Tumor dormancy is an important research topic because of its potential implications in metastasis and acquired resistance to chemotherapy. Dormant tumor cells at the secondary site may persist after chemotherapy and regrow as metastatic recurrence. Seminal studies by Dr. Bissell and Dr. Hendrix have shown that the tissue microenvironment can reprogram tumor cells into dormancy. Extracellular matrix (ECM) proteins are key components of tumor microenvironment and may regulate dormancy, and disruption of ECM homeostasis may cause recurrence and metastatic growth.

A major cytokine that regulates ECM remodeling is tumor growth factor-beta (TGF-beta), and it is implicated in modulating tumor cell dormancy. In the tissue microenvironment, serine protease HtrA1 was shown to inhibit TGF-beta signaling by targeting TGF-beta family ligands and receptors for proteolysis. Our lab showed that HtrA1 is down-regulated in ovarian cancer cells, and HtrA1 loss promotes resistance to anoikis and in vivo metastasis. Loss of HtrA1 is also associated with increased incidence of carcinomatosis. Finally, previous studies have shown that HtrA1 degrades ECM proteins. Therefore, we hypothesize that loss of HtrA1 in the ovarian tumor microenvironment enhances TGF-beta signaling and promotes ECM protein production by stromal cells, thereby contributing to ECM remodeling and tumor dormancy.

Ovarian cancer cells with endogenous HtrA1 were transduced with HtrA1-targeting shRNA (sh1) to evaluate the effect of HtrA1 downregulation on invasion, tumor dormancy, and ECM remodeling. Cells transduced with non-targeting shRNA (NT) served as a control. Cancer cells (NT and sh1) were plated over confluent mesothelial cell layer with or without 10 nM TGF-beta for 24 hours and 48 hours. No significant difference in cell invasion was observed between those cells cultured with or without TGF-beta. In order to test which matrix proteins are critical for cell proliferation versus cell cycle arrest, MicroMatrix 36 arrays were used to culture NT and sh1 cells. Cells were cultured for 48 hours on matrix slides, then fixed and stained with anti Ki67. It was determined matrix containing combinations of fibronectin showed cells that stained positive for Ki67, when compared to those not containing fibronectin or collagen together. Changes in intracellular signaling in the presence and absence of matrix were also investigated using kinase and MAPK arrays. Mesothelial cells were cultured until confluent, and cells were removed from culture plates using 2.0 mM EDTA in PBS. Matrix produced from mesothelial cells remained on the plates. NT and sh1 cells were then plated in the presence and absence of matrix and were allowed to culture until cell attachment was observed (3 hours at 37oC, 5% CO2). Cells were lysed and lysates were cultured on arrays according to manufactures instructions. ERK and Akt pathways were observed to be activated in cells plated with matrix compared to those without matrix. Taken together the evidence thus far suggests that matrix composition is critical for cell proliferation. Currently, Stable Isotope Amino Acid labeling is being employed to investigate changes in matrix remodeling on a global scale. In addition, Imaging Mass Spectrometry will also be used to evaluate formalin fixed paraffin embedded tissue samples from patients to identify at a molecular level changes in extracellular matrix composition between tumor and normal tissues.

Citation Format: Cooley, Megan K., Chien, Jeremy. Regulation of tumor dormancy by extracellular matrix remodeling as a result of increased TGF-beta signaling and loss of HtrA1 in ovarian cancer [abstract]. In: Proceedings of the 10th Biennial Ovarian Cancer Research Symposium; Sep 8-9, 2014; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2015;21(16 Suppl):Abstract nr POSTER-BIOL-1310.