ATP released from dying cancer cells stimulates P2X4 receptors and mTOR in their neighbors

Article summary

Resistance of tumors to anti-cancer drugs remains an ongoing challenge in cancer therapeutics. In a publication in Nature, Schmitt et al. showed that tumor cell death can alter neighboring cells via activating purinergic signaling toward their better survival. They used patient-derived tumor organoids (PDTOs) from human colorectal cancer (hCRC) cell lines and tumor organoids derived from Lgr5^eGFP-DTR+ mice as models of solid tumors. Following initial induction of tumor cell death, the surviving tumor cells of both had elevated phosphorylation of p70 S6 kinase, PRAS40, and ribosomal S6 protein, all indicative of the activation of the mechanistic (formally “mammalian”) target of rapamycin (mTOR) pathway [1]. The mTOR inhibitor, rapamycin, had little effect on tumor cells by itself. However, when applied in combination with a cell-death-inducing agent, rapamycin had compound effects on suppressing tumor cell survival in vitro and in vivo. These observations suggested that surviving tumor cells had acquired a new dependence on mTOR signaling. In the search for the underlying mechanism, the authors found that adenosine 5′-triphosphate (ATP), present in the supernatant of the organoid culture, could induce phosphorylation of ribosomal S6 in tumor organoids by itself. Screening at the mRNA level identified the P2X4 receptor as the most prominent purinergic receptor subtype expressed in hCRC and healthy colon organoids. Reducing the activity of P2X4 receptors by either administering 5-(3-bromophenyl)-1,3-dihydro-2H-benzofuro[3,2-e]-1,4-diazepin-2-one (5-BDBD, 1 μM) or doxycycline (DOX)-inducible short hairpin (sh)RNA knockdown, in combination with tumor-cell death induction, significantly reduced the survival of hCRC. The authors investigated in parallel why surviving tumor cells had become dependent on mTOR activation. When N-acetyl-l-cysteine (NAC) was used to scavenge reactive oxidative species (ROS) in hCRC tumors, the dependence of surviving tumor cells on mTOR was abolished. Collectively, this study suggests that ROS and ATP released from dying tumor cells can drive adjacent tumor cells to a new pathophysiological status, where the parallel stimulation of mTOR counteracts the proapoptotic effect of ROS by the ATP-activated P2X4 receptor. The end outcome is the better survival of neighboring tumor cells. The study implicates the therapeutic potential of P2X4 inhibitors as a part of combinatory anti-cancer therapy by enhancing the effect of standard chemotherapy agents for CRC and other types of tumors [1].

Commentary

The complex and dynamic cell-to-cell communication between cancer cells within the tumor microenvironment is thought to contribute to drug resistance in tumors, reducing the effects of anti-cancer treatment. Schmitt et al. have demonstrated a novel paracrine mechanism by which dying tumor cells influence the survival of neighboring tumor cells, mediated by P2X4 receptors and mTOR. mTOR is a serine/threonine protein kinase of the PI3K-related kinase family, known to form two different protein complexes, mTOR complex 1 (mTORC1) and 2 (mTORC2) [2]. mTORC1 contains partner proteins regulatory-associated protein of mTOR (raptor) and mammalian lethal with SEC13 protein 8 (mLST8), and two inhibitory proteins proline-rich Akt substrate of 40 kDa (PRAS40) and DEP domain containing mTOR interacting protein (DEPTOR). mTORC1 can promote protein synthesis through phosphorylation of p70S6 kinase 1 (S6K1) and eIF4E binding protein (4EBP) [2, 3]. Phosphorylation of ribosomal protein S6 is a known indicator of increased protein synthesis, and mTORC1 is one of the signaling pathways known to induce S6 phosphorylation [4]. Physiologically, mTOR is activated by several growth factors and mitogen-dependent signaling pathways. mTORC1 hyperactivation has been associated with tumorigenesis of many types of cancer, leading to the development of mTOR inhibitors as anti-cancer reagents, such as rapamycin and its derivatives [2, 3].

In this study, Schmitt et al. used PDTOs generated from hCRC cell lines, and cell death was induced by 5-fluorouracil (5-FU), a chemotherapy agent used to treat CRC since the 1990s. The authors also used mice tumor organoids of intestinal origin, azoxymethane, and dextran sodium sulfate (AOM-DSS) organoids, and AMO-DSS (ATKN) organoids generated from Lgr5^eGFP-DTR+ transgenic mice. These mice tumor organoids allowed inducible ablation of a subpopulation of intestine epithelial cells expressing Lgr5 [5] by diphtheria toxin (DT) to eliminate the possibility that 5-FU had cell-autonomous effects on surviving tumor cells. In both models, initial tumor cell death induced increased phosphorylation of ribosomal protein S6, p70 S6 kinase, and PRAS40, indicative of mTOR activation. Indeed, after 4 h of 5-FU treatment, surviving hCRC tumor cells had acquired sensitivities to mTOR inhibition. Rapamycin significantly suppressed tumor cell survival in vitro, their reseeding capacity in vitro, and the growth of tumor volume in the subcutaneous tumor xenograft model. mTORC1, but not mTORC2, seemed to be involved, as the tumor cell survival was significantly reduced by gene knockdown of raptor, but not DEPTOR [1]. Investigating the underlying mechanism, the authors showed that the supernatant obtained from DT-treated Lgr5^eGFP-DTR+ organoids triggered mTOR activation when applied to untreated organoids, and the supernatant contained ATP [1]. Furthermore, the addition of ATP by itself induced phosphorylation of ribosomal S6 in Lgr5^eGFP-DTR− organoids at 0.1–100 μM and in hCRC at 100 μM [1]. DT-induced phosphorylation of ribosomal S6 in Lgr5^eGFP-DTR+ organoids could be reduced by hydrolysis of ATP by apyrase [1]. Using a combination of mRNA expression profiling, a pharmacological approach, and sh-mediated gene knockdown, the authors narrowed down the purinergic receptor that regulated S6 phosphorylation to the P2X4 subtype. These data demonstrate that ATP released from dying tumor cells and the subsequent activation of P2X4 were the underlying mechanism for mTOR activation in the surviving tumor cells. As both P2X4 receptor inhibition and mTOR inhibition had little effect on tumor cells by themselves, the authors concluded that the dependence on the P2X4 receptor-mediated activation of mTOR is a trait that was acquired following the death of neighboring tumor cells [1].

The roles of purinergic receptor signaling in tissue homeostasis and cell survival have been known for a long time. The link between the mTOR and the purinergic signaling pathways has started to emerge only recently, mainly around the P2X7 subtype. Bian et al. used C57BL/6 murine MCA38 colon cancer cells to show that ATP (1–5 mM) and BzATP (0.1–2 mM) significantly reduce the viability and growth of these colon cancer cells. They also reported that gene knockdown of P2X7 receptors reduced the effects of ATP on cancer cells. ATP also reduced phosphorylation of mTOR-pathway proteins (PRAS40, ribosomal S6), suggesting that ATP-mediated activation of P2X7 receptors inhibited mTOR signaling and reduced cancer cell survival [6]. This is the opposite direction of mTOR regulation by ATP to the observation by Schmitt et al. It should be noted that the P2X7 receptor was not the only purinergic subtype expressed, as MCA38 cells also expressed P2X4 and P2X5 receptors at the mRNA level [6].

Another example is in human osteosarcoma. Zhang et al. showed that P2X7 receptor stimulation by BzATP (5–125 μM) activates mTOR signaling and promotes cancer survival and invasiveness [7]. Further support for the link between the P2X7 receptors and mTOR comes from a rat model of osteoarthritis [8, 9]. In these studies, the authors observed that moderate exercise intensity maintains a low level of activation of P2X7 receptors, which correlates with the mTOR pathway activation [9]. Under the moderate exercise, P2X7 receptor activation with BzATP decreased cell death and promoted autophagy [8]. However, in osteoarthritis, the effect of P2X7 receptors was the opposite in that P2X7 receptor inhibition by A740003 induced autophagy and decreased apoptosis [9]. As such, the authors suggest that the P2X7 receptor is the upstream regulator of the mTOR pathway, however, in a dynamic manner to balance between P2X7 receptor-mediated activation of inflammasomes and P2X7 receptor-mediated activation of mTOR [8, 9]. These studies share the consistent view that P2X7 receptors and ATP signaling can be the upstream regulator of mTOR signaling, although the downstream effect can differ. The current publication by Schmitt et al. is the first to point out the involvement of P2X4 receptors in the regulation of mTOR. It is interesting to note that the P2X4 subtype has been reported to interact physically and functionally with P2X7 receptors [10–12]. The association of P2X7 receptors with the NRP3 inflammasome [13] is a feature now shared with P2X4 receptors [14].

The exact mechanism of how P2X4 channel opening by ATP can lead to activation of the mTOR pathway remains to be investigated. The P2X4 receptor is an ionotropic P2 receptor that forms trimeric, non-selective cation channels with other P2X subtypes in homo- or heteromeric configurations [15]. The binding of ATP opens P2X channels, allowing non-selective cation conductance (Na⁺, Ca²⁺, and K⁺). Cell surface expression of P2X4 receptors is dynamic and can be regulated by ivermectin, which potentiates P2X4 receptor responses [16]. P2X4 receptors are often localized to internal organelles in the cytoplasms, such as observed in neurons [17], alveolar cells [18], and the ocular lens [19], to name a few. At least part of intracellular the P2X4 receptor population has been shown to be lysosomal and to play physiological roles in modulating the fusion of endolysosomes with other compartments [20, 21]. The P2X4 subtype exhibits pH sensitivity to require a higher concentration of ATP for activation under low pH in the lysosome [20, 22]. A similar pH-dependent response for P2X4 receptors has also been reported in the lamellar bodies of the alveoli [18]. The acidic pH environment has been suggested as the requirement for the re-sensitization of P2X4 receptors [23]. These observations collectively support the unique roles played by P2X4 receptors in endolysosomal compartments as a possible mediator capable of sensing extracellular ATP on the cell surface, as well as regulating internal Ca²⁺ balance in the lysosome. Interestingly, mTOR and associated complexes are reported to be localized on the surface of lysosome, which may be a key metabolic regulatory site [24]. It is possible that the lysosome may be the site of mTOR modulation by P2X4 receptors, and the regulation of local Ca²⁺ homeostasis by P2X4 receptors may influence mTOR signaling. Further investigations of the exact mechanism of P2X4 receptor-mTOR crosslink and identification of its intracellular location may be of interest to the purinergic community and the wider research field.

The study by Schmitt et al. suggests that chemotherapy-induced death of tumor cells results in the release of ROS and ATP into the tumor microenvironment. While the released ROS renders neighboring tumor cells more prone to apoptosis, this appeared to be counteracted by ATP-mediated paracrine activation of P2X4 receptors, which stimulates mTORC1, shifting the balance of these cells toward better survival. P2X4 receptors are known to play roles in tumor biology for different forms of cancers, including prostate cancer [25, 26] and breast cancer [27]. In both prostate cancer and breast cancer, P2X4 receptor activation is associated with increased survival and more aggressive growth of tumor cells [25, 27]. The implication of the study by Schmitt et al. is that the combinatory therapy targeting P2X4 receptors or mTOR may significantly amplify the effectiveness of existing anti-cancer therapies, and this may extend to other cancer types, such as breast and prostate cancers.

Author contribution

H.S-K. wrote the main manuscript text.

Funding

H.S-K. was supported by the Auckland Medical Research Foundation and the Eisdell Moore Centre.

Data availability

Data sharing is not applicable as in this case no datasets.

Declarations

Ethical approval

This article does not contain any studies with human participants or animals performed by the author.

Conflicts of interest

The author declares no competing interests.