| Background and aimsPurinergic signaling has three essential components:source of the extracellular nucleotides, specific receptors for these molecular transmitters (or their derivatives), and ectonucleotidases(e.g. CD39/ENTPD1) that modulate cell responses by hydrolyzing nucleotides to nucleosides.Extracellular adenosine triphosphate (ATP) functions as a novel danger signal that boosts antitumor immunity and can also directly kill tumor cells. ATP initiates intracellular signaling cascades that participate in many pathophysiological processes e.g. proliferation, differentiation, apoptosis, inflammation, and metabolism. There are two P2receptor families that associate with ATP and other nucleoside triphosphates and diphosphates:seven P2X receptors (P2X1-7) that are ATP-gated ion channels and eight G protein-coupled P2Y receptors (P2Y1,2,4,6,11-14)-Different P2receptors have differential agonist affinity/specificity whereby modulating different cellular functions.High levels of pericellular ATP are thought to exert antitumor activity through various molecular mechanisms. it has been shown that ATP promotes antitumor immune responses and exhibits direct cytotoxicity on many types of tumor cells. Among the fifteen P2receptors, five subtypes specifically P2X5, P2X7, P2Y1, P2Y2, and P2Y11(human exclusive), have been associated with the direct tumor-killing functions of ATP, but the precise molecular mechanisms remain somewhat unclear.mTOR is a well-known sensor for extracellular nutrients and growth factors converging many crucial signals e.g. Ras/mitogen-activated protein kinase (MAPK) and PI3K/AKT to control tumor cell growth. Signal transduction networks are very complex and blockade of one component often disrupts key negative feedback mechanisms resulting in aberrant activation of compensatory pathways. As such, further defining the components of cytotoxic ATP-initiated purinergic signaling pathways is critical for the development of effective cancer therapies.Material and Methods1. Animal research and cell culture2. MCA38and B16/F10proliferation and cell viability assay3. Real-time and Dynamic Monitoring of Cell Growth by xCELLigence RTCA MP System4. Clonogenic assay5. Antagonist-treatment experiments6. Protein Purification and Western Blotting7. Reverse transcription-PCR (RT-PCR)8. Generation of P2X7deficient cell lines by lentivirus P2X7shRNA9. Ethidium bromide uptake assay10.Intracellular nucleotide measurements by high-performance liquid chromatography (HPLC)11. Tumor inoculations12. Statistical analysisResult1ã€Antitumor actions of ATP:in vitro and in vivo studiesBrief exposure of eATP (as short as15min) was able to effectively kill both types of tumor cells (MCA38and B16/F10cell), in a time-and dose-dependent manner. Cytotoxic ATP impacts many aspects of tumor cell growth/survival:viability and proliferation as well as clonogenicity in vitro; and in vivo tumor growth in C57BL/6wild type mice. Tumor cells responded to ATP cytotoxicity immediately, as evaluated by real-time measurement of tumor cell growth. Noteworthily, tumor cell death observed here was exclusively attributed to the action of ATP as ADP or adenosine showed no effects on tumor growth even at very high concentrations (up to10mM). We also observed that ATP markedly enhanced LC3-â…¡ levels in tumor cells, a sensitive marker for autophagy, growing in complete growth medium containing10%serum, also in a time-and dose-dependent fashion.2. Signaling regulatory networks of PI3K/AKT, AMPK and mTOR impacted by ATPWe first examined the phosphorylation of pathway components of mTOR (inclusive of mTOR, p70S6kinase (S6K), S6ribosomal protein, and translation repressor protein4E-BP1), PI3K/AKT (AKT and PRAS40), and AMPK (AMPKα, AMPK(3and acetyl-CoA carboxylase (ACC)) in response to eATP. Time-and dose-dependent decreases in phosphorylated levels of mTOR and PI3K/AKT signaling molecules concomitant with increases in AMPK cascade were noted. These effects occurred regardless of the cell growing conditions (serum-starvation medium or complete growth medium).To further delineate regulatory networks for ATP-induced alterations in mTOR, PI3K/Akt and AMPK signaling, we performed the following experiments using pharmacologic inhibitors to these pathways. Phosphorylation of S6K was used to indicate changes in activation status of mTOR. First, okadaic acid (OA), a potent inhibitor of protein phosphatase1(PP1) and protein phosphatase2A (PP2A) preventing p-Akt depletion, fully abolished ATP-mediated depletion of p-AKT whereas with no effects on depletion of its conventional target p-PRAS40(Thr246). However, okadaic acid failed to reverse ATP-induced inhibition on mTOR and cell growth. In parallel, inhibitor to PI3K (LY294002) could almost completely block basal phosphorylation of AKT-PRAS40-S6K. As expected, rapamycin (inhibitor of mTOR) suppressed S6K phosphorylation, to a similar extent as ATP.Blockade of AMPK activation by compound C (CC) completely rescued ATP-elicited de-phosphorylation of S6K. ATP-mediated depletion of p-PRAS40,(as suggested by its name proline-rich AKT substrate of40KDa) a well-recognized classical downstream target of AKT, was near wholly abolished by compound C, as similar to that of p-S6K. These data are implicative of a novel AMPK-PRAS40-mTOR axis responding to tumoricidal ATP.Finally, we explored the links between mTOR signaling and tumor cell autophagy that are altered by eATP. Indeed, mTOR is thought to suppress autophagy. We observed that rapamycin dramatically enhanced LC3-â…¡ levels in tumor cells, similar to ATP. Furthermore, okadaic acid or compound C could substantively abrogate LC3-â…¡ at basal levels or induced by ATP. However, okadaic acid or compound C alone failed to rescue ATP-elicited tumor cell death. Indeed, compound C treatment alone slightly but significantly reduced cell growth. Moreover, blockade of either PI3K/AKT (by LY294002) or mTOR (by rapamycin), or both, was able to induce cell death but with much less potency when compared to ATP.3ã€P2X7receptor is the control nexus of ATP-elicited tumor cell deathWe analyzed mRNA expression of P2receptor panels in MCA38cells using RT-PCR. MCA38cells expressed mRNA transcripts for P2X3(minimal), P2X4, P2X5, P2X6(weaker), P2X7, P2Y1, and P2Y12-14. We then treated tumor cells with BzATP, a more potent prototypic P2X7receptor agonist. BzATP triggered growth inhibition and signaling alterations of tumor cells near exactly recapitulates the same pattern observed with ATP. In contrast, UTP, an agonist of P2X5, P2Y2, P2Y4, and P2Y6, exhibited no effects on tumor cell growth. Moreover, the global P2receptor antagonist suramin (inhibiting P2X1-3, P2X5, P2Y1-2, P2Y6, and P2Y11-13) was unable to relieve the suppressive actions of ATP on phosphorylation of AKT, PRAS40and S6K as well as tumor cell growth, although it could block ATP-induced activation of AMPK signaling. In parallel, selective P2X7receptor antagonist KN62virtually counteracted ATP-mediated signaling alterations, in a time-and dose-dependent manner.4ã€Knockdown of P2X7protects tumor cells from antitumor function of ATPWe generated stable P2X7deficient tumor cells using lentiviral shRNA method. Out of four P2X7shRNAs, one effectively blocked P2X7protein expression in both tumor cell lines. These are defined as P2X7KD and used for all subsequent analyses. P2X7deficiency near fully deviated all cytostatic ATP-mediated effects on tumor cells:signaling regulatory networks, cell growth, colony-forming capacity, and autophagy. Further, knockdown of P2X7conferred tumor cells with higher growth and clonogenic potential, as compared to control knockdown (Ctrl KD) cells. 5ã€Characterization of extracellular ATP-elicited P2X7receptor channel functionsFirst, we examined the expression of P2X7splice variants P2X7(a) and P2X7(k) in tumor cells, and both MCA38and B16/F10cells only express the P2X7(a) variant at the mRNA level. Second, P2X7channel function, specifically non-selective pore activity, was evaluated using ethidium bromide uptake assays. We noted that ATP markedly stimulates ethidium bromide uptake in both cell lines. Third, we then measured levels of intracellular ATP, ADP and AMP by HPLC in tumor cells pulsed treated with ATP for15and30min. Substantial decreases in intracellular ATP levels were observed, concurrent with significant increases in intracellular ADP and AMP levels as well as AMP/ATP ratios.Fourth, by performing antagonist studies, we then confirmed that PI3K/AKT and AMPK/mTOR signaling altered by P2X7activation are not linked to other native pore-forming transporters (inclusive of pannexin-or connexin-type channels) nor reactive oxygen species (ROS), as previously reported. Neither carbenoxolone (a pharmacological inhibitor of pannexin-or connexin-type channels) nor N-acetyl-cysteine (a broad anti-oxidant) could substantively impact ATP-induced P2X7signaling cascades.We now show that P2X7activation can also induce tumor cell autophagy. Importantly, none of the specific pharmacological inhibitors (e.g. Z-VAD-fmk, a pan-caspase inhibitor or the necroptotic inhibitor necrostatin-1; various doses of compounds ranging from12.5to200μM were tested) when tested alone could counteract ATP-induced tumor cell death.6ã€Calcium signaling is not involved in ATP-P2X7mediated tumor-killing activityTo determine whether the influx of Ca2+is involved in ATP-evoked cellular effects, two experimental approaches were employed using two chemical compounds that differently alter cytosolic Ca2+levels:BAPTA-AM, a selective cell-permeable calcium chelator to block intracellular Ca2+stores, and thapsigargin (TG), a potent inhibitor of endoplasmic reticulum Ca2+-ATPase causing an immediate increase in cytoplasmic Ca2+levels. We found that BAPTA-AM failed to relieve ATP-induced mTOR inactivation and LC3-â…¡ increase. Instead, BAPTA-AM completely abolished basal levels of PI3K/AKT and mTOR signaling and LC3-â…¡ with no effects on AMPK signaling, associated with marked induction of cell death. TG were able to induce immediate cell death and activation of AMPK pathway, but exerted no effects on PI3K/AKT and mTOR signaling transduction nor on autophagyConclusion:1ã€Tumor cell growth and autophagy that are co-responsive to cytotoxic ATP, in a time-and dose-dependent fashion.2ã€The cytotoxicity of ATP is mediated by two independent signaling pathways, namely AMPK-PRAS40-mTOR and PI3K/AKT, that synergistically perturb (otherwise tightly controlled) balance between growth and autophagy in tumor cells whereby eventually causing cell death.3ã€P2X7receptor is the upstream control node of ATP-initiated signaling regulatory networks. ATP-P2X7elicited tumor cell death is mediated via both apoptosis and necrosis, in keeping with fluxes in intracellular nucleotides.4ã€Calcium signaling is not involved in ATP-P2X7mediated tumor-killing activityIn summary, we infer that high-levels of extracellular ATP act on one novel P2X7-AMPK-PRAS40-mTOR axis and one conventional P2X7-PI3K/AKT axis to induce autophagy and inhibit growth that ultimately result in tumor cell death. |