Background: Mesenchymal stem cells (MSCs) serve as an attractive vehicle for cell-directed enzyme prodrug therapy (CDEPT) due to their unique tumour-nesting ability. Such approach holds high therapeutic potential for treating solid tumours including glioblastoma multiforme (GBM), a devastating disease with limited effective treatment options. Currently, it is a common practice in research and clinical manufacturing to use viruses to deliver therapeutic genes into MSCs. However, this is limited by the inherent issues of safety, high cost and demanding manufacturing processes. The aim of this study is to identify a facile, scalable in production and highly efficient non-viral method to transiently engineer MSCs for prolonged and exceptionally high expression of a fused transgene: yeast cytosine deaminase::uracil phosphoribosyl-transferase::green fluorescent protein (CD::UPRT::GFP).
Methods: MSCs were transfected with linear polyethylenimine using a cpg-free plasmid encoding the transgene in the presence of a combination of fusogenic lipids and β tubulin deacetylase inhibitor (Enhancer). Process scalability was evaluated in various planar vessels and microcarrier-based bioreactor. The transfection efficiency was determined with flow cytometry, and the therapeutic efficacy of CD::UPRT::GFP expressing MSCs was evaluated in cocultures with temozolomide (TMZ)-sensitive or TMZ-resistant human glioblastoma cell lines. In the presence of 5-fluorocytosine (5FC), the 5-fluorouracil-mediated cytotoxicity was determined by performing colometric MTS assay. In vivo antitumor effects were examined by local injection into subcutaneous TMZ-resistant tumors implanted in the athymic nude mice.
Results: At > 90% transfection efficiency, the phenotype, differentiation potential and tumour tropism of MSCs were unaltered. High reproducibility was observed in all scales of transfection. The therapeutically modified MSCs displayed strong cytotoxicity towards both TMZ-sensitive and TMZ-resistant U251-MG and U87-MG cell lines only in the presence of 5FC. The effectiveness of this approach was further validated with other well-characterized and clinically annotated patient-derived GBM cells. Additionally, a long-term suppression (> 30 days) of the growth of a subcutaneous TMZ-resistant U-251MG tumour was demonstrated.
Conclusions: Collectively, this highly efficient non-viral workflow could potentially enable the scalable translation of therapeutically engineered MSC for the treatment of TMZ-resistant GBM and other applications beyond the scope of this study.
Keywords: Mesenchymal stem cells; Prodrug therapy; Temozolomide-resistant glioma; Transfection.