Elevated TIMP-1 expression is associated with a prometastatic phenotype, disease relapse, and poor survival in neuroblastoma

// Pritha Paul 1, 4, * , Eric J. Rellinger 1, 4, * , Jingbo Qiao 1, 4 , Sora Lee 1, 4 , Natasha Volny 1, 4 , Chandrasekhar Padmanabhan 1 , Carmelle V. Romain 1, 4 , Bret Mobley 3 , Hernan Correa 3 and Dai H. Chung 1, 2, 4 1 Section of Surgical Sciences, Department of Surgery, Vanderbilt University Medical Center, Nashville, TN 37232, USA 2 Department of Cancer Biology, Vanderbilt University Medical Center, Nashville, TN 37232, USA 3 Department of Pathology, Vanderbilt University Medical Center, Nashville, TN 37232, USA 4 Department of Pediatric Surgery, Vanderbilt University Medical Center, Nashville, TN 37232, USA * These authors have contributed equally to this work Correspondence to: Dai H. Chung, email: dai.chung@vanderbilt.edu Keywords: TIMP-1, neuroblastoma, metastasis, liver, LM2 Abbreviations: TIMP, tissue inhibitor of metalloproteinases Received: August 16, 2016      Accepted: May 04, 2017      Published: July 28, 2017 ABSTRACT Approximately two-thirds of patients with neuroblastoma are found to have metastatic disease at time of diagnosis with frequent skeletal, lymph node, central nervous system, and liver involvement. Using a serial in vivo splenic injection model, we have isolated an aggressive subclone (BE(2)-C/LM2) from MYCN -amplified neuroblastomas that demonstrate an enhanced propensity to develop metastatic liver lesions. BE(2)-C/LM2 subclone cells demonstrate increased adherent, soft agar colony and tumorsphere growth in vitro . Components of the tumor microenvironment regulate cancer progression, via networks of cytokines and growth factors. Cytokine array analysis identified increased TIMP-1 in the plasma of mice injected with BE(2)-C/LM2 subclone cells, leading us to hypothesize that TIMP-1 may play a role in our observed prometastatic phenotype. Immunoblotting and ELISA demonstrated enhanced endogenous TIMP-1 expression in our isolated neuroblastoma subclone. Silencing endogenous TIMP-1 successfully blocked in vitro proliferation, soft agar colony formation and tumorsphere formation by BE(2)-C/LM2 cells. Stable RNA interference of endogenous TIMP-1 failed to reverse the prometastatic phenotype of our BE(2)-C/LM2 subclone in our liver metastasis model, suggesting that endogenous TIMP-1 levels may not be an essential component of this in vivo behavior. Notably, tissue microarray analysis and Kaplan-Meier by gene expression demonstrates that elevated TIMP-1 expression is correlated with increased disease relapse and mortality in patients with neuroblastoma. Taken together, our study identifies TIMP-1 as a novel soluble factor that is associated with a prometastatic phenotype in our in vivo model and adverse outcomes in patients with neuroblastoma.