Search

Extracellular adenosine accumulates in tumors and causes suppression of immune cells. Suppressive adenosine signaling is achieved through adenosine A2A and A2B receptors, which are Gs coupled, and their activation elevates cAMP levels. Gs-coupled GPCR signaling causes cAMP accumulation, which plays an anti-inflammatory role in immune cells. Protein kinase A (PKA) and exchange protein directly activated by cAMP (Epac) are two intracellular receptors of cAMP. In this study we showed that adenosine receptor signaling polarizes activated murine dendritic cells (DCs) into a tumor-promoting suppressive phenotype. Adenosine receptor signaling activates cAMP pathway and upregulates the negative regulators of NF-κB but does not influence phosphorylation of immediate inflammatory signaling molecules downstream of TLR signaling. Pharmacologic activation of both PKA and Epac pathways by specific cAMP analogues phenocopied the effects of adenosine signaling on murine DCs, such as suppression of proinflammatory cytokines, elevation of anti-inflammatory IL-10, increased expression of regulators of NF-κB pathway, and finally suppression of T cell activation. Inhibition of effector cytokine, IL-12p40 production, and increased immunosuppressive IL-10 production by adenosine signaling is significantly reversed only when both PKA and Epac pathways were inhibited together. Adenosine signaling increased IL-10 secretion while decreasing IL-12p40 secretion in human monocyte-derived DCs. Stimulation of both PKA and Epac pathways also caused combinatorial effects in regulation of IL-12p40 secretion in human monocyte-derived DCs. Interestingly, PKA signaling alone caused similar increase in IL-10 secretion to that of adenosine signaling in human monocyte-derived DCs. Our data suggest adenosine/cAMP signaling targets both PKA/Epac pathways to fully differentiate DCs into a suppressive phenotype.

The accumulation of high levels of adenosine in tumors activates A(2A) and A(2B) receptors on immune cells and inhibits their ability to suppress tumor growth. Deletion of adenosine A(2A) receptors (A(2A)ARs) has been reported to activate antitumor T cells, stimulate dendritic cell (DC) function, and inhibit angiogenesis. In this study, we evaluated the effects of intermittent intratumor injection of a nonselective adenosine receptor antagonist, aminophylline (AMO; theophylline ethylenediamine) and, for the first time to our knowledge, a selective A(2B)AR antagonist, ATL801. AMO and ATL801 slowed the growth of MB49 bladder and 4T1 breast tumors in syngeneic mice and reduced by 85% metastasizes of breast cancer cells from mammary fat to lung. Based on experiments with A(2A)AR(-/-) or adenosine A(2B) receptor(-/-) mice, the effect of AMO injection was unexpectedly attributed to A(2B)AR and not to A(2A)AR blockade. AMO and ATL801 significantly increased tumor levels of IFN-gamma and the IFN-inducible chemokine CXCL10, which is a ligand for CXCR3. This was associated with an increase in activated tumor-infiltrating CXCR3(+) T cells and a decrease in endothelial cell precursors within tumors. Tumor growth inhibition by AMO or ATL801 was eliminated in CXCR3(-/-) mice and RAG1(-/-) mice that lack mature T cells. In RAG1(-/-) mice, A(2B)AR deletion enhanced CD86 expression on CD11b(-) DCs. Bone marrow chimera experiments demonstrated that CXCR3 and A(2B)AR expression on bone marrow cells is required for the antitumor effects of AMO. The data suggest that blockade of A(2B)ARs enhances DC activation and CXCR3-dependent antitumor responses. The Journal of Immunology, 2012, 188: 198-205.

Inhibition of proximal TCR signals by adenosine A2A receptors (A2AR) is important for T cell development and maintaining T cell quiescence in vivo. Deletion of A2ARs causes accelerated growth of B16 and MB49 tumors, which is associated with reduced T cell accumulation in tumors and decreased CD127 expression and interleukin-7 responsiveness. Here we showed that ectopic expression of CD127 in A2AR deficient mice restores normal numbers of thymic populations and IL-7 responsiveness of T cells. Ectopic expression of CD127 in A2AR deficient mice also caused accumulation of similar numbers of T cells in the periphery as compared to A2AR proficient mice. Growth of B16 tumors in A2AR deficient mice was accelerated as compared to A2AR proficient mice. However, growth of B16 tumors was comparable between A2AR deficient or proficient mice ectopically expressing CD127. Adenosine A2A receptor deletion or ectopic CD127 expression did not change Vbeta selection suggesting changes in T cell repertoire may not be involved in increased tumor growth and reduced T cell numbers in A2AR deficient mice. Interestingly, tumor growth in A2AR and CD127 double deficient mice was slower as compared to CD127 deficient mice. Finally, adenosine preconditioning of engineered T cells significantly reduced lung colonization of B16OVA tumors and persistence of engineered T cells. Overall these results suggest that CD127 regulation by adenosine is important for T cell maintenance and functional responses in vivo and blockade of A2AR to improve anti-tumor immune responses should be further promoted by providing signals for T cell survival.

Polarization of T helper subsets into different functional phenotypes strongly affects the progression of immune-related diseases. One of the hallmarks of immune cell activation and inflammation is the elevation of extracellular adenosine and increased expression of adenosine A2A receptors in activated immune cells. Adenosine strongly inhibits activation of helper T cells by elevating cAMP concentrations, which activate PKA and EPAC proteins to regulate cellular responses. Here we showed that adenosine strongly inhibits T cell accumulation rather than differentiating into functional subsets in the presence of polarizing conditions. Different aspects of adenosine mediated T cell suppression were phenocopied by specific activation of PKA and/or EPAC pathways. Adenosine can suppress phospho-activation of Akt pathway to promote T cell quiescence and to inhibit immediate downstream events after TCR stimulation. One of the targets for Akt is Foxo1. Foxo1 is known to suppress T cell proliferation. Our results showed that adenosine receptor stimulation reduced inhibitory phosphorylation of Foxo1 downstream of Akt. Foxo1 inhibitor, AS1842856, completely reversed the inhibition of T cell accumulation by adenosine signaling. Our results suggest that adenosine by activating PKA and/or EPAC pathways and by sustaining Foxo1 activation suppresses T cell activation and accumulation. These findings have important implications to develop novel interventions to regulate helper T cell responses in different pathological conditions.

Metabolic and immune related activities converge as main triggers of adenosine accumulation in extracellular space. Adenosine by engaging adenosine A2A and A2B receptors strongly suppresses innate and adaptive immune responses. Although adenosine receptors are being targeted in preclinical and clinical studies how different danger signals are regulated by adenosine is poorly understood. Here we showed that adenosine receptor stimulation strongly inhibited inflammatory responses while sparing type-I interferon responses downstream of different danger signals in dendritic cells and macrophages. Mechanistically, danger signals associated with MyD88-dependent inflammatory pathways such as LPS and CpG but not the danger signals associated with IRF3/Type-I interferon pathways such as pA:U and cGAMP increase the expression adenosine A2A and A2B receptors. Adenosine was shown to increase the expression of NR4A nuclear hormone receptors to inhibit NF-κB activation. Although adenosine did not influence NF-κB phosphoactivation expression of anti-inflammatory NR4A1 was increased after adenosine receptor stimulation in the presence of TLR ligands known to activate MyD88 pathway but not in the presence of cGAMP and pA:U. Overall these results indicate that there is a differential modulation of danger signaling by adenosine rather than overall supression. Our results have important implications for developing combinatorial approaches to target adenosine and danger signaling pathways to cure immune-related diseases.

Activation of immune cells is associated with increased expression of adenosine receptors. Extracellular adenosine is present in primary and secondary lymphoid organs and limit the amplitude of immune responses suggesting that antagonists for adenosine receptors can be targeted to improve the efficacy of vaccines. Here we showed that vaccine adjuvant Monophosphoryl Lipid A (MPLA) strongly increased adenosine A2A and adenosine A2B receptor expression in primary macrophages and dendritic cells. Adenosine A2A but not adenosine A2B receptor blockade significantly increased both numbers and percentages of antigen specific endogenous T cells in both spleen and draining lymph nodes after primary immunizations with MPLA and ovalbumin. We observed a similar increase in both endogenous or adoptively transferred T cells in mice with myeloid-specific deletion of A2A receptors after primary immunization. Adenosine receptor blockade significantly increased the total IgG titers and IgG2c/IgG1 ratio after rechallenge with ovalbumin. These results suggest that adenosine A2A receptor blockade potentially improve vaccine efficacy by promoting both cellular and humoral immune responses and promoting Th1 type immunity. Our results also suggest that A2ARs on antigen presenting cells are important targets to improve vaccine efficacy by adenosine A2A receptor blockade. These findings have important implications for the design of novel and more efficacious vaccine formulations.

Extracellular adenosine concentrations and expression adenosine receptors are elevated after cell death, injury and inflammation. Antigen presenting cells are one of the primary targets for adenosine to suppress T and NK cell responses. However, molecular mechanisms for adenosine regulation of dendritic cells are poorly understood. Here we show that adenosine receptor stimulation strongly suppresses dendritic cell activation. Adenosine did not affect phosphoactivation of inflammatory signaling pathways such as MAPKs, NF-κB and IRF3. Adenosine receptor signaling increased intracellular cAMP accumulation. Specific cAMP analogs for EPAC and PKA pathways phenocopied the effects of adenosine on dendritic cells especially when used together. Adenosine-mediated suppression of inflammatory/effector cytokines is associated with increased expression of anti-inflammatory c-Fos and NR4A receptors. NR4A and cFos expression were also increased by EPAC and PKA-specific cAMP analogs. Addition of conditioned mediums from dendritic cells prestimulated with stable adenosine analog or PKA and EPAC specific cAMP analogs together inhibited the IFNγ production by activated T cells. Overall our data suggest that adenosine can potentially target both PKA and EPAC pathways and increase the expression of NR4A nuclear orphan receptors and cFos to regulate dendritic cell responses. Our findings have important implications for the development of novel therapies for immune related diseases such as cancer by targeting adenosine receptors.

ATP-binding Cassette Transporter G1 (ABCG1) promotes cholesterol efflux from cells for intracellular cholesterol homeostasis. ABCG1 has been shown to regulate macrophage and lymphocyte immune responses. In this study, we demonstrate a role of ABCG1 as a novel modulator of tumor immunity. MB49 bladder carcinoma or B16 melanoma cells were injected into Abcg1-/- or B6 mice subcutaneously. Both groups of mice were fed with either high-fat diet or chow diet starting a week before injection of tumor cells. Abcg1-/- mice on high-fat diet had dramatically reduced (80 %) tumor growth compared to B6 mice. Abcg1-/- mice on high-fat diet also showed diminished tumor metastasis and prolonged survival. Furthermore, we showed that reduced tumor size in Abcg1-/- mice on high-fat diet was associated with an increase in the frequency of NK cells and T cells and decrease in the frequency of macrophages and Tregs in tumor. Selective inhibition of ABCG1 in myeloid cells, but not in T cells, reduced tumor growth in vivo, demonstrating that impaired tumor growth in the absence of ABCG1 is mediated through myeloid cell-intrinsic factors. Moreover, ABCG1 deficiency caused an increase in apoptosis of macrophages and a shift of macrophages to an M1 phenotype in the tumor under high-fat diet conditions. Overall, our study shows that absence of ABCG1 inhibits tumor growth through regulation of macrophage function within the tumor and provides a link between diet, cholesterol homeostasis, and cancer.

Extracellular adenosine is generated as part of immune escape programming orchestrated by tumor cells and regulatory cells. Adenosine A2A receptors (A2AR) on innate and adaptive immune cells are high affinity targets for adenosine-mediated immune suppression. In this study we demonstrate that myeloid-deletion of adenosine A2A receptor gene (adora2a) strongly inhibit tumor growth and metastasis. This was associated with increased APC function of macrophages and increased the numbers and activation of cytotoxic lymphocytes. Surprisingly, lymphoid deletion of adora2a markedly increased tumor growth and decreased the numbers of effector/memory T cells in the tumor. Mechanistic studies revealed that adora2a deletion in T cells severely impacts generation of the long/lived memory/effector T cell population within the tumor consistent with recent studies indicating that A2AR activation reduces activation-induced cell death. The results of this study have important implications for the development cell-based therapies targeting adenosine receptors.

Adenosine is generated in response to cellular stress such as hypoxia and inflammation and acts as a “metabokine” to modulate immune responses and prevent tissue damage. The accumulation of high levels of adenosine in tumors activates adenosine A2A and A2B receptors on immune cells to attenuate tumor rejection. Previously, we have shown that A2BR blockade inhibits growth of syngeneic tumors by enhancing innate and adaptive immune responses. However, the cellular targets that mediate inhibition of tumor growth after A2B receptor blockade are unknown. In this study, we tested the activity of innate and adaptive immune cells isolated from bladder or melanoma tumors grown in mice reconstituted with mixed bone marrows from A2BR-/- and A2BR+/+ animals. While A2BR deficiency did not influence the activation of NK cells and T cells, A2BR-/- antigen presenting cells were activated more than A2BR-/- cells. Selective deletion of A2BRs on myeloid cells or CD11c expressing cells significantly delayed the growth of syngeneic melanoma. The results show that antigen-presenting cells are direct cellular targets for tumor growth inhibition after A2BR blockade. The results of this study have important implications for therapeutic use of A2B receptor blockade or deletion for treatment of cancer.

Adenosine generated as a result of normal metabolic activity is elevated in response to tissue stress, damage, or hypoxia. By stimulating adenosine A2A receptors (A2ARs) adenosine links metabolic activity to regulation of normal physiological functions such as sleep, blood flow, and inflammation. Although stimulation of A2ARs on activated T cells and APCs acutely suppresses inflammation and immune killing of tumor cells, global A2AR deletion has variable effects on the growth rate of syngeneic tumors, suggesting that A2AR signaling has opposing effects on immune function. Here we show that A2ARs are required for homeostatic maintenance of naïve T cells. In global A2AR-deficient mice naïve T cells undergo apoptosis, which drastically lower their number in blood and in peripheral lymph nodes. Lymphoid deletion of A2ARs, adoptive co-transfer and mixed bone-marrow reconstitution experiments showed that cell intrinsic A2AR signaling is required for T cell homeostasis. A2AR signaling is also required for interleukin 7 to optimally facilitate T cell survival ex vivo. In mice injected with syngeneic melanoma cells, lymphoid-specific A2AR deletion decreases T cell numbers in the periphery and in the tumor, reduces the proportion of tumor infiltrating T cells with an activated/memory phenotype, and markedly accelerates tumor growth. The results indicate that homeostatic adenosine production and A2AR signaling are required in naïve T cell maintenance and immune defense against tumors.

The activation of adenosine A2A or A2B receptors on immune cells inhibits their ability to suppress tumor growth. In this study, we evaluated the anti-tumor effects of aminophyllilne (AMO) a nonselective antagonist of A2A and A2B receptors that is used clinically to treat asthma. Intra-tumor injection of AMO inhibited the growth of syngeneic MB49 bladder cells subcutaneously injected into C57BL/6 mice and the growth and lung metastasis of luminescent syngeneic 4T1 breast tumor cells inoculated into the mammary fat pats of Balb/c mice. Although AMO inhibited the growth of MB49 in A2AR-/- mice it failed to do so in Rag1-/-, A2BR-/- or CXCR3-/- mice. The selective deletion of the A2BR or CXCR3 from bone marrow cells also blocked tumor suppression by AMO. Transcript levels of interferon-γ and CXCL10, and activation of CD4+ and CD8+ T cells, were increased in tumors treated with AMO, while numbers of CD34+/CD45- endothelial cell precursors were decreased. We conclude that: 1) the anti-tumor effects of non-selective adenosine receptor blockade by AMO are primarily mediated by the A2BR; 2) CXCR3 receptors are required for anti-tumor effects of AMO and influence both T cell trafficking and chemokine-mediated angiogenesis; and 3) AMO may be useful for proof-of-principle studies to determine the efficacy of adenosine receptor antagonists as tumor suppressors alone or in conjunction with tumor vaccines.

TLR4 uses both MyD88 and TRIF adaptors to induce inflammation and to enhance antigen presentation. Previously, we reported that the low-toxicity vaccine adjuvant monophosphoryl Lipid A (MPLA) is a TRIF-biased TLR4 agonist compared to LPS. However, not all TRIF-dependent genes were optimally induced after MPLA stimulation. Further complications were that both MPLA and LPS are heterogeneous and have several structural differences compared to each other. We now report that homogeneous preparations of synthetic MPLA (sMLA) selectively induce TRIF-dependent genes such as IFIT2 but not other TRIF-dependent genes such as IFNβ or MyD88/TRIF co-dependent genes such as COX-2 when compared to synthetic diphosphoryl Lipid A (sDLA). Consideration of the promoters for these genes revealed that IFIT2 requires fewer transcription factors for it to be active (IRF-3 and p65), whereas requirements for COX-2 and IFNβ transcription are more complex (p50/p65, AP-1, IRF-3). sMLA weakly induced signaling pathways leading to p50/p65 and AP-1 activation, but activation of IRF3 and p38 was largely unaffected. Thus, low toxicity sMLA may drive Trif-biased signaling because the promoters of its targeted genes are more likely to be 'simple', while those associated with MyD88 are more often complex. Results of this study identify a new model for TRIF-biased stimulation of TLR4 by sMLA. This work was supported by the NIH/NIAID and the AHA.

TLR4 stimulation leads to the activation of both MAL/MyD88 and TRAM/TRIF pathways. Recently we identified synthetic monophosphoryl Lipid A (sMLA) as a Trif-biased agonist of TLR4. However, not all TRIF dependent genes were induced optimally after sMLA stimulation. In this study we show that there is a MyD88-dependent inhibition of IFNβ expression after sMLA but not after synthetic diphosphoryl Lipid A (sDLA) stimulation in monocytic dendritic cells (DCs). The IFNβ promoter has binding sites for NF-κB, AP-1 and IRF3, transcription factors activated by IKKα/β, MAPKs (JNK, p38 and ERK) and TBK1, respectively. Functional studies with specific inhibitors showed that JNK and p38 but not ERK1/2 played a major role in IFNβ expression both in MyD88KO and Wild type (WT) DCs. However, phospho-activation of these MAPKs as well as of IKKα/β were decreased in MyD88KO vs. WT DCs after sMLA or sDLA stimulation. However, sMLA but not sDLA induced stronger IRF3 phospho-activation in MyD88KO cells when compared to WT suggesting a MyD88 dependent regulation of TBK1 activation. Preliminary studies indicate differential kinetics in MyD88 recruitment after sMLA stimulation, which may be responsible for this inhibitory effect via an unknown mechanism. These findings identify a novel role for MyD88 in regulation of IFNβ expression based on ligand structure. This work was supported by the NIH/NIAID and the AHA.