The Effects Of Ecto-Cd39-Cd73-Adenosinergic Signaling Pathway On Malignant Glioma Associated Immunosuppression And Its Therapeutic Potentials

BackgroundAs the most prevalent form of primary brain tumor in humans, malignant gliomas (WHO Grade III and IV) are characterized by aggressive infiltration and dynamic angiogenesis associated with high morbidity and poor prognosis. Even with the enormous progresses in basic scientific research and clinical practice during the past decades, patients with newly diagnosed glioblastoma multiforme (GBM) have a median survival of14.6months, and2-3years for patients with anaplastic astrocytoma. By initiating and amplifying specific anti-tumor response, immunotherapeutic strategies bring a promising approach in the battle against this devastating neoplasm. In fact, increased disease-free survival and overall survival have been demonstrated by preliminary immunotherapeutics trials in glioblastoma patients. However, malignant gliomas are associated with numerous immunomodulatory properties, including absent tumor specific antigen expression, immune checkpoint overexpression, immunosuppressive cytokine secretion, effector lymphocyte anergy, recruitment and induction of regulatory T cells (Tregs) and M2macrophages, as well as immune inhibitory factor accumulation. Thus, anti-tumor immunity is suppressed profoundly and manipulated to promote tumor progression.Adenosine has been identified as a universal and potent immune suppressor. Briefly, pro-inflammatory ATP released from stressed or damaged cells to extracellular space is sequentially converted into AMP and adenosine by surface-expressing CD39/ectonucleoside triphosphate diphosphohydrolase-1(ENTPD1) and CD73/ecto-5’-nucleotidase (NT5E), respectively. Adenosine deaminase (ADA) is responsible for the rapid deactivation of anti-inflammatory adenosine. Thus, adenosine balance maintained by the ectoenzymes CD39-CD73-ADA is pivotal for immune homeostasis. However, excess adenosine accumulation is associated with certain pathological circumstances, such as chronic inflammations and tumors, which activates Gs-protein-coupled adenosine A2aR receptors expressed by a variety of immune cells and elevates the intracellular cAMP level. Consequently, potent immunosuppression is induced, causing abrogated T cell proliferation, Th1/Th2shifting, Treg induction as well as macrophage activation inhibition. To date, CD39-CD73-adenosine pathway has been recognized as a critical imnnunosuppressive mechanism with promising therapeutic prospect in oncology. However, several details have not been well characterized yet. For instance, while the importance of tumor-derived CD73in tumorigenesis, metastasis and immunosuppression has been revealed by in vitro and in vivo evidence, glioma CD39activity hasn’t been confirmed. On the other hand, despite co-expression of CD73with CD39on murine CD4+CD25+Foxp3+Tregs, the expression and activity of this ectoenzyme in human CD4+CD39+T lymphocytes were found almost negative.In this study, the phenotypic and functional characteristics of ectoenzymes expressed by glioma cells and glioma-infiltrating CD4+T lymphocytes were evaluated in details. Based on the distinct but complementary ectoenzyme status possessed by these two populations, we demonstrated that CD73+glioma cells contribute to the local adenosinergic immunosuppression in synergy with infiltrating CD39+T lymphocytes in the glioma microenvironment and thus might be of great potential in the malignant glioma treatment.PART I Ectoenzyme phenotypic characteristics of glioma cells1.1The ectoenzyme transcriptional expression in glioma cells To ascertain the ectoenzyme expression profiling of glioma cells, we extracted the total RNA from glioma cell lines U-87MG and T98G, and assessed the transcriptional levels of ectonucleoside triphosphate diphosphohydrolase family members, including CD39/ENTPD1, CD39L1/ENTPD2and CD39L3/ENTPD3, as well as5’-nucleotidase/CD73by reverse transcription PCR. Interestingly, both U-87MG and T98G exhibited the similar preferential expression of CD73. In contrast, expressions of CD39family members were almost negative. Quantitative RT-PCR assays confirmed this expression pattern. Despite the distinct transcriptional levels between the two cell lines, it was the dominant ectoenzyme form expressed by glioma cells. Notably, expression of CD39was low or even hardly detectable. These results were consistent with the previous study reporting the almost negative CD39expression in rat C6glioma cells. Similar transcriptional pattern was also found in another glioma cell line U-251.1.2The ectoenzyme phenotype of glioma cellsThe preferential expression of CD73rather than CD39was further validated by flow cytometry on protein level. Particularly, U-87MG expressed the highest level of CD73, while T98G had the lowest.1.3The ectoenzyme phenotype of glioma cancer stem cellsWe also examined the surface ectoenzyme expression in11different glioma cancer stem cell (gCSCs) lines (including9confirmed gCSC lines and2newly isolated gCSC lines) with flow cytometry. Among them,8(72.7%) lines possess CD73expression, while only1/11(9.1%) exhibits CD39expression. We have added the malignant glioma CD39/CD73immunohistochemical data as Figure1C. Among the19malignant glioma specimenswe tested (16glioblastoma multiforme and3anaplastic astrocytoma), CD73was expressed in89.5%(17/19) of cases, but CD39expression was merely found in21.1%(4/19), which confirmed our previous observation1.4Immunohistochemistry staining of resected malignant glioma specimensTo verify our finding in primary brain tumors, immunohistochemistry detecting ectoenzymes CD39and CD73expression in malignant glioma tissues was performed. Among all the19specimens we tested (16glioblastoma multiforme and3anaplastic astrocytoma), CD73was expressed in89.5%(17/19) of cases, but CD39expression was merely found in21.1%(4/19)(Fig.1C).Taken together, we demonstrate that glioma cells possess the preferential expression of CD73; however, the expression of CD39is almost negative, which is the rate-limiting enzyme in the nucleotide hydrolysis cascade.PART II:Ectoenzyme phenotypic characteristics of glioma-infiltrating CD4+T cells2.1CD39/CD73phenotypic characteristics of glioma-infiltrating CD4+T cellsDespite the low frequency (range:-0.2%) in the numerous cell populations in malignant gliomas, infiltrating CD4+T cells are known for their notorious role in inducing glioma associated immunosuppression. To explore the elusive effect they might exert in the local adenosinergic pathway, we collected peripheral blood and tumor specimens from newly-diagnosed malignant glioma patients, isolated the peripheral and tumor-infiltrating CD4+T lymphocytes and compared the phenotypic characteristics between these two populations (n=9). Healthy donor peripheral CD4+T cells were also included as control. As shown in Table1and Fig2A, malignant glioma patients did not exhibit ectoenzyme dysregulation in the peripheral CD4+T lymphocytes compared with healthy donors. Astonishingly, robust CD39expression was observed in the infiltrating CD4+T lymphocytes, with the prevalence of61.8±19.3%compared to8.0±5.7%in matched peripheral CD4+T lymphocytes. Considering the marked inter-individual differences, our data might also reveal the heterogeneity of adenosinergic pathway status in the malignant glioma patients. In contrast, CD73level was not altered in glioma-infiltrating CD4+T lymphocytes as CD39was. Together, our data indicate that tumor-infiltrating CD4+T lymphocytes associated with dramatically increased CD39expression exhibit distinct but complementary properties compared with the glioma cells.2.2CD26/CD73expression on human CD4+CD39+T cellsAs indicated above, glioma-infiltrating CD4+T lymphocytes are CD39highCD73low. Unlike the concordant expression in mice, surface expression and hydrolysis activity of CD73in human CD4+CD39+T lymphocytes are still in controversy. Thus, we next investigated the detailed ectoenzyme profiling of this particular CD4+CD39+T subset. Adenosine deaminase (ADA) is recognized as a pivotal negative regulator of adenosinergic signaling by adenosine deactivation. Detection of surface-bound glycoprotein CD26, which is the dominant ADA anchoring protein on human lymphocytes, provides an alternative approach to evaluate ADA expression31. As shown in Fig.3A, CD26was predominantly expressed on CD4+CD39-responder T cells rather than on the CD4+CD39+population, which suggested CD4+CD39+T cells were in favor of adenosine accumulation. However, CD73surface expression in the CD4+CD39+T cell population was11.7±6.99%, even lower than23.5±12.8%for CD4+CD39-responder T cells. Since nucleotide cascade depends on both CD39and CD73, this "defective" phenotype suggests that CD4+CD39+T cells can be insufficient to generate adequate adenosine from ATP to exert suppressive effect.2.3CD73expression on human CD4+Foxp3+regulatory T cellsIt has been reported that the prevalence of Foxp3+Treg within gliomas is correlated with tumor pathology and WHO grade32. Although the concordant surface expression of the ectoenzymes CD39and CD73had been validated in murine Tregs, our in situ data highly suggest that tumor-infiltrating T lymphocytes, many of which were Tregs, lacked CD73expression. Therefore, we moved forwards to verify the CD73surface expression on the natural Tregs (nTregs) from glioblastoma patients as well as adaptive Tregs (iTregs) induced by suppressive TGF-β respectively. Compared to CD4+Foxp3-conventional T cells, peripheral natural CD4+Foxp3+Tregs from patients with glioblastoma exhibited higher expression of Treg marker CD39, but not CD73. Similarly, adaptive CD4+Foxp3+iTregs also expressed low level of CD73. In summary, we observe that human CD39+T cells lack CD73surface expression, which suggests that certain compensatory mechanism probably exists to induce local adenosinergic immunosuppression within glioma microenvironment. PART III:The enzymatic activities of glioma cells and CD4+T lymphocyte subsetsDespite the defective phenotypic characteristics of both glioma cells and CD4+CD39+T cells as we indicated above, whether they are associated with nucleotide hydrolysis deficiencies remains undefined. To understand their functional status better, we determined the inorganic phosphate generated during the ectoenzyme-mediated nucleotide dephosphorylation via malachite green-molybdate spectrophotometry assays.3.1The enzymatic activities of glioma cellsOf note, U-87MG and T98G glioma cells were tested here since they were found with the highest and lowest levels of CD73expression respectively. As expected, both glioma cells hydrolyzed exogenous AMP robustly, which could be abrogated by a specific CD73inhibitor APCP. Consistent with higher CD73level, U-87MG cells exhibited much higher5’-nucleotidase (-12.0folds of T98G). Interestingly, neither CD39-deficent U-87MG norT98G induced significant ATP hydrolysis.3.2The enzymatic activities of CD4+T lymphocyte subsetsOn the contrary, single cell-sorted CD4+CD39+T lymphocytes exhibited significant ENTPDase activity, which could be blocked by a CD39inhibitor ARL67156. However,5’-nucleotidase activity was not observed. CD4+CD39-T lymphocytes were deficient in either ENTPDase or5’-nucleotidase activities, which was not surprising since they were categorized as conventional/responder T lymphocytes without abundant ectoenzyme expressions.3.3The synergic effect between heterogenous CD39and CD73Till now, we have demonstrated that the deficient phenotypic and functional characteristics possessed by glioma cells and infiltrating CD4+CD39+T lymphocytes respectively are distinct but complementary, which inspired us to evaluate whether they could work synergistically to induce local adenosinergic immunosuppression. To mimic the CD73activity, we utilized the soluble5’-nucleotidase purified from Crotalus atrox venom33, which allowed us to focus on the synergistic effect between CD39and CD73without considering the complexity of entire glioma cells. Here, we found that much more phosphate was generated from AMP by CD39+T cells in the presence of soluble5’-nucleotidase than the cells alone. This synergistic effect was specifically blocked by CD73inhibitor APCP. We also determined high phosphate concentration derived from CD39+T cells in the presence of soluble5’-nucleotidase compared to T cells alone. Because soluble5’-nucleotidase cannot hydrolyze ATP directly, this particular piece of evidence indicates that the synergic effect between CD39and CD73did exist, which can be abrogated by APCP.PART IV:The synergic effect between CD39and CD73on CD4+T cell proliferationTo ascertain our hypothesis that immunoregulatory CD4+CD39+T lymphocytes could induce more significant immunosuppressive effect in synergy with glioma cells, CFSE-labeled single-sorted CD4+CD39-responder T lymphocytes were co-cultured with autologous CD4+CD39+T lymphocytes in the absence or presence of glioma cells for4days. The suppression%of responder cell proliferation was then calculated. Before co-culture, glioma cells were pre-treated with mitomycin C, an anti-mitotic agent to arrest cell division, to minimize the potential nutritional deprivation of responder cells.Consistent with previous reports, the proliferation of CD4+CD39-responder T lymphocytes was inhibited by autologous CD4+CD39+T lymphocytes, with the suppression%of28.5±4.0%. In contrast, U-87MG glioma cells alone did not affect the proliferation of CD4+CD39-responder T lymphocytes. Interestingly, more significant proliferation suppression was induced by CD4+CD39+T lymphocytes in the presence of U-87MG glioma cell (47.8±3.5%vs.28.5±4.0%). Either CD39inhibitor ARL67156or CD73inhibitor APCP could alleviate this synergistic suppression, suggesting that CD39+lymphocyte and CD73+glioma interacted with each other and prompted proliferation suppression. Likewise, glioma cell T98G induced a similar but much less significant effect, which is consistent with a much lower CD73expression. The proliferation inhibition was also abrogated by adenosine receptor A2aR antagonist SCH58261, indicating the participation of this Gs protein-coupled receptors in the adenosinergic signaling.PART V:Large-scale ectoenzyme-related prognosis analysis in patients with glioblastomaThe Cancer Genome Atlas (TCGA) is a project, begun in2005, to catalogue genetic mutations responsible for cancer, using genome sequencing and bioinformatics. TCGA represents an effort in the War on Cancer that is applying high-throughput genome analysis techniques to improve our ability to diagnose, treat, and prevent cancer through a better understanding of the genetic basis of this disease.The large-scale prognosis analysis using TCGA GBM dataset implied a significant correlation between tumor CD73expression and disease prognosis. Glioblastoma patients with CD73mRNA downregulation had a prolonged median disease-free survival of10.4months, compared to6.7months of those without CD73downregulation. CD73mRNA downregulation also benefited the median overall survival but was not statistically significant (15.3months vs.14.0months). The survival evidence indicates that CD73is not only present in glioma, but also of considerable therapeutic potential.