Nitric oxide synthases (NOS) are important mediators of progrowth signaling in tumor cells, as they regulate angiogenesis, immune response, and immune-mediated wound healing. Ionizing radiation (IR) is also an immune modulator and inducer of wound response. We hypothesized that radiation therapeutic efficacy could be improved by targeting NOS following tumor irradiation. Herein, we show enhanced radiation-induced (10 Gy) tumor growth delay in a syngeneic model (C3H) but not immunosuppressed (Nu/Nu) squamous cell carcinoma tumor-bearing mice treated post-IR with the constitutive NOS inhibitor NG-nitro-l-arginine methyl ester (L-NAME). These results suggest a requirement of T cells for improved radiation tumor response. In support of this observation, tumor irradiation induced a rapid increase in the immunosuppressive Th2 cytokine IL10, which was abated by post-IR administration of L-NAME. In vivo suppression of IL10 using an antisense IL10 morpholino also extended the tumor growth delay induced by radiation in a manner similar to L-NAME. Further examination of this mechanism in cultured Jurkat T cells revealed L-NAME suppression of IR-induced IL10 expression, which reaccumulated in the presence of exogenous NO donor. In addition to L-NAME, the guanylyl cyclase inhibitors ODQ and thrombospondin-1 also abated IR-induced IL10 expression in Jurkat T cells and ANA-1 macrophages, which further suggests that the immunosuppressive effects involve eNOS. Moreover, cytotoxic Th1 cytokines, including IL2, IL12p40, and IFNγ, as well as activated CD8+ T cells were elevated in tumors receiving post-IR L-NAME. Together, these results suggest that post-IR NOS inhibition improves radiation tumor response via Th1 immune polarization within the tumor microenvironment. Cancer Res; 75(14); 2788–99. ©2015 AACR.
Radiotherapy remains a primary mode of treatment for more than 50% of cancer patients in North America (1). At the molecular level, ionizing radiation (IR) exerts its antitumor effects by inducing direct DNA damage in the form of DNA double-strand breaks as well as indirect damage by the generation of reactive oxygen species (2). Although DNA damage has a central role in radiation-induced tumor cell death, it does not fully account for tumor response to local radiation. In addition to stimulation of DNA repair, IR induces multiple cellular signaling pathways. Importantly, cell survival depends upon the ratio of activated pro- and antiproliferative pathways, suggesting that irradiated cells, which evade death, survive and progress to more aggressive and therapeutically resistant tumors (3). Radiation-induced signaling pathways associated with cancer progression include elevated epidermal growth factor receptor, hypoxia inducible factor-1 (HIF1), upregulation, and/or activation of matrix metalloproteinases (MMP), and overexpression of cytokines, including vascular endothelial growth factor (VEGF) and other immunosuppressive mediators that promote cancer survival, invasion, and metastasis (4). Thus, the biology of sublethally irradiated tumor cells favor survival, invasion, and angiogenesis, suggesting that therapeutic efficacy could be improved by combining radiation treatment with agents that target these or other progrowth pathways induced by radiation (5).
Nitric oxide (NO) is an important mediator of many progrowth signaling cascades in cancer (6–9). Nitric oxide synthases (NOS) catalyze the production of NO by the five-electron oxidation of a guanidino nitrogen atom of the substrate l-arginine, which requires NADPH, FAD, FMN, heme, and O2 as cofactors (10). Three NOS isoforms are known to exist; neuronal NOS (nNOS or NOS1), inducible NOS (iNOS or NOS2), and endothelial NOS (eNOS or NOS3). NO has many diverse roles in normal physiology and tumor biology, which are spatially, temporally, and concentration dependent. The constitutive isoforms eNOS and nNOS are tightly regulated by Ca2+/calmodulin, and produce low flux (pmol/L) NO over short periods of time. In contrast, the inducible isoform iNOS is Ca2+ independent and generates higher flux NO over a longer period of time that can range from nmol/L to μmol/L in concentration, depending upon the stimulant (11). NOS has been studied extensively in carcinogenesis. Although elevated NOS3 expression has a role in tumor angiogenesis, increased NOS2 expression predicts poor therapeutic response, tumor progression, and decreased patient survival (9, 12–15). To date, our molecular signatures suggest that NO-mediated prosurvival, cell migration, angiogenesis, and stem cell marker (i.e., ERK, Akt, IL8, IL6, S100A8, CD44) signaling in tumors and tumor cells occurs at ≥400 nmol/L steady-state NO (6, 9). Together, these observations suggest that the NOS enzymes are exploitable therapeutic targets.
NO produced by the constitutive eNOS isoform controls blood flow and is a key mediator of the pro-angiogenic effects of vascular endothelial growth factor (VEGF; ref. 16). A clinical study demonstrated reduced tumor blood volume within 1 hour of administration of the competitive NOS inhibitor nitro-l-arginine (L-NNA), which lasted for 24 hours in all patients studied (17). Side effects of NOS inhibition included bradycardia and hypertension, which were not study limiting and suggest that NOS inhibition may be a beneficial therapeutic option for combined modalities (17). Advances in radiotherapy have included the co-administration of anti-angiogenic (AA) drugs, which radiosensitize endothelial cells (18). IR also activates constitutive NOS, as well as ERK1/2 kinase prosurvival signaling, both of which were blocked by L-NNA (19). In addition, the administration of L-NNA 24 hours prior to 10 Gy irradiation of tumor-bearing mice demonstrated reduced tumor blood flow and increased tumor cell apoptosis when compared to mice receiving radiation or L-NNA alone, further supporting NOS inhibition as a target to improve radiation therapeutic response (20).
In addition to targeting tumor vasculature, IR modulates host immunity and mimics vaccine response by enhancing the release of damage-associated molecular patterns (DAMP) from dying cells, which then activate cytotoxic lymphocytes (CTL) through toll-like receptor activation (21). IR facilitates antigen-presenting cell and T-cell penetration into the tumor (22), and also impacts host immunity through modulation of both pro- and antitumor responses depending upon the Th1 (cytotoxic) versus Th2 (immunosuppressive) cytokine milieu and associated immune cell mediators (21). Macrophages exposed to Th1 cytokines exhibit increased levels of pro-inflammatory cytokine production, antigen presentation, and cytotoxic activity. In contrast, macrophages exposed to Th2 cytokines exhibit an immunosuppressive phenotype associated with blocked CTL activity, increased angiogenesis, tissue restoration, and wound healing response (23). T-regulatory cells (Tregs) are pivotal mediators of immune suppression and the development of immunologic tolerance through their ability to limit antitumor immune responses (24). Tregs mediate tumor immune tolerance in part through the secretion of IL10, TGFβ, or IL35 immunosuppressive molecules (24). Indeed, IL10 and TGFβ derived from Tregs promote tumor progression through antitumor immune suppression (25). NO also mediates both cGMP-dependent and -independent Th1–Th2 immune transition and may be important in the tumor response to radiation-induced injury (26, 27). To explore this hypothesis, we examined cytokine expression profiles and T-cell activation during radiation-induced tumor growth delay in a syngeneic model treated post-IR with the NOS inhibitor NG-nitro-l-arginine methyl ester (L-NAME).
Materials and Methods
Jurkat cells (Jurkat clone E6-1) were obtained from the ATCC and maintained in 5% CO2, RPMI-1640 culture medium with 10% FBS and 100 IU/mL Pen Strep antibiotics (Life Technologies). Jurkat cells were treated with IR (0, 1, or 5 Gy) in the presence or absence of various concentrations of DETA/NO (0, 30, 60, 100, 300, 500, and 100 μmol/L), L-NAME (0, 500, or 1,000 μmol/L), or the guanylyl cyclase inhibitor ODQ (10 μmol/L) or thrombospondin-1 (TSP-1: 1 μg/mL). The NONOate donors, including DETA/NO, are stable when maintained in basic conditions (10 mmol/L NaOH) but release NO at defined rates at physiologic pH (28); 10 mmol/L NaOH served as vehicle control in experiments utilizing the NO donor DETA/NO. The ANA-1 macrophage cell line used in this study was established by immortalization of bone marrow macrophages from C57BL/6 mice with J2 recombinant retrovirus-expressing v-myc/v-raf oncogenes (29). ANA-1 cells were grown in DMEM supplemented with 10% FBS and 1% penicillin–streptomycin, plated at a density of 5 × 105 per well in a 12-well plate and grown overnight.
Suppression of IL10
Silencing of IL10 protein translation was accomplished by using an antisense 25-mer oligo (Gene Tools) designed specifically to block the AUG translational start site of mouse IL10 (GenBank accession no. NM_010548: oligo sequence, 5-AGCTCTCTTTTCTGCAAGGCTGCTT). This oligo complements the sequence from −31 to −6 relative to the initiation codon. Suppression of secreted IL10 protein levels were verified in lipopolysaccharide (LPS)-stimulated Raw 267.4 (30) cells pretreated with control or IL10 morpholino. Cell culture media was collected at 24 and 48 hours and IL10 protein levels were measured by ELISA assay (R&D Systems) according to the manufacturer's recommendations.
In vivo mouse tumor model
The Animal Care and Use Committee (National Cancer Institute, NIH, Bethesda, MD) approved mouse protocols. Female C3H/Hen or athymic nude mice were supplied by the Frederick Cancer Research and Development Center Animal Production Area (Frederick, MD). The animals were received at 6 weeks of age, housed five per cage, and given autoclaved food and water ad libitum. Experiments were performed at 9 to 10 weeks of age and in accordance with principles outlined in the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Research Council, Washington, DC). Squamous cell carcinoma VII/SF tumor cells (SCC) were derived from spontaneous abdominal wall squamous cell cancer (obtained from Dr. T. Phillips, UCSF, San Francisco, CA) and propagated in C3H/Hen mice (31). For growth delay studies, 2 × 105 viable SCC cells were injected into the subcutaneous space of the right hind leg of 8-week-old C3H/Hen or nude mice and grown for 1 week when tumor size reached approximately 200 mm3 in size. Similarly, human colon carcinoma HT29 tumor xenografts were grown in nude mice injected with 1 × 106 cells. C57BL/6 WT or eNOS−/− (The Jackson Laboratory Stock No. 002684) mice on the same background were injected with 1 × 106 B16 melanoma cells. SNP analysis (DartMouse, The Geisel School of Medicine at Dartmouth, Dartmouth, NH) demonstrated background purities of C57BL/6 WT and eNOS−/− mice to be 99.8% and 98.9%, respectively, when compared to the in-house control. Tumor volume was measured by caliper and calculated as mm3 = [width2 × length]/2 where width was the smaller dimension. Tumor irradiation was accomplished by securing each animal in a specially designed Lucite jig fitted with lead shielding that protected the body from radiation while allowing exposure of the tumor-bearing leg. A Therapax DXT300 X-ray irradiator (Pantak, Inc.) using 2.0 mm A1 filtration (300 KVp) at a dose rate of 2.53 Gy/min was used as the X-ray source. Irradiated tumors received one 10 Gy dose. Designated groups of animals were treated with NOS inhibitor L-NAME or IL10 suppressing agents. NOS inhibition was achieved by administering L-NAME post-IR in the drinking water at a concentration of 0.5 g/L for the duration of the experiment (20). IL10 protein levels were suppressed using an IL10 morpholino; mice were injected with a 750 μL volume of 10 μmol/L IL10 morpholino (Gene Tools) or a four base-mismatched control morpholino in saline 48 hours prior to irradiation. After irradiation, the mice were returned to their cages, and tumors were measured three times each week thereafter to assess tumor growth. Animals were euthanized when tumor growth approached the maximum allowable limit.
Control and irradiated tumors (±L-NAME) were collected at 0, 0.25, 1, 2, 3, 4, and 7 days post-irradiation. Cytokine protein expression was evaluated by Q-Plex multiplex ELISA arrays (QUANSYS Biosciences).
Isolation of leukocytes from spleen
Spleens were harvested from tumor-bearing animals, placed in sterile saline, and filtered through a two-chamber sterile Filtra-Bag (Fisher Scientific). Splenocytes were counted by Sysmex KX-21 (Roche Diagnostics).
Isolation of tumor-infiltrating leukocytes
Tumors were dissected and filtered through a two-chamber sterile Filtra-Bag (Fisher Scientific), then digested in RPMI containing 5% fetal calf serum, 700 units/mL collagenase (Invitrogen), 100 μg/mL DNAse I (Boehringer Mannheim), and 1 mmol/L EDTA (pH 8.0), at 37°C for 45 minutes. The homogenate was then processed in a tissue stomacher-80 (Seward) for 30 seconds, washed with HBSS (BioWhittaker), and resuspended in 40% Percoll (Amersham Pharmacia) in DMEM medium (BioWhittaker). The suspension was underlaid with 80% Percoll and centrifuged for 25 minutes at 1,000 × g. Leukocytes were collected from the interphase, washed, and counted.
Cells (1 × 106) were incubated in cell staining buffer (0.1% BSA, 0.1% sodium azide) containing 250 μg/mL 2.4G2 ascites, which blocks nonspecific Fc receptor antibody binding for 15 minutes. Cells were stained with fluorescently conjugated antibodies (BD Pharmingen) for 20 minutes. Labeled cells were washed twice in cell staining buffer and analyzed on a BD Facs Canto II flow cytometer (Becton Dickinson).
RNA extraction, reverse transcription, and quantitative real-time PCR
Total RNA was extracted with TRizol (Invitrogen) according to the manufacturer's protocol. RNA samples were reverse transcribed into cDNA using Sprint RT Complete 8-well strips (Clontech) according to the manufacturer's recommendation. Primer pairs were designed for IL10 that recognized F: TTAAGGGTTACCTGGGTTGC and R: GCCTGAGGGTCTTCAGGTTC sequences using the IL10 gene Ref Seq sequences by Primer3 (32). All quantitative real-time PCR reactions were designed to follow a universal real time PCR condition: 94°C 2 minutes, 45 cycles of 94°C 30 seconds, and 60°C 30 seconds in PowerSYBR Master Mix (Applied Biosystems). Amplicon specificity was checked by BLAST search and on a 2% FlashGel (Lonza) to ensure that neither nonspecific amplicons nor primer–dimer complexes were formed. Relative expression was calculated using the ddCt formula. Amplification efficiencies for primers were checked against the 18S housekeeping gene to ensure signals were detected in PCR exponential phase and that primers had similar amplification efficiencies.
Jurkat and CD47-deficient JinB8 Jurkat cells (∼1 × 106; ref. 33) were plated on 12-well plates (Corning) using RPM1 medium + 2% FBS at 37°C in 5% CO2 and treated with 1 μg/mL TSP-1 for 6 hours. Untreated cells were used as controls. Total RNA was extracted using TriPure Isolation Reagent (Roche). The first strand cDNA was made using Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Thermo Scientific). Primer sequences for hypoxanthine phosphoribosyltransferase 1 (HPRT1; 5′-ATT GTA ATG ACC AGT CAA CAG GG-3′/5′-GCA TTG TTT TGC CAG TGT CAA-3′) and IL10 (5′-AAA TTA GCC GGG CAT GGT GG-3′/5′-CTG CAA CTT CCA TCT CCT GGG T-3′) were used for real-time PCR performed using SYBR Green (Roche) on an MJ Research Opticon I instrument (Bio-Rad) with the following amplification program: 95°C for 15 minutes, followed by 40 cycles of 95°C for 15 seconds, 58°C for 20 seconds, 72°C for 25 seconds, and 72°C for 1 minute. Melting curves were performed for each product from 30°C to 95°C, reading every 0.5°C with a 6-second dwell time. Fold change in mRNA expression was calculated by normalizing to HPRT1 mRNA level. Two-factor ANOVA with replication was used for statistics analysis.
Automated capillary Western blot
Western blots were performed using WES, an automated capillary-based size sorting system (ProteinSimple; ref. 34). All procedures were performed with manufacturer's reagents according to their user manual. Briefly, 8 μL of diluted protein lysate was mixed with 2 μL of 5× fluorescent master mix and heated at 95°C for 5 minutes. The samples (1 μg), blocking reagent, wash buffer, primary antibodies, secondary antibodies, and chemiluminescent substrate were dispensed into designated wells in a manufacturer provided microplate. The plate was loaded into the instrument and protein was drawn into individual capillaries on a 25 capillary cassette provided by the manufacturer. Protein separation and immunodetection was performed automatically on the individual capillaries using default settings. The data were analyzed using Compass software (ProteinSimple; ref. 34). Primary antibodies used were nNOS, eNOS (Cell Signaling), and iNOS (Santa Cruz Biotech) HPRT was used as loading control (Santa Cruz Biotech).
All results are expressed as the mean ± SEM. The differences in means of groups were determined by the Student t test with the minimum level of significance set at P ≤ 0.05.
NOS inhibition enhances radiation-induced tumor growth delay in syngeneic mice
Tumor growth delay is described by the substance enhancement ratio (SER), which is the ratio of time required for treated versus control tumors to reach a defined size (1,000 mm3). The effect of NOS inhibition by L-NAME, a NOS inhibitor that is more selective for the constitutive isoforms (eNOS and nNOS; ref. 35), on radiation-induced tumor growth delay was examined in a syngeneic murine model of SCC tumor-bearing C3H mice. L-NAME was administered in the animals' drinking water (0.5 g/L) following tumor irradiation (post-IR, 10 Gy). Because NO regulates vascular tonicity, we chose post-IR administration of L-NAME to minimize vascular constriction and maintain tumor pO2 prior to and during tumor irradiation while targeting vascular constriction post-IR. Figure 1A demonstrates enhanced radiation-induced tumor growth delay in mice that received post-IR L-NAME (SER ∼3.3) when compared to tumors treated with 10 Gy IR alone (SER ∼1.8). The iNOS-specific inhibitor aminoguanidine was also tested, and yielded an SER of 2.1. Interestingly, L-NAME–mediated NOS inhibition had no effect on the radiation-induced tumor growth delay of HT29 human adenocarcinoma cells or SCC xenografts in immunosuppressed nude mice lacking T cells (Fig. 1B and C, respectively). These results indicate a requirement of T cells for L-NAME potentiation of radiation-induced tumor growth delay.
Effect of L-NAME on radiation-induced cytokine expression in syngeneic mice
T cells are lymphocytes that direct cell-mediated immunity and are distinguished from other lymphocytes by the presence of cell surface T-cell receptors. There are several subsets of T cells, each with a distinct function; proliferating helper T cells differentiate into two major types of effector T cells known as Th1 and Th2 cells, which secrete specific cytokines that mediate different immune responses. Th1 cells are proinflammatory, mediate host immunity to foreign pathogens, and are induced by IL2, IL12, and their effector cytokine IFNγ (36). In contrast, Th2 cells are immunosuppressive, secrete IL4, IL5, IL10 as well as TGFβ, and mediate wound resolution following pro-inflammatory assault (37). To explore a potential role for NOS-derived NO during radiation-induced T-cell response, QPlex was used to examine alterations in tumor cytokine expression. Supplementary Table S1 summarizes the impact of NOS inhibition by L-NAME, on the trend of Th1 versus Th2 cytokine protein expression induced by 10 Gy tumor irradiation, whereas Supplementary Table S2 summarizes pg/mg cytokine levels as well as P-values and fold-change, as a function of time after irradiation. The trend of Th1 versus Th2 cytokine protein expression summarized in Supplementary Table S1 suggests that tumors receiving radiation alone rapidly acquire (within 24 hours) an overall Th2 signaling profile as defined by early elevation of IL10 (Day 1) followed by increased IL5, IL3, and IL4 tumor expression (Day 2–4). In contrast, tumors from animals that received post-IR L-NAME exhibited a Th1 profile as defined by elevated IL2 (6 hours, Day 1), IL12, and IFNγ (Day 3, 4, 7) tumor expression. The most profound observation pertained to the dramatic early induction of IL10 24-hour post-IR, which was abated by L-NAME (Fig. 2A). These results suggest the rapid induction of an IL10-mediated immunosuppressive phenotype in response to tumor irradiation, which was abolished by post-IR NOS inhibition. We also examined tumor NOS isoform protein expression 24-hour post-IR. When compared to control, Fig. 2B shows increased iNOS protein expression in the 10 and 10 Gy + L-NAME tumors. This is an interesting observation considering that constitutive NOS inhibition by L-NAME was more effective in extending the radiation-induced tumor growth delay. This may be explained by the findings of Connelly and colleagues (38) who demonstrated that eNOS is required for the full activation of iNOS.
NO–induced IL10 expression in Jurkat T cells and ANA-1 macrophages
IL10 is generally produced by differentiated monocytes and lymphocytes (i.e., macrophages and T cells, respectively). To further examine the involvement of NO during radiation-induced IL10 expression, we used Jurkat cells, which are T lymphocytes that express IL10 and are commonly employed to study T-cell signaling. Cytokine expression profiles were examined in cells exposed to the slow releasing NO donor DETA/NO, which mimics NO flux under inflammatory conditions. Figure 3 demonstrates NO concentration-dependent induction of IL10 mRNA (A) and protein (B) in Jurkat cells, which peaked at 300 μmol/L DETA/NO. Next, Jurkat cells were exposed to 1 Gy irradiation, then treated with or without L-NAME, and incubated overnight to mimic the tumor xenograft irradiation protocol. Figure 3C shows a greater than four-fold increase in Jurkat IL10 expression 24 hours after 1 Gy irradiation, which was abated by L-NAME and is similar to the L-NAME effect on radiation-induced tumor IL10 expression shown in Fig. 2. Interestingly, the L-NAME suppressed IL10 levels re-accumulated to that induced by 1 Gy irradiation in the presence of the exogenous NO donor DETA/NO at concentrations of 100 to 500 μmol/L or ∼400 nmol/L steady-state NO (Fig. 3C; refs. 6, 7, and 12). To date, our breast cancer biomarker signatures suggest that NO-mediated prosurvival, cell migration, angiogenesis, and stem cell marker (i.e., ERK, Akt, IL8, IL6, S100A8, CD44) signaling in tumors and tumor cells occurs at ≥400 nmol/L steady-state NO (6, 7, 9, 12, 39). When considering this molecular signature, the results shown in Fig. 3A–C are consistent with our earlier reports and suggest that ∼400 nmol/L steady-state NO modulates radiation-induced IL10 expression in Jurkat cells. Also, the NO flux-dependent regulatory trend of IL10 shown in Fig. 3C resembles a bell-shaped curve, which is consistent with low flux NO regulation of wound response vs. high flux NO-mediated toxicity (11, 40–42).
L-NAME is more selective for the constitutive NOS isoforms, which implicates possible eNOS/cGMP-dependent signaling (35, 43). To explore the potential of cGMP-dependent signaling during radiation-induced IL10 expression, Jurkat cells were exposed to 1 Gy IR ± the guanylyl cyclase inhibitor ODQ, which completely abolished IL10 expression induced by 1 Gy IR (Fig. 3D). TSP-1 inhibits NO signaling through its receptor CD47 by inhibiting eNOS activation and abating NO-dependent cGMP synthesis and cGMP-dependent protein kinase signaling in vascular cells and Jurkat T cells (41, 42). Exogenous TSP-1 also blocked radiation-induced Jurkat IL10 expression, suggesting eNOS/cGMP-dependence for this process (Fig. 3D). Inhibition of IL10 expression by TSP-1 is CD47-dependent because inhibition of basal IL10 mRNA expression was lost in the CD47-deficient Jurkat mutant JinB8 (Fig. 3E). The 2-fold stimulation of IL10 mRNA by TSP-1 in the CD47 mutant is consistent with reported positive effects of TSP-1 on IL10 expression mediated by the TSP-1 receptor CD36 (44). Radiation also induced IL10 in murine ANA-1 macrophages, which was abated by L-NAME, ODQ and TSP-1, indicating that cGMP-dependent regulation of IL10 is not restricted to T cells (Fig. 3F). Collectively, these results indicate that radiation-induced IL10 expression in T cells is tightly controlled by low constitutive NOS-derived NO flux.
IL10 suppression enhances radiation-induced tumor growth delay
Post-IR NOS inhibition by L-NAME enhanced radiation-induced tumor growth delay and abated radiation-induced IL10 expression (Figs. 1A and 2, respectively). To examine a role of IL10 in the recovery from post-IR tumor growth delay, IL10 protein translation was suppressed by treatment with an IL10 morpholino (45, 46). Confirmation of the morpholino efficacy for IL10 protein suppression was verified using LPS-stimulated Raw 267.4 cells because LPS is a strong inducer of IL10 in these cells (30) as shown in Fig. 4A. Next, tumor-bearing animals were treated with IL10 or control morpholino 48 hours prior to tumor irradiation. IL10 morpholino treatment enhanced the radiation-induced tumor growth delay (SER ∼2.7) as shown in Fig. 4B in a manner similar to that observed by L-NAME (Fig. 1A) but had no effect on tumor growth in the absence of radiation. These results suggest that radiation-induced tumor growth delay can be improved by inhibiting IL10-mediated immunosuppressive signaling in the C3H/SCC syngeneic model.
L-NAME increases tumor-associated CD8+ cytolytic T cells post-IR
Cytotoxic T lymphocytes are a subgroup of T cells that when activated kill invading pathogens and tumor cells. These cells are commonly referred to as CD8+ T cells because they express cell surface CD8 glycoprotein. Importantly, immunosuppressive molecules, including IL10, can inactivate CD8+ T cells. To identify the presence of CD8+ T cells, markers of tumor lymphocyte infiltration were examined in control and 10 Gy ± L-NAME tumors, as well as spleen from tumor-bearing mice. Figure 5A shows increased tumor-associated CD8+ T cells in irradiated tumors treated with L-NAME but not spleen taken from the same animals (Fig. 5B). CD69 is a marker of T-cell activation. Figure 5C demonstrates increased CD8+ CD69+ mean fluorescence intensity in infiltrating lymphocytes from irradiated tumors treated with L-NAME but not in spleen taken from the same animals (Fig. 5D). Importantly, these results implicate a localized tumor response culminating in the elevation of activated cytotoxic T cells in post-IR NOS inhibited tumors. Neutrophils, dendritic cells, and immature myeloid cells from post-IR treated L-NAME tumors were elevated on day 3, when compared to irradiated tumors (Fig. 5E–G). In contrast, Tregs and natural killer cells did not change (Fig. 5H and I). Collectively, these results demonstrate that altered NO flux via L-NAME–mediated NOS inhibition can improve the efficacy of therapeutic radiation by immune polarization favoring a proinflammatory phenotype within the tumor microenvironment in an SCC/C3H syngeneic model.
L-NAME is more selective for inhibiting the constitutive NOS isoforms (eNOS and nNOS) and our cell culture results implicate eNOS/cGMP-dependent signaling (Fig. 3) in the immune polarization shown in Fig. 5. To further explore a role of eNOS in the potentiation of radiation therapeutic efficacy, we examined the radiation-induced tumor growth delay of murine B16 melanoma xenografts in wild-type (WT) and eNOS knockout (eNOS−/−) mice on the C57BL/6 background (Fig. 6A). When compared to control, the irradiated tumor in WT mice yielded an SER of ∼2 (Fig. 6A), which is consistent with the SCC/C3H syngeneic model shown in Fig. 1A. Remarkably, the irradiated tumor in eNOS−/− animals exhibited an SER of ∼5.5 when compared to the WT-irradiated tumor (Fig. 6A). Interestingly, IL10 protein was not detected in B16 xenografts grown in WT C57BL/6 or eNOS−/− on the same background under these conditions; however, the cytokine protein expression profile of irradiated tumors in eNOS−/− mice exhibited significantly elevated Th1 cytokines, including IL2, TNFα, and IFNγ, as shown in Fig. 6B–D. Importantly, radiation has been shown to induce eNOS expression, which promotes tumor recovery from radiation injury (47). Together, these results suggest that eNOS has a vital role in acute tumor radiation response and that improved pharmacology of NOS inhibitors in combination with radiation may be therapeutically beneficial.
The role of NO flux within the tumor microenvironment as it relates to therapeutic efficacy is complex. Studies have shown that steady-state NO modulation within the tumor microenvironment leads to improved radiation therapeutic efficacy (20, 47, 48). NO mediates numerous effects at the cellular and physiological levels. Similar to molecular O2, the outer π orbital of NO has an unpaired electron, which imparts a high affinity for other radicals, including radiation-induced carbon radicals on DNA, which leads to fixed DNA damage and enhanced NO-mediated radiosensitivity (49). In support of this early observation, a recent study demonstrated that the presence of low steady-state NO dramatically enhanced radiosensitization as well as increased the time of DNA repair, when compared to anoxic and aerated control tumors (50). Similarly, site-specific iNOS transgene expression driven by the radiation-inducible pE9 promoter demonstrated enhanced tumor radiation response under hypoxic conditions (51). In addition to these direct effects of NO-mediated radiosensitization, altered NO gradients prior to tumor irradiation have been shown to normalize tumor vasculature, which increased tumor oxygen tension and tumor response to radiation (52). In contrast, administration of the NOS inhibitor L-NAME before and after radiation minimized the cytotoxic effect of NO under conditions of hypoxia (48). In this context, NO improved radiation therapeutic efficacy by enhanced tumor perfusion and oxygen effect (53). Thus, NO modulation prior to and at the time of radiation is therapeutically beneficial. Together, these studies demonstrate the contextual dependence of timing and distinct mechanisms directed by NO flux for improved tumor response to radiotherapy.
Although the modulation of tumor NO flux prior to irradiation improves tumor oxygenation and radiation efficacy, NO also promotes angiogenesis in the context of immune-mediated wound response (40–42), which may facilitate post-irradiation recovery of a sublethally irradiated tumor. Indeed, macrophages employ NO generated by both eNOS and iNOS during wound response (40, 54) and in vivo models have shown delayed wound closure in iNOS knockout mice (55). Toward this end, IR-induced angiogenesis (56) through NO signaling (47), which promoted tumor recovery following radiation injury. These observations suggest that post-irradiation inhibition of angiogenesis may be beneficial. Thus, we hypothesized that improved radiation therapeutic efficacy and extended tumor growth delay may be achievable by targeting NO flux through NOS inhibition following tumor irradiation.
Interestingly, post-IR administration of the constitutive NOS inhibitor L-NAME extended radiation-induced tumor growth delay and was more effective than the selective iNOS inhibitor aminoguanidine (Fig. 1A). Moreover, L-NAME extended the radiation-induced tumor growth delay only in syngeneic mice but not nude mice. This observation implicates the involvement of innate immunity and cytotoxic T cells in enhanced radiosensitivity, which is regulated by NO flux, and further supported by the cytokine expression profile of post-IR NOS-inhibited tumors that expressed high levels of cytotoxic Th1 cytokines, including IL2, IFNγ, and IL12p40, as summarized in Fig. 7. In contrast, tumors receiving radiation alone exhibited immunosuppressive Th2 signaling, as indicated by increased IL10, IL5, and IL4 cytokine expression (Supplementary Tables S1 and S2). Moreover, tumor cytokine expression analysis revealed enhanced IL10 protein levels 24 hours following tumor irradiation in SCC-tumor bearing C3H mice, which was abolished by L-NAME (Fig. 2) and confirmed in irradiated Jurkat T lymphocytes and ANA-1 macrophages (Fig. 3). Importantly, in vivo IL10 protein suppression extended radiation-induced tumor growth delay in C3H mice in a manner similar to that of L-NAME. These findings implicate a novel role for NO as a stimulator of IL10-mediated tumor immunosuppressive signaling, which accelerates tumor recovery and regrowth in response to radiation injury in the C3H model.
Cytokine expression analysis of ANA-1 macrophages, and Jurkat T cells demonstrated increased IL10 expression 24 hours after 1 Gy irradiation, which was abated by L-NAME, suggesting that radiation-induced IL10 could come from these cell types. Although we used a variety of sensitive detection methods, including flow cytometry analysis of IL10 associated with markers of specific immune cell populations, as well as flow cytometry analysis of GFP-IL10–tagged mice, we were unable to confirm the specific cellular source of IL10 in our experiments. In addition, no significant changes in Treg cell populations were observed that might account for the cellular source of increased IL10 levels following irradiation. Ongoing studies are aimed at identifying the relative contributions of leukocyte subsets to IL10 following tumor irradiation.
Flow cytometry analysis of immune cell mediators demonstrated increased CD8+ expression and CD8+/CD69+ MFI (indicative of cytotoxic CD8+ T-cell activation) in post-IR NOS-inhibited C3H tumor xenografts but not spleen (Fig. 5A–D), implicating a localized immune response at the irradiated tumor site. The results herein further support a key role for NO flux-dependent regulation of IFNγ and cytotoxic T-cell activation for improved radiation therapeutic efficacy (Supplementary Tables S1 and S2; Figs. 5 and 6). Indeed, cytotoxic CD8+ T cells mediate cell killing through increased IFNγ (57), and inhibition of either IFNγ or CD8+ T cells abolished the therapeutic efficacy of radiation in colon adenocarcinoma tumor-bearing mice (58). In addition, our results indicate that L-NAME potentiation of radiation treatment efficacy is eNOS/cGMP-dependent. Suppression of eNOS/cGMP-dependent signaling by the guanylyl cyclase inhibitors ODQ or TSP-1 abolished radiation-induced IL10 expression in Jurkat T lymphocytes and ANA-1 macrophages (Fig. 3). Also, radiation-induced tumor growth delay was dramatically enhanced in eNOS−/− tumor xenografts, which exhibited increased Th1 cytokine expression (Fig. 6). Thus, radiation-induced tumor injury promotes a Th2 immunosuppressive profile that is eNOS/cGMP-dependent and involves low NO flux.
The post-IR NOS-inhibited tumor exhibited enhanced expression of IL2, IFNγ, and IL12p40 Th1 mediators (Supplementary Table S1). IL2 is a pleiotropic cytokine that has pivotal roles during immune regulation in response to foreign pathogens (59). IL2 is produced primarily by CD4+ T cells and promotes the differentiation, expansion, and cytolytic activation of cytotoxic T cells. Importantly, IL2 effects are receptor mediated; IL2 interaction with IL12Rβ2 leads to upregulation of IFNγ and IL12 during Th1-cell differentiation (59). Interestingly, a cGMP-dependent role of low flux NO in the selective upregulation of IL12Rβ2 has been reported (26). IL12 is also important for sustaining memory/effector T cells, which promote long-term protection against pathogens and tumors. IL2 also interacts with IL2Rα to promote CD8+ T-cell differentiation and activation (59). Importantly, these cytokine activation profiles are consistent with the time course analysis showing elevated IL2, IFNγ, and IL12p40 in the post-IR NOS-inhibited tumors summarized in Supplementary Table S1, as well as CD8+ T-cell regulation shown in Fig. 5. In contrast, tumors receiving irradiation alone demonstrated increased IL10 followed by elevated Th2 mediators IL5, IL3, and IL4. Interestingly, IL2 was also elevated in these irradiated tumors and may have played a role in the upregulation of IL4 and IL5 Th2 cell differentiation, which is IL4Rα dependent (59). The proinflammatory cytokine IL1α was also observed in the irradiated tumor. Despite its proinflammatory status, IL1α released by solid tumors acts as a chemoattractant to facilitate malignancy-associated inflammatory responses (60).
Collectively, the results herein suggest a novel mechanism of low flux NO during Th1–Th2 transition, tumor immunosuppressive signaling, and accelerated wound recovery in the tumor response to ionizing irradiation. Importantly, CD8+ T-cell regulation and IFNγ expression seem to be determined by NO flux-dependent IL10 versus IL2 signaling cascades, which can be modulated by pharmacological NOS inhibition and may provide a novel immunotherapeutic approach for improved radiation therapeutic efficacy.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Conception and design: L.A. Ridnour, R.Y.S. Cheng, D.D. Roberts, H.A. Young, J.B. Mitchell, G. Trinchieri, D.A. Wink
Development of methodology: R.Y.S. Cheng, J.M. Weiss, D.R. Soto-Pantoja, A.H. Kesarwala, D.A. Wink
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L.A. Ridnour, R.Y.S. Cheng, J.M. Weiss, S. Kaur, D.R. Soto-Pantoja, D. Basudhar, J.L. Heinecke, W. DeGraff, A.L. Sowers, J.B. Mitchell
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): L.A. Ridnour, R.Y.S. Cheng, J.M. Weiss, S. Kaur, D.R. Soto-Pantoja, D. Basudhar, C.A. Stewart, D.A. Wink
Writing, review, and/or revision of the manuscript: L.A. Ridnour, R.Y.S. Cheng, D. Basudhar, J.L. Heinecke, A.H. Kesarwala, D.D. Roberts, H.A. Young, G. Trinchieri, R.H. Wiltrout, D.A. Wink
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): W. DeGraff
Study supervision: D.D. Roberts
Other (performed animal experiments): A. Thetford
This research was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, and Center for Cancer Research (D.A. Wink, D.D. Roberts, H.A. Young, J.B. Mitchell, G. Trinchieri, and R.H. Wiltrout).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.