No abstract

In the tumor microenvironment (TME) ATP and its receptor P2X7 exert a pivotal influence on cancer growth and tumor–host interactions. Here we analyzed the different effect of P2X7 genetic deficiency versus its antagonism on response against P2X7-expressing implanted tumors. We focused on immune cell expression of ATP degrading enzymes CD39 and CD73 and in vivo measured TME’s ATP. The immune infiltrate of tumors growing in P2X7 null mice shows a decrease in CD8+ cells and an increased number of Tregs, overexpressing the fitness markers OX40, PD-1, and CD73. A similar Treg phenotype is also present in the spleen of tumor-bearing P2X7 null mice and it is paralleled by a decrease in proinflammatory cytokines and an increase in TGF-β. Differently, systemic administration of the P2X7 blocker A740003 in wild-type mice left unaltered the number of tumor-infiltrating CD8+ and Treg lymphocytes but increased CD4+ effector cells and decreased their expression of CD39 and CD73. P2X7 blockade did not affect spleen immune cell composition or ectonucleotidase expression but increased circulating INF-γ. Augmented CD73 in P2X7 null mice was mirrored by a decrease in TME ATP concentration and nucleotide reduced secretion from immune cells. On the contrary, TME ATP levels remained unaltered upon P2X7 antagonism, owing to release of ATP from cancerous cells and diminished ectonucleotidase expression by CD4+ and dendritic cells. These data point at P2X7 receptor as a key determinant of TME composition due to its combined action on immune cell infiltrate, ectonucleotidases, and ATP release.

Chemotherapy-induced immunogenic cell death can favor dendritic cell (DC) cross-priming of tumor-associated antigens for T cell activation thanks to the release of damage-associated molecular patterns, including ATP. Here, we tested the hypothesis that in acute myeloid leukemia (AML), ATP release, along with its well-known immune stimulatory effect, may also contribute to the generation of an immune suppressive microenvironment. In a cohort of AML patients, undergoing combined daunorubicin and cytarabine chemotherapy, a population of T regulatory cells (Tregs) with suppressive phenotype, expressing the immune checkpoint programmed cell death protein 1 (PD-1), was significantly increased. Moving from these results, initial in vitro data showed that daunorubicin was more effective than cytarabine in modulating DC function toward Tregs induction and such difference was correlated with the higher capacity of daunorubicin to induce ATP release from treated AML cells. DCs cultured with daunorubicin-treated AML cells upregulated indoleamine 2,3-dioxygenase 1 (IDO1), which induced anti-leukemia Tregs. These data were confirmed in vivo as daunorubicin-treated mice show an increase in extracellular ATP levels with increased number of Tregs, expressing PD-1 and IDO1+CD39+ DCs. Notably, daunorubicin failed to induce Tregs and tolerogenic DCs in mice lacking the ATP receptor P2X7. Our data indicate that ATP release from chemotherapy-treated dying cells contributes to create an immune suppressive microenvironment in AML.

No abstract

Acute myeloid leukemia (AML) is a common adult leukemia often arising from a preexistent myelodysplastic syndrome (MDS). High mortality rates of AML are caused by relapse and chemoresistance; therefore, we analyzed the role of P2X7 receptor (P2X7R) splice variants A and B in AML progression and response to chemotherapy. The expression of P2X7RA and P2X7RB was investigated in samples obtained from MDS and AML untreated subjects or AML patients in relapse or remission after chemotherapy. Both P2X7RA and P2X7RB were overexpressed in AML versus MDS suggesting a disease-promoting function. However, in relapsing patients, P2X7RA was downmodulated, while P2X7RB was upmodulated. Treatment with daunorubicin (DNR), one of the main chemotherapeutics for AML, upregulated P2X7RB expression while reducing P2X7RA mRNA in AML blasts. Interestingly, DNR administration also caused ATP release from AML blasts suggesting that, following chemotherapy, activation of the receptor isoforms via their agonist will be responsible for the differential survival of blasts overexpressing P2X7RA versus P2X7RB. Indeed, AML blasts expressing high levels of P2X7RA were more prone to cell death if exposed to DNR, while those overexpressing P2X7RB were more vital and even protected against DNR toxicity. These data were reproducible also in HEK-293 cells separately expressing P2X7RA and B. P2X7RA facilitation of DNR toxicity was in part due to increased uptake of the drug inside the cell that was lost upon P2X7RB expression. Finally, in an AML xenograft model administration of DNR or the P2X7R antagonist, AZ10606120 significantly reduced leukemic growth and coadministration of the drugs proved more efficacious than single treatment as it reduced both P2X7RA and P2X7RB levels and downmodulated c-myc oncogene. Taken together, our data suggest P2X7RA and P2X7RB as potential prognostic markers for AML and P2X7RB as a therapeutic target to overcome chemoresistance in AML relapsing patients.