A New Therapeutic Hypothesis: Nonsense-Mediated Decay Is an Exploitable Target in Multiple Myeloma

Background

Nonsense-mediated decay (NMD) is a cellular quality control system that degrades mRNAs containing premature termination codons (PTCs) as well as ~10% of normal mRNAs (Kurosaki and Maquat, 2016). NMD thus prevents translation of misfolded proteins, and potential activation of the unfolded protein response (UPR). Mutations in splicing factors such as SF3B1, SRSF2, U2AF1 and ZRSR2 found in hematological as well as solid tumors, can lead to generation of aberrant mRNAs that contain PTCs. Aberrant splicing patterns in cancer cells can possibly result in increased pressure on the NMD machinery. CC-115, a potent inhibitor of mTOR kinase (TORK) and of DNA-dependent protein kinase, (DNA-PK; Mortensen et al., 2015; Tsuji et al., 2017), is in clinical development for the treatment of solid and hematologic malignancies (Thijssen et al., 2016). Preclinical data revealed an additional target of CC-115 and its differential effect on NMD. Our hypothesis was that a subset of tumor cells, especially hematologic tumors with high protein production and/or splicing factor mutations, would be susceptible to NMD inhibition by CC-115.

Methods

In total, 141 cell lines were screened for sensitivity to CC-115-mediated inhibition of proliferation and induction of cell death, in comparison to specific inhibition of TORK (CC-223). Isogenic DNA-PK knockout cell lines HCT116/HCT116 DNA-PK-/- and M059K/M059J DNA-PK-/- were treated with CC-115 and CC-223. Activity on NMD in vivo was tested using HCT-116 xenograft tumors treated with Vehicle or CC-115. Dependence on CC-115 sensitivity was determined using CRISPR/Cas9 technology of apoptosis or UPR genes in various MM cell lines. RNA sequencing was used for identification of potential targets in sensitive and resistant cell lines.

Results

A subset of cancer cell lines underwent cell death at sub-micromolar concentrations of CC-115 due to inhibition of NMD, but this was independent of mutations in splicing factors such as SF3B1. We next focused on MM cells as these generally produce high levels of (immunoglobulin) proteins and are prone to ER stress, and therefore potentially susceptible to NMD inhibition. Indeed, treatment with CC-115 resulted in activation of the UPR independent of TORK and DNA-PK inhibition, and cell death in 11/12 MM cell lines. Activity of CC-115 correlated strongly with cell death by the known ER-stress inducer, thapsigargin. Cell death by CC-115 occurred by the mitochondrial pathway of apoptosis, as it depended on caspase activity and the presence of Bax-Bak. Analysis of RNA sequencing data is ongoing and has indicated potential targets dictating sensitivity to CC-115-mediated cell death.

Conclusions

We describe that hematologic tumors with high protein production are specifically sensitive to CC-115, a novel and clinically exploitable inhibitor of NMD. This might lead to application in malignancies that depend on NMD to avoid excessive protein stress, such as multiple myeloma.