via Venezian, 1, 20133 Milan, Italy
3 Cancer Research Technology Limited, Wolfson Institute for Biomedical research, Gower Street, London, WC1E 6BT, UK
4 CompChem Solutions, St John's Innovation Centre, Cowley Road, Cambridge, CB4 0WS, UK
BMC Genomics 2010, 11:532 doi:10.1186/1471-2164-11-532Published: 1 October 2010
Cellular senescence is a major barrier to tumour progression, though its role in pathogenesis of cancer and other diseases is poorly understood in vivo. Improved understanding of the degree to which latent senescence signalling persists in tumours might identify intervention strategies to provoke "accelerated senescence" responses as a therapeutic outcome. Senescence involves convergence of multiple pathways and requires ongoing dynamic signalling throughout its establishment and maintenance. Recent discovery of several new markers allows for an expression profiling approach to study specific senescence phenotypes in relevant tissue samples. We adopted a "senescence scoring" methodology based on expression profiles of multiple senescence markers to examine the degree to which signals of damage-associated or secretory senescence persist in various human tumours.
We first show that scoring captures differential induction of damage or inflammatory pathways in a series of public datasets involving radiotherapy of colon adenocarcinoma, chemotherapy of breast cancer cells, replicative senescence of mesenchymal stem cells, and progression of melanoma. We extended these results to investigate correlations between senescence score and growth inhibition in response to ~1500 compounds in the NCI60 panel. Scoring of our own mesenchymal tumour dataset highlighted differential expression of secretory signalling pathways between distinct subgroups of MPNST, liposarcomas and peritoneal mesothelioma. Furthermore, a pro-inflammatory signature yielded by hierarchical clustering of secretory markers showed prognostic significance in mesothelioma.
We find that "senescence scoring" accurately reports senescence signalling in a variety of situations where senescence would be expected to occur and highlights differential expression of damage associated and secretory senescence pathways in a context-dependent manner.