Activation of the Ataxia Telangiectasia Mutated/Autophagy pathway by a G-quadruplex ligand links senescence with apoptosis

ABSTRACT

G-quadruplex (G4) ligands have shown anti-tumor activity by stabilizing G4 structures. We recently reported that the G4 ligand 20A promotes senescence in cancer cells through Ataxia Telangiectasia Mutated (ATM) and autophagy activation. Inhibition of both pathways directs cells to apoptosis, positioning ATM/autophagy axis as a linchpin between senescence and apoptosis.

Autophagy, a catabolic lysosomal pathway, has gained increasing interest in cancer research as this process can influence cancer development and cancer therapy.^1-3 However, little is known about the mechanism of cell fate regulation by autophagy in response to drugs that bind to DNA, in particular to Guanine-quadruplexes (G4).⁴ G4 are a family of peculiar three-dimensional structural arrangements adopted by guanine-rich DNA and RNA strands. This structure is characterized by the stacking of several contiguous G-quartets formed by the self-association of four guanine residues in a planar arrangement. Research into these unusual conformations has sustained a strong momentum because G4 are stable under physiological conditions. G4-prone sequences are distributed throughout the human genome at repetitive regions, including telomeres, as well as a number of promoters, especially in key oncogenes.⁵

G4 offer attractive structural features for targeting by small molecules. A variety of compounds can selectively target G-quadruplexes with little or no binding to other DNA or RNA structures. Some of these ligands show selective cytotoxicity toward cancer cells in vitro, and several G4 ligands (G4L) have been successfully used in cancer xenograft models alone or in combination with other cancer drugs.⁵ Our team developed a series of novel 2,4,6-triarylpyridines G4L with fair to excellent selectivity towards G-quadruplexes.⁶

In a recent study,⁷ using a wide set of analysis methods including transcriptomic and proteomic analyses, cell imaging, and biochemical studies, we showed that one of these 2,4,6-triarylpyridines called 20A elicits a significant activation of biological pathways related to DNA Damage Response (DDR), autophagy, and cell growth arrest. These observations prompted us to investigate the links between these three processes in response to 20A.

Our results revealed that 20A causes growth arrest in in vitro and in vivo cancer models. Notably, 20A promotes induction of senescence and apoptotic cell death through Tumor Protein p53 (TP53)-independent mechanisms. We found that 20A treatment leads to induction of DNA damage with subsequent activation of two DDR pathways, Ataxia Telangiectasia Mutated (ATM)/Checkpoint 2 (CHK2) and Ataxia Telangiectasia, and Rad3-related (ATR)/Checkpoint 1 (CHK1) (Figure 1). The activation of ATM by 20A is responsible for the induction of autophagy and leads ultimately to senescence. In turn, autophagy controls DDR by regulating the activation of CHK1. Such regulation between DDR and autophagy may constitute an amplification loop for maintaining DDR in response to G4L and promoting cell survival. Interestingly, we found that disruption of ATM and autophagy impairs the senescent onset and drives cells to apoptotic cell death. Although senescence was first considered as a tumor suppressive mechanism, recently growing body of evidence suggests that this process can also be a tumor promotor mechanism.⁸

Figure 1.

Links between the DNA damage response, autophagy, and senescence in response to G-quadruplex ligands. The G-quadruplex ligand 20A causes global DNA Damage that results in the activation of the DNA Damage Response pathways (DDR). (a) In particular, the activation of Ataxia Telangiectasia Mutated (ATM) leads to the induction of autophagy which, in turn, controls DDR by regulating Checkpoint kinase 1 (CHK1). The ATM/autophagy axis is responsible for the induction of senescence and the protection of cells against death. (b) Conversely, disruption of either ATM or autophagy following 20A treatment impairs the induction of senescence and drives cells to apoptotic cell death highlighting the potential therapeutic interest of targeting this pathway to achieve the optimal efficacy of the G-quadruplex ligand 20A.

The ability of G4L to induce senescence was initially attributed to stabilization of telomeric G4 structures, resulting in telomerase inhibition and telomere erosion. However, recent findings revealed that some G4L can induce senescence through telomere-independent mechanisms. We found no increase in telomeric damage in cell lines subjected to 20A; we rather observed an increase in the number of metaphase chromosomes covered with gamma H2A histone Family X (γ-H2AFX) (which we refer to as global DDR), as expected for a compound targeting DNA. These results support the notion that, while 20A does target DNA, telomeres, and telomerase are not the main sites of its action. Another persistent idea found in the literature regarding the role of G4L as modulators of transcription is that G4L act only as transcriptional inhibitors by competing with transcription factors for binding to gene promoters. The mechanism is probably not as simple since transcriptomic analyses of other compounds also revealed that a number of genes are actually upregulated in response to G4L, as found in our study for 20A. Whether or not such a regulation relies on the binding of 20A to particular G4 structures warrants further investigation.

Taken together, our study brings evidence that the activation of the ATM/autophagy pathway by the G4L 20A plays a pivotal role in regulating cell fate choice between senescence and apoptosis. From a therapeutic point of view, inhibiting the ATM/autophagy pathway impairs senescence, and drives cells to apoptotic cell death, which offers a promising strategy to optimize the potential therapeutic efficacy of the G4L 20A in cancers. In the same vein, we speculate the G4L 20A might be particularly effective in cancer cells that are deficient for ATM such as those developed in cancer patients carrying mutations in ATM.

Autophagy was also reported to regulate cell death in response to two G4L compounds unrelated to 20A^9,10 through a mechanism that is not fully understood. Further investigation will be required to decipher the precise mechanism through which autophagy and ATM (its upstream effector) regulate senescence and protect cells against apoptosis induced by 20A. Moreover, It will be interesting to examine if the stated results are also observed with other G4L. A better understanding of the molecular mechanisms of action of G4L will hopefully identify novel targetable pathways to improve the anticancer effects of these ligands and design efficient drugs combination for cancer treatment.

Funding Statement

This work was supported by funds from the Institut National de la Santé et de la Recherche Médicale (INSERM), the University of Bordeaux, Institut Curie, the Bergonié Institute, the Conseil Regional d’Aquitaine, the Ligue contre le Cancer, Comité de Dordogne and the SYMBIT project (reg. no. CZ.02.1.01/0.0/0.0/15_003/0000477) financed by the ERDF.

Abbreviations

	Ataxia Telangiectasia Mutated
ATR	Ataxia Telangiectasia and Rad3-related
CHK(1/2)	Checkpoint (1/2)
G4L	Guanine-quadruplex (G4) ligand
DDR	DNA Damage Response
TP53	Tumor protein p53 (best known as p53)

Acknowledgments

The authors would like to thank the members of the ARNA and ACTION laboratories for useful discussions.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.