Apoptotic DNase Research Articles

Abstract 4295: Apoptotic DNase DFFB-induced DNA damage within cancer persister cells underlies acquired resistance to targeted therapies across multiple tumor types

Abstract Cancer acquired drug resistance contributes to a large proportion of patient deaths. A major obstacle to the development of effective therapies which prevent resistance is our lack of understanding of the relevant fundamental molecular adaptive processes. Here we report the discovery that tumor cells rely upon sublethal engagement of apoptotic death machinery and apoptotic DNase DFFB to acquire resistance to targeted therapies in multiple tumor types in cell culture and in vivo. We found that while DFFB wild type tumor cells readily acquire resistance to targeted therapies by escaping from the quiescent “persister” state into proliferating resistant cells, persister cells lacking functional DFFB remain indefinitely in a quiescent state on treatment. Mechanistically, we found that DFFB, which is activated by apoptotic caspases in surviving cancer persister cells, induces persistent DNA damage which promotes resistance through acquired mutations and nongenetic resistance through suppression of growth-arresting interferon signaling. During drug treatment acute interferon induction arises from sublethal mitochondrial outer membrane permeabilization and cytoplasmic exposure of mitochondrial nucleic acids which activate pattern recognition receptors. However, during longer treatments we found that DFFB knockout persister cells exhibit strikingly elevated levels of STAT1 and STAT2 and interferon stimulated gene sets. This interferon signaling is growth arresting because treatment of DFFB knockout persister cells with a JAK1/2 inhibitor rescues their ability to regrow. Furthermore, in DFFB wild type persister cells we found that DFFB-dependent DNA damage activates stress response transcription factor ATF3 which blocks the transcriptional activation of interferon stimulated genes following initial interferon induction. This is consistent with prior data in other contexts demonstrating that ATF3 is induced by DNA damage and that ATF3 directly binds to promoters of multiple interferon genes repressing their expression. These findings reveal that DFFB activation in persister cells prevents chronic interferon signaling and enables escape into proliferating drug resistant tumor cells. Therefore, DFFB-mediated interferon suppression represents a novel mechanism that is fundamental to acquired drug resistance. In addition, high expression of DFFB and ATF3 correlates with worse overall survival in a pan-cancer analysis and DFFB knockout mice develop normally indicating DFFB is a nonessential gene in normal tissue. Therefore, we propose that DFFB is a promising therapeutic target to prevent acquired resistance to targeted therapy. Citation Format: August Finley Williams, David A. Gervasio, Claire E. Turkal, Anna E. Stuhlfire, Michael X. Wang, Brandon E. Mauch, Ariel H. Nguyen, Michelle H. Paw, Mehrshad Hairani, Cooper P. Lathrop, Sophie H. Harris, Jennifer L. Page, Matthew J. Hangauer. Apoptotic DNase DFFB-induced DNA damage within cancer persister cells underlies acquired resistance to targeted therapies across multiple tumor types [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4295.

Abstract 6321: Developing a lenalidomide-inducible safety switch for CAR T cell therapy

Abstract While CAR T cell therapy is a medical breakthrough against various hematological malignancies, yet treatment failures often occur at least in part due to excessive engineered T cell activity leading to severe toxicities. To combat this, strategies pairing CAR T cells with safety systems, for example, suicide switches engineered to induce cell death are being actively pursued. However, existing designs have limited performance characteristics and generally use unapproved small molecule controllers or non-human sequences. To overcome these issues, we have developed a novel suicide switch leveraging the well-tolerated, FDA-approved anti-cancer drug lenalidomide, as a controller of the stability of fusion proteins tagged with a “degron” sequence. Specifically, we engineered a stoichiometric pair of transgenes using the apoptotic DNase pair, caspase activating DNase (CAD) and its degron tagged inhibitor (ICAD-degron). We hypothesized that the overexpressed ICAD-degron could restrain CAD, until lenalidomide treatment degrades ICAD to unleash CAD for apoptosis. We found that our system was stably overexpressed and rapidly induced cell death even at sub-nanomolar concentrations of lenalidomide. A comparison of our design with the frontline safety switch, inducible Caspase 9 in primary human T cells revealed that the lenalidomide-induced suicide switch enabled more complete depletion of engineered cells. We tested the effects of our suicide switch on CAR T cell function and found comparable proliferation and tumor cell cytolysis with fast depletion upon lenalidomide addition. In vivo, lenalidomide suicide switch CAR T cells exhibited similar anti-tumor function as control CAR T cells and were rapidly depleted after drug treatment. In summary, utilizing all human sequences and a clinically approved drug controller, we developed a suicide switch that is well tolerated and can rapidly induce cell death even at sub-therapeutic concentrations of lenalidomide. More broadly, chemogenetic regulation of stoichiometric protein pairs is a generalizable strategy for post-translational control of highly active transgenic elements. Citation Format: Ditsa Sarkar, William Lin, Joanna Y. Kim, Nelson H. Knudsen, Isabel C. Lane, Tamina Kienka, Michael C. Kann, Amanda A. Bouffard, Andy Cheng, Marcela V. Maus, Robert T. Manguso, Max Jan. Developing a lenalidomide-inducible safety switch for CAR T cell therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6321.

Abstract 99: Apoptotic DNase DFFB mediates cancer persister cell mutagenesis and acquired drug resistance

Abstract Tumors frequently acquire resistance to therapy resulting in disease relapse and patient mortality. Therapeutic approaches to overcome acquired resistance are hindered by limited treatment options and the presence of multiple parallel resistance mechanisms. Rather than treating resistance after it emerges, it may possible to prevent it by inhibiting the adaptive processes which initiate resistance. However, these processes and the genes which control them are poorly understood. Here we report that drug-tolerant cancer persister cells, which constitute residual disease and seed relapse, undergo drug stress-induced sublethal apoptotic signaling resulting in activation of apoptotic DNase DFFB without cell death. DFFB induces DNA damage and mutagenesis in persister cells, revealing DFFB as a key regulator of stress-induced mutagenesis in persister cells. Furthermore, we found in multiple tumor types that DFFB is required for acquired resistance to targeted therapies. Mechanistically, DFFB induces multiple putative resistance mutations and may also promote epigenetic changes which result in resistance. These results reveal DFFB as a promising potential therapeutic target to prevent acquired resistance. In addition, this stress-sensing mutagenic mechanism may promote mutagenesis in other contexts including normal tissues because sublethal apoptotic signaling may result from a variety of physiological stresses. Citation Format: August Finley Williams, David Gervasio, Claire Turkal, Anna Stuhlfire, Michael Wang, Brandon Mauch, Ariel West, Michelle Paw, Mehrshad Hairani, Cooper Lathrop, Sophie Harris, Jennifer Page, Matthew Hangauer. Apoptotic DNase DFFB mediates cancer persister cell mutagenesis and acquired drug resistance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 99.

Abstract PR003: Apoptotic DNase DFFB mediates stress-induced mutagenesis

Abstract It was recently discovered that actively proliferating and post-mitotic normal human tissues acquire somatic mutations at the same rate. This indicates that somatic mutagenesis is primarily independent of cell proliferation and therefore cannot be based on DNA replication errors. Furthermore, the mutagenic mechanisms which contribute to ubiquitous mutational signatures such as SBS5 remain unknown. It has been hypothesized that tissues accumulate mutations through unknown physiologic processes which contribute to DNA damage but the molecular details remain unknown. These observations have revealed our lack of understanding of the fundamental processes of mutagenesis. Stress-induced mutagenesis has been previously characterized in bacteria which respond to stress by transiently increasing the rate of mutagenesis to promote the emergence of resistance mutations. This phenomenon was recently discovered in human cancer cells but the mechanism is unknown. Here, we present our discovery that cancer cells under drug stress undergo sublethal apoptotic signaling resulting in activation of apoptotic DNase DFFB. DFFB, which canonically acts to fragment chromosomal DNA during apoptosis, induces DNA strand breaks in surviving cancer cells resulting in error prone repair and mutagenesis. In the context of targeted therapy-treated cancer, we found that DFFB is required for acquired resistance to emerge in multiple tumor types. We also found that CD8 T cells induce a similar process in targeted tumor cells which survive. Therefore, sublethal apoptotic signaling may be a general phenomenon which promotes mutagenesis in human cancer. Based on these observations and prior literature showing sublethal apoptotic signaling occurs in normal cells in response to various stress stimuli, we propose that this mechanism may be a previously unknown fundamental mutagenic process which contributes not only to acquired resistance in cancer but also to tumor initiation, aging and other conditions. Citation Format: August F. Williams, David A.G. Gervasio, Claire E. Turkal, Mehrshad Hairani, Michael X. Wang, Anna E. Stuhlfire, Ariel H. West, Brandon E. Mauch, Michelle H. Paw, Cooper P. Lathrop, Sophie H. Harris, Jennifer L. Page, Matthew J. Hangauer. Apoptotic DNase DFFB mediates stress-induced mutagenesis [abstract]. In: Proceedings of the AACR Special Conference: Aging and Cancer; 2022 Nov 17-20; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2022;83(2 Suppl_1):Abstract nr PR003.

Apoptotic DNase network: Mutual induction and cooperation among apoptotic endonucleases.

DNA fragmentation produced by apoptotic DNases (endonucleases) leads to irreversible cell death. Although apoptotic DNases are simultaneously induced following toxic/oxidative cell injury and/or failed DNA repair, the study of DNases in apoptosis has generally been reductionist in approach, focusing on individual DNases rather than their possible cooperativity. Coordinated induction of DNases would require a mechanism of communication; however, mutual DNase induction or activation of DNases by enzymatic or non‐enzymatic mechanisms is not currently recognized. The evidence presented in this review suggests apoptotic DNases operate in a network in which members induce each other through the DNA breaks they produce. With DNA breaks being a common communicator among DNases, it would be logical to propose that DNA breaks from other sources such as oxidative DNA damage or actions of DNA repair endonucleases and DNA topoisomerases may also serve as triggers for a cooperative DNase feedback loop leading to elevated DNA fragmentation and subsequent cell death. Therefore, mutual induction of apoptotic DNases has serious implications for studies focused on activation or inhibition of specific DNases as a strategy for therapeutic intervention aimed at modulation of cell death.

DR396, an apoptotic DNase γ inhibitor, attenuates high mobility group box 1 release from apoptotic cells

High mobility group box1 (HMGB1) is a non-histone chromatin chromosomal protein playing an important role in chromatin architecture and transcriptional regulation. Recently, HMGB1 has been shown to be secreted into extracellular milieu in necrosis and apoptosis, and involved in inflammatory responses. However, the mechanism by which apoptotic cells release HMGB1 is unclear. In this study, to investigate the mechanism of HMGB1 release, we searched inhibitors of HMGB1 release from apoptotic cells. As a result, three compounds, 4-(4,6-dichloro-[1,3,5]-triazin-2-ylamino)-2-(6-hydroxy-3-oxo-3H-xanthen-9-yl)-benzoic acid (DR396), Pontacyl Violet 6R (PV6R), and Fmoc-D-Cha-OH (FDCO) in our in-house chemical library were found to inhibit HMGB1 release from staurosporine (STS)-induced apoptotic HeLa S3 cells. Interestingly, these three compounds have been previously categorized into apoptotic DNase γ inhibitors. Therefore, we examined whether apoptotic nucleosomal DNA fragmentation is involved in the release of HMGB1 during apoptosis. Expectedly, DR396, which is the most potent and specific inhibitor of DNase γ, was found to almost completely inhibit both HMGB1 release and internucleosomal DNA cleavage in HeLa S3 cells transfected with DNase γ expression vector and stably expressing DNase γ (HeLa S3/γ cells). These results clearly suggest that nucleosomal DNA fragmentation catalyzed by DNase γ is critical in the release of HMGB1 from apoptotic cells.

Activation of cell death pathways in the inner ear of the aging CBA/J mouse

We have previously demonstrated that oxidative stress increases in the inner ear of aging CBA/J mice and might contribute to the loss of function of the sensory system. We now investigate the activation of cell death pathways in the cochleae of these animals. Middle-aged (12 months) and old (18–26 months) mice with hearing deficits displayed outer hair cell nuclei with apoptotic and, to a lesser extent, necrotic features. Both intrinsic and extrinsic cell death pathways were activated by translocation or post-translational modification of proteins in the aging cochlea as compared to young (3 months) animals. Cytosolic cytochrome c increased, formed a complex with, and activated caspase 9. Endonuclease G translocated to the nuclei of aging outer hair cells suggesting its function as an apoptotic DNase. The cleaved (and hence active) forms of calpain I and calpain II increased while active cathepsin D was transiently elevated in middle-aged but not old animals. Finally, increases in the phosphorylation of p38 MAPK and JNK implicated the additional involvement of the MAPK pathway. The results suggest that multiple cell death pathways, all potentially linked to oxidative stress, are activated in hair cells of the auditory organ in aging mice.

Regulation of the human apoptotic DNase/RNase Endonuclease G: involvement of Hsp70 and ATP

Endonuclease G (EndoG) is a mitochondrial enzyme that becomes an apoptotic nuclease when released from the mitochondrial intermembrane space. EndoG will digest either DNA or RNA, but at physiological ionic strength, RNA is a much more favorable substrate as compared to chromatin. This indicates that EndoG's major in vivo function(s) may be: (i) an apoptotic RNase, and/or (ii) an apoptotic DNase in the presence of additional co-activators. In the present study we have searched for factors that modulate the activity of human EndoG on DNA substrates. We demonstrate that EndoG forms complexes with AIF and FEN-1 but not with PCNA. Interestingly, heat shock proteins 70 interact with EndoG and are involved in the regulation of its activity. Purified Hsp70 prevented stimulation of EndoG DNase activity by other nuclear factors in the ATP-dependent manner.

Identification of two functional nuclear localization signals in DNase gamma and their roles in its apoptotic DNase activity.

Among DNase I family members, only DNase gamma causes DNA fragmentation during apoptosis. However, the molecular basis for this functional feature of DNase gamma is poorly understood. Here we describe the identification of functional NLSs (nuclear localization signals) in DNase gamma and their roles in its apoptotic function. DNase gamma contains two NLSs: a classical bipartite-type NLS (NLS1) located in the N-terminal half, and a short basic domain (NLS2) at the C-terminus. No potential NLSs are found in the primary structures of other DNase I family DNases. Inactivation of either NLS1 or NLS2 causes reduced DNA ladder-producing activity in DNase gamma. Disruption of NLS2 suppresses ladder formation more effectively than disruption of NLS1. DNase gamma doubly mutated in both NLSs is enzymically active, but no longer catalyses apoptotic DNA fragmentation. Although DNase I fails to produce ladder formation during apoptosis, DNase I fused to NLS2 of DNase gamma through its C-terminus is able to catalyse DNA fragmentation in apoptotic cells. These results indicate that the presence of either NLS1 or NLS2 is necessary for the apoptotic function of DNase gamma, and that the most important domain for this function is NLS2. These findings also explain the lack of apoptotic DNase activity in the other DNase I family DNases.

A Role for EndoG in Apoptosis

During apoptosis, DNA is cleaved by enzymes, which assures that the death event is inexorable. Two proteins with known nuclease activity have been implicated in apoptosis: the Caenorhabditis elegans DNAse II-like protein termed NUC-1, and the mammalian DNAse termed CAD (caspase-activated DNAse). Now another nuclease can be added to the list: endonuclease G (EndoG). Li et al. found that isolated mitchondria treated with apoptosis-inducing Bid released EndoG. EndoG activation apparently did not require the proteolytic digestion of a bound inhibitory protein; rather, release from the mitochondria was sufficient for its activation. Additionally, in CAD-deficient cells, EndoG was able to carry out caspase-independent digestion of DNA, indicating that EndoG activity alone may be sufficient for DNA fragmentation. Parrish et al. showed that CPS-6, a C. elegans protein that bears considerable sequence similarity to EndoG, also participates in cell death in the worm. cps-6 mutant worms had decreased cell death during their development. These results were also seen in experiments using RNA interference to reduce the amount of CPS-6 expressed in worms. Hengartner writes a synopsis of the papers and discusses the implications of these articles on apoptosis research. L. Y. Li, X. Luo, X. Wang, Endonuclease G is an apoptotic DNase when released from mitochondria. Nature 412 , 90-94 (2001). [Online Journal] J. Parrish, L. Li, K. Klotz, D. Ledwich, X. Wang, D. Xue, Mitochondrial endonuclease G is important for apoptosis in C. elegans . Nature 412 , 95-99 (2001). [Online Journal] M. O. Hengartner, DNA destroyers. Nature 412 , 27-29 (2001). [Online Journal]

A Role for EndoG in Apoptosis

During apoptosis, DNA is cleaved by enzymes, which assures that the death event is inexorable. Two proteins with known nuclease activity have been implicated in apoptosis: the Caenorhabditis elegans DNAse II-like protein termed NUC-1, and the mammalian DNAse termed CAD (caspase-activated DNAse). Now another nuclease can be added to the list: endonuclease G (EndoG). Li et al. found that isolated mitchondria treated with apoptosis-inducing Bid released EndoG. EndoG activation apparently did not require the proteolytic digestion of a bound inhibitory protein; rather, release from the mitochondria was sufficient for its activation. Additionally, in CAD-deficient cells, EndoG was able to carry out caspase-independent digestion of DNA, indicating that EndoG activity alone may be sufficient for DNA fragmentation. Parrish et al. showed that CPS-6, a C. elegans protein that bears considerable sequence similarity to EndoG, also participates in cell death in the worm. cps-6 mutant worms had decreased cell death during their development. These results were also seen in experiments using RNA interference to reduce the amount of CPS-6 expressed in worms. Hengartner writes a synopsis of the papers and discusses the implications of these articles on apoptosis research.L. Y. Li, X. Luo, X. Wang, Endonuclease G is an apoptotic DNase when released from mitochondria. Nature 412, 90-94 (2001). [Online Journal]J. Parrish, L. Li, K. Klotz, D. Ledwich, X. Wang, D. Xue, Mitochondrial endonuclease G is important for apoptosis in C. elegans. Nature 412, 95-99 (2001). [Online Journal]M. O. Hengartner, DNA destroyers. Nature 412, 27-29 (2001). [Online Journal]

Endonuclease G is an apoptotic DNase when released from mitochondria.

Nucleosomal fragmentation of DNA is a hallmark of apoptosis (programmed cell death), and results from the activation of nucleases in cells undergoing apoptosis. One such nuclease, DNA fragmentation factor (DFF, a caspase-activated deoxyribonuclease (CAD) and its inhibitor (ICAD)), is capable of inducing DNA fragmentation and chromatin condensation after cleavage by caspase-3 (refs 2,3,4). However, although transgenic mice lacking DFF45 or its caspase cleavage site have significantly reduced DNA fragmentation, these mice still show residual DNA fragmentation and are phenotypically normal. Here we report the identification and characterization of another nuclease that is specifically activated by apoptotic stimuli and is able to induce nucleosomal fragmentation of DNA in fibroblast cells from embryonic mice lacking DFF. This nuclease is endonuclease G (endoG), a mitochondrion-specific nuclease that translocates to the nucleus during apoptosis. Once released from mitochondria, endoG cleaves chromatin DNA into nucleosomal fragments independently of caspases. Therefore, endoG represents a caspase-independent apoptotic pathway initiated from the mitochondria.

Inhibition of Apoptosis-Associated DNA Fragmentation Activity in Nonapoptotic Cells: The Role of DNA Fragmentation Factor-45 (DFF45/ICAD)

We have investigated the mechanism whereby nuclear DNA fragmentation activity emerging during early apoptosis is inhibited during normal cell life. In a cell-free system, cytosol fractions from diverse nonapoptotic human cell lines (Jurkat T-cell leukemia, HeLa carcinoma, SK-N-MC neuroblastoma, and WI-38 embryonic lung fibroblast) potently neutralized the nuclear DNA fragmentation activity of cytosol from apoptotic anti-Fas treated Jurkat cells. Recombinant human DNA fragmentation factor 45 kDa subunit (DFF45/ICAD), an inhibitor of the caspase-activated DNase DFF40/CAD, substituted for healthy cytosol in inhibiting DNA fragmentation. An antiserum against human DFF45 detected 44 and 34 kDa proteins (major and minor, respectively) in the cytosols but not in the nuclear or membrane fractions of various cultured human cells. Cytosols depleted of DFF45/ICAD by immunoadsorption had little or no inhibitor of nuclear DNA fragmentation activity and no caspase-activated DNA fragmentation activity. We conclude that immunoreactive DFF45/ICAD is the principal inhibitor of apoptotic DNase activity in the cytosol of healthy cells.

Purification and characterization of a DNase gamma-like endonuclease from Xenopus laevis liver.

We recently found that two apoptotic DNase gamma-like endonucleases (36 and 38 kDa DNases) were present in Xenopus laevis larval and adult liver cell nuclei and that their activities increased in metamorphic climax. Here, we purified the main DNase gamma-like endonuclease from Xenopus laevis liver cell nuclei and characterized its physical and enzymatic properties in detail. The molecular mass of Xenopus liver nuclear endonuclease was 38,000 daltons as determined by SDS-polyacrylamide gel electrophoresis. A native molecular mass of 35,000 was estimated by gel filtration. The purified Xenopus liver endonuclease was a neutral one and required both Ca2+ and Mg2+ for DNase activity. Unlike the mammalian DNase gamma, the Ca2+/Mg2+ requirement could not be supplied by Mn2+. The inhibition profiles by aurintricarboxylic acid, sodium citrate and divalent metal ions such as Co2+, Ni2+, Cu2+ and Zn2+ were similar to those of mammalian DNase gamma. These results suggest that this endonuclease is a Xenopus laevis homolog of the mammalian apoptotic endonuclease DNase gamma.