Acquired resistance to cancer therapy is a formidable obstacle preventing cancer

treatments from fully curing patients. Across multiple cancer types, analysis of tumor shrinkage in response to treatment reveals a residual, quiescent, and drug tolerant cancer cell population termed cancer persister cells. Persister cells initially survive drug treatment through reversible, non-genetic mechanisms but subsequently acquire resistance-conferring mutations and regrow to seed the emergence of drug resistant tumors. The mechanisms by which persister cells acquire mutations are unknown. Here, we investigate our hypothesis that persister cells experience drug stress-induced sublethal apoptotic signaling which results in activation of apoptotic DNases. These DNases, which normally serve to fragment chromosomal DNA during apoptosis, instead promote DNA damage and mutagenesis within surviving persister cells allowing for regrowth into drug-tolerant expanded persister cell (DTEP) colonies. We observed that during extended treatment with targeted therapies, persister cells require apoptotic DNase (DFFB) to regrow into drug-resistant proliferating cells. Consistent with the hypothesis that persister cells are mutationally active, we also observed that persister cells exhibit elevated levels of DNA damage. This DNA damage was absent in DFFB KO persister cells, suggesting that DFFB activation is required to induce DNA damage in persister cells. These findings reveal that sublethal apoptotic signaling and activation of DFFB may play a role in persister cell mutagenesis and tumor relapse.

The development of oncogene targeted therapies over the past few decades facilitated by cancer driver gene identification through next generation sequencing has resulted in impressive clinical responses against late-stage cancers. However, robust responses are hindered by acquired drug resistance. This is the scenario where a tumor initially responds to therapy, but then regrows even in the presence of continued drug treatment. Here, we investigate cancer persister cells as a model of acquired drug resistance to identify processes that allow tumors to overcome drug pressure. Single cell transcriptomic profiling revealed well-characterized drug resistance processes occurring in persister cells such as epithelial-mesenchymal transition. Interestingly, it also revealed apoptotic signaling as a process shared across multiple persister cell models. We found that cancer persister cells survive caspase 3 activation and minority mitochondrial outermembrane permeabilization (MOMP). This sublethal apoptotic signaling activates the nuclease DFFB. DFFB causes mutagenic DNA damage resulting in persister regrowth through both genetic and nongenetic mechanisms. Inhibition of DFFB blocked persister cell regrowth both in cell culture and in vivo. Given the need for targets against stress adaptation mechanisms, DFFB represents a promising therapeutic target that could extend the efficacy of already available anti-cancer therapies.

Following initial therapy responses, tumors often relapse leading to patient mortality. How cancer cells change from a therapy sensitive to a therapy resistant state is poorly understood, particularly in the context of cytotoxic CD8 T cells mediated immunotherapy. Multiple studies have identified mechanisms by which residual tumor cells avoid CD8 T cell activation, but thus far there have been no studies focused on whether residual tumor cells survive continuous exposure to and attack by activated CD8 T cells. We hypothesized that in addition to commonly proposed evasive mechanisms, cancer cells can enter an immunotherapy-tolerant persister cell state to survive activated CD8 T cell attack. Here, we report the observation of a subpopulation of quiescent immunotherapy persister cells which survive through a reversible, non-genetic mechanism. Upon extended cytotoxic T cell pressure, a subset of immunotherapy persister cells reenter the cell cycle and regrow into overtly resistant colonies which may represent the initial events of acquired resistance and tumor recurrence. These findings suggest that cancer cells may survive initial T cell cytotoxicity exposure through a quiescent persister state for several weeks prior to relapse. Interestingly, we found that immunotherapy persister cells survive despite continual T cell activation and experience sublethal activation of apoptotic signaling. Together, these studies reveal a novel population of tumor cells which survive extended CD8 T cell attack and may seed tumor recurrence during acquired resistance to immunotherapy.

The landscape of cancer treatment options has grown to include more precise and less invasive strategies. However, acquired drug resistance often results in transient responses. Cancer persister cells are a heterogeneous subpopulation of tumor cells which survive lethal doses of anti-cancer therapy and seed tumor recurrence. Persister cells survive treatment by non-genetic mechanisms that are not well understood. Additionally, persister cells undergo stress induced mutagenesis and rapid genetic adaptation resulting in the emergence of genetically drug resistant cells. Therapeutically targeting persister cells is therefore a promising approach to prevent acquired resistance. In this thesis, we describe our investigations into persister cell vulnerabilities and mutagenesis. First, we discovered that HDAC inhibitors sensitize various types of persister cells to ferroptosis, an oxidative cell death process. Second, we describe our efforts to elucidate the mechanism by which the clinically utilized anti-alcoholism drug, Disulfiram kills persister cells. Lastly, we explore the mechanism by which persister cells undergo mutagenesis that allows for acquired genetic resistance to cancer drugs.

One major problem plaguing the medical community is patient relapse of cancer following targeted drug therapy. In a majority of patients with non-small-cell lung carcinoma, this process was shown to occur in as little as one year following treatment. Here we investigate the role that molecules, such as antioxidants, and cellular processes, such as DNA death, damage, and response, have in the underlying mechanistic basis for acquiring drug resistance. Our findings suggest that antioxidants are not capable of adequately preventing the acquisition of drug resistance, pointing toward R.O.S.-independent mechanisms of acquisition of drug resistance mutations. Additionally, we show that while the inhibition of DNA damage response and repair pathways significantly prevent the outgrowth of cancerous cells in the presence of drug, there is no difference in response from drug naïve versus drug-tolerant cancer cells. From these findings, we conclude that despite their reported disabled DNA repair machinery, drug-tolerant persister cells are not sensitized to death via inhibition of DNA damage response genes. Additionally, we elucidate the mechanism through which Disulfiram, a drug clinically approved for alcoholism which was recently reported to kill persister cells, induces persister cell-specific lethality. We find that Disulfiram does not kill persister cells through ALDH inhibition, as previously reported, but rather through an oxidative-apoptotic mechanism. By furthering the understanding of factors involved in the tumor’s acquisition of drug resistance, we provide insight into potential mechanisms to target through the development of new treatments aimed at preventing the occurrence of cancer relapses.

Background

Methods

Results

Conclusions