The LLC tumor cell line was purchased from the American Type Culture Collection (ATCC® CRL-1642™; Manassas, VA, USA) and maintained in the Department of Immunology, Peking Union Medical College and Chinese Academy of Medical Sciences (Beijing, China). The B16F10 melanoma control cells were provided by Dr Lieping Chin (Department of Immunology, School of Medicine, Yale University, New Haven, CT, USA) (17). These murine tumor cells were maintained in Dulbecco's modified Eagle's medium/F12 medium containing 10% heated-inactivated fetal calf serum (FCS), 2 mM L-glutamine, 10 mM HEPES, penicillin (100 U/ml)-streptomycin (50 µg/ml) solution and 1% sodium pyruvate solution (Gibco Life Technologies, Carlsbad, CA, USA).

Female 6–8 week-old C57BL/6 mice (n=20) were obtained from the Experimental Animal Institute of Peking Union Medical College. Mice were subjected to a 12 h light/dark cycle and maintained in a specific pathogen-free environment at 18–23°C with 40–60% humidity. Food and water were accessible at all times. All protocols involving animals used in the present study were approved by the Animal Research Ethics Committee of the Cancer Institute and Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences (no. 20140002, Beijing, China).

LLC tumor cells (1×10⁶) were resuspended in 100 µl phosphate-buffered saline (PBS) and subcutaneously inoculated into the right flanks of four C57BL/6 mice. These LLC tumor-bearing mice were enrolled in the study when the primary tumor reached 4–8 mm in diameter. The monoclonal antibody (mAb) Gr-1 (RB6–8C5; 200 µg in 100 µl PBS; BioLegend, Inc., San Diego, CA, USA) was intraperitoneally administered on days 5, 10 and 15 to deplete MDSCs. Control LLC tumor-bearing mice (n=1) were treated with immunoglobulin G (rat IgG_2b; BioLegend, Inc.). Tumor dimensions were measured two or three times per week with digital calipers (Cangzhou Yongkang Medical Devices Co., Ltd., Yongkang, China), and the tumor volume was calculated using the formula: Tumor volume (mm³) = 0.52 × (length × width²). Overall animal survival was monitored following conclusion of the treatment. Subsequently, mice were sacrificed 19 days after tumor cell inoculation, using carbon dioxide and tumor tissues were resected. A total of 5 mice were used per group, and all experiments were repeated twice. All experiments were conducted according to animal studies ethics committee guidelines.

Cells from the bone marrow or spleen of LLC tumor-bearing mice were harvested 19 days after tumor cell inoculation, under sterile conditions. Single-cell suspensions were prepared, and red blood cells were removed using Tris lysis buffer (144 mM NH₄Cl, 17 mM Tris-HCl; pH 7.2; Zhongshan Golden Bridge Biotechnology Co., Ltd., Beijing, China). Mononuclear cells isolated from the bone marrow or spleen were incubated at 4°C for 30 min with the following fluorescently conjugated rat mAbs purchased from BioLegend, Inc. [0.2 µg Abs per 10⁶ cells (100 µl)]: Anti-CD45 (allophycocyanin; 30-F11), anti-CD11b [fluorescein isothiocyanate (FITC); M1/70], anti-Gr-1 [phycoerythrin (PE), RB6–8C5], anti-CD8 (PE; 53–6.7) and anti-CD3 (PE/Cy7; 145-2C11). Subsequently, CD45⁺/CD11b⁺/Gr-1⁺ cells (MDSCs) or CD45⁺/CD3⁺/CD8⁺ cells were purified by cell sorting using a FACS Calibur (BD Biosciences, Franklin Lakes, NJ, USA) instrument to >90% purity.

MDSC culture was performed as described previously with modifications (18). Briefly, sorted bone marrow populations (CD11b⁺/Gr-1⁺ and CD3⁺/CD8⁺) were cultured in six-well plates (1×10⁵ cells/well) for 3 days with the addition of recombinant mouse granulocyte-macrophage colony-stimulating factor (GM-CSF; 50 ng/ml; R&D Systems, Minneapolis, MN, USA), prior to harvesting the supernatant for use. CD8⁺ T cells obtained from the spleen were used as effector cells for a cytotoxic assay.

For surface staining, murine leukocytes from heparinized whole tail blood (40 µl) collected on days 8 and 19 post-tumor challenge (n=4) were incubated with the following antibodies: Anti-CD45, anti-CD11b, anti-Gr-1, anti-CD8, anti-CD4 and anti-CD3 (BioLegend, Inc.), at a concentration of 0.2 µg Abs per 10⁶ cells (100 µl) at 4°C for 30 min, according to standard protocols (19). Subsequently, 500 µl OptiLyse® lysis solution (Beckman Coulter, Brea, CA, USA) was added to each tube and incubated at room temperature for 10 min. Following washing with PBS, the samples were subjected to flow cytometric analysis. Additional analysis was performed using FlowJo software (version 7.6.5; Tree Star, Inc., Ashland, OR, USA).

LLC cells were incubated with 5 µM 5,6-carboxyfluorescein diacetate succinimidyl ester (CFSE; Invitrogen Life Technologies, Carlsbad, CA, USA) at 37°C for 10 min in serum-free media, and then washed in RPMI-1640 media containing 10% FCS (Gibco Life Technologies). CFSE-labeled cells were cultured in complete media with the addition of supernatant from MDSCs in 24-well plates (1 ml/well, complete media and supernatant ratio, 1:1). Following 30 h of incubation at 37°C and 5% CO₂, the number of cell divisions undertaken by LLC cells was determined by flow cytometry.

Protein was extracted from LLC tumor tissues from C57BL/6 tumor-bearing mice, as well as LLC and B16F10 tumor cells, and analyzed by western blotting. Briefly, tumor tissues from C57BL/6 mice were flash-frozen in liquid nitrogen immediately following collection, while LLC or B16F10 tumor cells were harvested and washed with PBS. Ground tumor tissues and LLC or B16F10 cells were then lysed in pre-chilled radioimmunoprecipitation assay lysis buffer (Thermo Fisher Scientific, Waltham, MA, USA) containing protease and phosphatase inhibitor cocktails (Roche Diagnostics, New Providence, NJ, USA). Following centrifugation at 10,000 × g for 15 min at 4°C, the supernatant was removed and protein concentration determined using Coomassie brilliant blue G-250 (Sigma-Aldrich, St. Louis, MO, USA). Total protein (20 µg) was prepared for denaturing gel electrophoresis. For western blotting, proteins were separated on 12% SDS-PAGE gels (Bio-Rad Laboratories, Inc., Hercules, CA, USA) and transferred onto Immobilon-FL polyvinylidene difluoride membranes (EMD Millipore, Billerica, MA, USA) using a wet transfer apparatus (Bio-Rad Laboratories, Inc.). Following transfer, the blots were rinsed in washing buffer [Tris-buffered saline with Tween-20 (TBST): 20 mM Tris base, 137 mM NaCl, 0.1% Tween-20, pH 7.6] and blocked in blocking buffer (5% non-fat dry milk in TBST; both from Bio-Rad Laboratories, Inc.) for 1 h at room temperature. Membranes were then washed in TBST and incubated with primary antibodies (1:1,000; Cell Signaling Technology, Inc., Danvers, MA, USA) diluted in TBST with 5% bovine serum albumin overnight at 4°C. Subsequent washes in TBST were followed by incubation with secondary antibodies [goat anti-rabbit IgG-horseradish peroxidase (HRP), cat. no. 7074, or goat anti-mouse IgG-HRP, cat. no. 7076; 1:2,000; Cell Signaling Technology, Inc.] diluted in blocking buffer for 1 h at room temperature. Subsequently, the membranes were washed in TBST and the protein bands were visualized using an enhanced chemiluminescence western blotting system (GE Healthcare Life Sciences, Piscataway, NJ, USA). GAPDH and β-actin were used as internal controls for consistent protein loading. The following primary antibodies were used: Rabbit polyclonal MAGE-A4 antibody (AV54410), rabbit phosphorylated (p)-AKT⁴⁷³ XP® mAb (Ser473; D9E; 4060), rabbit total (t)-AKT mAb (pan; C67E7; 4691), rabbit p-signal transducer and activator of transcription 3 (STAT3)⁷⁰⁵ XP® mAb (Tyr705; D3A7; 9145), rabbit p-STAT3⁷²⁷ mAb (Ser727; 9134), rabbit t-STAT3 mAb (79D7; 4904), mouse p-extracellular signal-regulated kinase (ERK)1/2 mAb [p44/42 MAPK (Erk1/2); Thr202/Tyr204; E10; 9106], rabbit t-ERK1/2 mAb [p44/42 MAPK (Erk1/2); 9102], mouse β-actin mAb (8H10D10; 12262) and rabbit GAPDH mAb (14C10; 2118). MAGE-A4 rabbit polyclonal antibody was purchased from Sigma-Aldrich, while all other primary antibodies mentioned above were obtained from Cell Signaling Technology, Inc. NuPAGE® LDS Sample Buffer (NP0007), NuPAGE® MOPS SDS Running Buffer (NP0001-02) and Novex® Tris-Glycine Transfer Buffer (LC3675) for western blotting were purchased from Invitrogen Life Technologies. The intensities of bands on the membranes were determined using Quantity One software version 4.6.2 (Bio-Rad, Laboratories, Inc., Hercules, CA, USA).

LLC tumor cells (target cells) were labeled with 5 µM CFSE. Sorted CD8⁺ T cells were used as effector cells. Effector cells (4×10⁵ cells/well) were cultured in 48-well plates with 2×10⁴ labeled tumor target cells/well in 500 µl RPMI-1640 containing 10% FCS. Following 4 h of incubation, propidium iodide (PI; 2 µg/ml, Sigma-Aldrich) was added to identify killed target cells. Flow cytometric analysis identified PI⁺CFSE⁺ cells (killed targets) and PI⁻CFSE⁺ cells (viable targets). A total of ~5,000 target cells were acquired. The percentage of specific lysis was calculated in relation to the proportion of basal lysis in untreated cells using the following formula: Percentage of specific lysis = proportion of lysis per sample - proportion of basal lysis / (1 - proportion of basal lysis) × 100.

GraphPad Prism V (GraphPad Software Inc., La Jolla, CA, USA) was used for statistical analyses. Data are presented as the mean ± standard error of the mean, unless indicated otherwise. For comparisons between two groups, statistical analyses were performed using a Student's t-test, and P<0.05 was considered to indicate a statistically significant difference. Survival curves were estimated using the Kaplan-Meier method, and differences among these were evaluated using the log-rank test.

To establish a mouse model of lung cancer, which highly expressed the CT antigen MAGE-A4, MAGE-A4 expression was first detected in LLC tumor cells. LLC tumor cells highly expressed the MAGE-A4 antigen, as did positive control B16F10 cells (Fig. 1A and B). Subsequently, the number of MDSCs present in the blood of tumor-bearing animals at various stages of tumor development was determined by flow cytometry, and an increased MDSC population was identified in advanced tumors (Fig. 1C).

To investigate the impact of MDSCs on MAGE-A4 expression, MDSCs were depleted in LLC tumor-bearing animals. Western blotting revealed that targeted depletion of MDSCs decreased MAGE-A4 expression in tumor-bearing mice. To elucidate the mechanism underlying how MDSCs enhanced MAGE-A4 expression in LLC tumor-bearing mice, p-AKT⁴⁷³, t-AKT, p-STAT3⁷⁰⁵, p-STAT3⁷²⁷, t-STAT3, p-ERK1/2 and t-ERK1/2 expression were evaluated by western blotting. The results presented in Fig. 2A revealed that depletion of MDSCs significantly inhibited MAGE-A4 expression, and that this effect was associated with the p-AKT⁴⁷³ and p-STAT3⁷⁰⁵ pathways.

The survival of mice bearing LLC tumors following depletion of MDSCs was subsequently evaluated. Tumor masses did not generally decrease in size during the treatment period (Fig. 2B). However, depletion of CD11b⁺Gr1⁺ cells conferred a significant survival advantage, with median survival of 68 days, whereas all control mice treated with control isotype IgG succumbed by day 50 (median survival, 41 days; Fig. 2C). Although this result may be interpreted as indicating a lack of efficacy, acute increases in tumor size following immune therapy have frequently been attributed to the recruitment and infiltration of immune cells, rather than the progressive growth of cancer cells (20).

Administration of the mAb Gr-1 markedly reduced the percentage and number of circulating MDSCs in LLC tumor-bearing C57BL/6 mice (Fig. 3A). The percentage of circulating CD4⁺ T cells increased ~2-fold (13.11±2.59 in control mice vs. 23.5±1.3 in RB6–8C5-treated mice; n=4 per group; P=0.053; Fig. 3B); however, this result was not statistically significant. Consistent with this trend, the percentage of circulating CD8⁺ T cells (Fig. 3C) in RB6–8C5-treated mice increased ~2-fold when compared with that of untreated mice, and this result was statistically significant (7.63±1.23 in control mice vs. 17.15±1.25 in RB6–8C5-treated mice; n=4 per group; P=0.032).

To evaluate the effect of MDSCs on the efficacy of the cytotoxic immune response in the LLC animal model, CD8⁺ T cell cytotoxic activity was assessed by flow cytometry. CD8⁺ T cell cytotoxicity towards LLC cells was found to be higher in the MDSC depletion group than that of the control group (Fig. 4A and B). This confirmed that MDSC depletion in tumor-bearing mice resulted in functional activation of CD8⁺ T cells, which, in combination with other factors, results in tumor regression.

These results demonstrated that cancer-conditioned MDSCs may impede the recognition of tumor antigens by CD8⁺ T cells and impair effector T cell activity in tumor-bearing mice. Furthermore, targeted depletion of MDSCs with Gr-1 therapy may enhance the endogenous anti-tumor immune response.

The correlation between MAGE-A4 antigen expression and MDSC infiltration was further assessed by incubating LLC tumor cells with MDSC supernatant. MDSC supernatant significantly enhanced LLC cell proliferation (Fig. 5A and B; P=0.011) and MAGE-A4 expression (Fig. 5C and D).