In silico-identified putative inhibitors of paraspeckle assembly induce cell death by apoptosis in Multiple Myeloma cells and lead to the activation of the cGAS/STING pathway.

Background: Paraspeckles (PSs) are membraneless nuclear bodies formed by the interaction between the lncRNA NEAT1 and PS proteins (PSPs), among which NONO and SFPQ. We reported that NEAT1 is significantly overexpressed in plasma cells of multiple myeloma (MM) patients, and that its silencing disrupts PSs, reducing MM cells proliferation and viability. On these bases we hypothesized that PSs may represent a novel targetable vulnerability in MM, which remains an incurable disease. Hence, a better understanding of PSs involvement in MM pathogenesis could have a significant impact, paving the way to innovative combinatorial therapeutic strategies. Aim: Given the translational limitations of RNA-targeting, we decided to disrupt PS structure through approaches with greater future translational potential, thus focusing on the protein components. Starting from an in-silico screening of drug-like small molecules targeting key interaction points between NEAT1, NONO and SFPQ, we aimed to characterize in vitro the anti-MM activity of the top two compounds (I5 and I6). These compounds are thought to mimic the effect of NEAT1 silencing. Methods: The biological activity of the 2 compounds was evaluated in a panel of 6 cytogenetically different MM cell lines (HMCLs) and 4 haematological, not-HMCLs. Cell viability was investigated through CCK-8 assay. Confocal microscopy of combined NEAT1 RNA-FISH with NONO IF and WB were used to determine PSs structural modifications upon treatments. Apoptosis was assessed by Caspase 3/7 Glo Assay and FACS analysis. Transcriptome analysis was performed using RNA-Seq. Drug synergism evaluation was conducted with CompuSyn software. Results: We highlighted a significant difference in IC50 values among MM and not-MM cell lines after 5 days of treatment (I5: HMCLs median= 1.28µM, Not-HMCLs median= 4.76µM; I6: HMCLs median= 2.14µM, Not-HMCLs median= 8.42µM), suggesting a specific anti-MM activity of both inhibitors. The significant reduction in cell viability was confirmed in 3 HMCLs through CCK-8 assay. In line with effector caspases activation, FACS analysis revealed a dose-dependent increase of both early and late apoptotic cells (≈2/5-fold increment, depending on the treated HMCL), suggesting a pro-apoptotic effect of both inhibitors on MM cells. Importantly, the presence of toxic off-target effects was excluded by analysing cell viability and apoptosis induction in healthy donors-derived activated PBMC samples. In addition to a ≈45% reduction in the number of PSs per cell in treated HMCLs, which confirmed the on-target activity of both inhibitors, we observed the post-transcriptional downregulation of core-localizing PSPs, NONO and SFPQ. Finally, RNA-Seq analysis of two HMCLs treated with both inhibitors revealed significant positive modulation of the cytosolic DNA sensing pathway. Given the involvement of the cGAS/STING system in MM treatment response, we assessed potential synergistic effects of both inhibitors with commonly used MM drugs. Treatment with small molecules enhanced drug sensitivity in all tested HMCLs to Melphalan and Bortezomib. Conclusion: Our data highlight the molecular mechanism of PSs disruption caused by NONO-SFPQ heterodimers inhibition, leading to the observed biological effects in HMCLs. These promising results emphasize the need to better characterize PSs role in MM and in treatment response, to determine whether PSs could represent a possible new druggable vulnerability in MM.