Meta-Mining of Gene Expression Profiles Identifies MG132 and Cantharidin as Therapeutic Agents with In Vitro Activity Against Leiomyosarcoma
B Edris, RB West, CH Lee, M Zhu, J Fletcher, M Van de Rijn, AH Beck. Stanford University, Palo Alto; Brigham and Women's Hospital, Boston; Univ. of British Columbia, Vancouver, Canada
Background: Few therapeutic agents exist for the treatment of leiomyosarcoma (LMS). In a prior study, we performed molecular profiling of LMS and identified 3 LMS subtypes. In the present study, we used the gene expression signatures of the LMS subtypes to identify novel therapeutic compounds for the treatment of LMS, and we used in vitro experiments to test their efficacy.
Design: Lists of the top 200 differentially expressed genes in each of the 3 LMS subtypes (n=49) compared with a set of leiomyomas (n=19) was uploaded into the Connectivity Map (CMap, www.broadinstitute.org/cmap), a resource designed to find connections between gene expression profiles and drug response signatures. We selected 4 highly ranked candidate therapeutic compounds with strong negative or positive enrichment with at least 1 of the LMS subtypes. We tested the effectiveness of the compounds against the 3 LMS cell lines. Experiments were carried out in quadruplicate using 2-fold serial dilutions of each drug. Viability was assessed using the XTT Cell Proliferation Kit II (Roche Diagnostics, Mannheim, Germany).
Results: Expression profiling demonstrated that the 3 LMS cell lines each showed the most similarity with LMS subtype II. Cantharidin, a protein phosphatase inhibitor, showed the strongest negative enrichment across all 3 subtypes in the CMap analysis and was effective in inhibiting the growth of the 3 LMS cell lines. Tyrophostin AG-825, an inhibitor of the HER-2/neu tyrosine kinase, showed a negative enrichment with LMS subtypes I and II had no significant effect on cell viability in the 3 LMS cell lines. MG132, a proteasome inhibitor with a strong positive enrichment with LMS subtype II, showed a strong negative effect on cell viability in the 3 LMS cell lines. Doxorubicin, the most commonly used chemotherapeutic treatment for LMS, showed moderately negative enrichment with LMS subtype II and showed a moderate, though non-uniform, inhibitory effect against the 3 LMS cell lines.
Conclusions: The CMap analysis revealed novel candidate drugs for the treatment of LMS. The drug sensitivity experiments demonstrated that MG132 and Cantharidin show activity against LMS cell lines. These experiments provide a starting point for further mechanistic studies of LMS drug sensitivities in vitro and in vivo and represent a first step towards the development of personalized therapies for LMS patients.
Category: Bone & Soft Tissue
Monday, March 22, 2010 1:00 PM
Poster Session II # 23, Monday Afternoon