C57BL/6 (B6) mice were obtained from Beijing University Experimental Animal Center (Beijing, China). Mice were maintained in a specific pathogen-free facility. All experimental manipulations were undertaken in accordance with the Institutional Guidelines for the Care and Use of Laboratory Animals, Institute of Zoology (Beijing, China) and were approved by the Committee for Animal Care and Use in Institute of Zoology.

Anti-mCD11b-PE-Cy5 and anti-mCD206-PE were purchased from BD Biosciences PharMingen (San Diego, CA, USA). Anti-F4/80-FITC was procured from Tianjin Sungene Biotech (Tianjin, China). Recombinant mouse IL-4 was purchased from PeproTech (Rocky Hill, NJ, USA). The primary antibodies against IRF4 and arginase1 were purchased from Cell Signaling Technology. All of these antibodies were diluted at 1:1,000 in PBS with 5% bovine serum albumin. Anti-β-Actin mAb (1:50,000) was purchased from Sigma-Aldrich.

Bone marrow cells were harvested and cultured with DMEM containing 10% (v/v) FBS and 10 ng/mL of mouse M-CSF for 7 days to obtain bone marrow-derived macrophages (BMDMs) (20, 21). Primary mouse peritoneal macrophages were obtained from the peritoneal exudates of 4–6-week-old mice. The peritoneal exudate cells were washed twice with PBS solution and adjusted to 1 × 10⁶ cells/mL in DMEM cultured for 3–4 h at 37°C and 5% CO₂ (22). The non-adherent cells were removed by washing with warm PBS. The purification of macrophage was analyzed by flow cytometry (Beckman, CA, USA), using mouse macrophage markers CD11b and F4/80. The adherent cells constituted more than 90% of CD11b⁺F4/80⁺ macrophages.

The arginase (Arg) activity assay was performed as described previously (23, 24). Briefly, the cells were lysed in 0.1% Triton X-100. Tris–HCl was then added to the cell lysates at a final concentration of 12.5 mM, and MnCl₂ was added to obtain a 1 mM final concentration. The Arg was activated by heating for 10 min at 56°C, and the l-arginine substrate was used at a final concentration of 250 mM. The reactions were incubated at 37°C for 30 min and stopped by the addition of H₂SO₄/H₃PO₄. After the addition of a-isonitrosopropiophenone and heating for 30 min at 95°C, the urea production was measured as the absorbance at 540 nm, and the data were normalized to the total protein content.

Cell death was measured using the Annexin V-FITC apoptosis detection kit (Abcam, Mountain View, CA, USA), according to the manufacturer’s instructions. After treatment with IL-4 and/or 2-DG for 48 h, the cells were harvested and washed twice with cold PBS (pH = 7.4). The cells were then incubated with 200 µL binding buffer containing Annexin V-FITC (40 µL/mL) and propidium iodide (PI; 1 µg/mL) for 15 min at room temperature in the dark. The population of PI and Annexin V-positive cells was analyzed by flow cytometry (Epics XL, Beckman Coulter Inc., Pasadena, CA, USA).

In order to determine the cellular toxicity of 2-DG, the levels of lactate dehydrogenase (LDH) released from macrophages were measured. After 48 h exposure of IL-4 and/or 2-DG treatment, cell-free supernatant aliquots were separated cells in each experimental sample by centrifuge, and supernatants were transferred to clean flat-bottom plate for enzymatic analysis. LDH in the culture supernatants was measured using commercially available LDH cytotoxicity detection kit-PLUS (Roche Applied Science, Mannheim, Germany). All samples were assayed by a microplate spectrophotometer (Thermo MK3, MA, USA).

Chitin (Sigma) was washed three times in PBS and then sonicated with a UR-20P device (Tomy) for 30 min on ice. After filtration with 100 µM cell strainer, chitin was diluted in 50 mL PBS. About 800 ng chitin with or without 2-DG (1,000 mg/kg) was intraperitoneally injected. The peritoneal macrophages were collected 2 days after administration.

Total RNAs were isolated with Trizol (Invitrogen), and reverse transcription was performed with M-MLV superscript reverse transcriptase according to the manufacturer’s instructions. Real-time PCR kit (SYBR Premix Ex Taq™, DRR041A) was purchased from Takara Bio Inc. PCR was done on CFX96 (Bio-Rad). To determine the relative mRNA expressions in response to IL-4, the mRNA expression levels of Arg, Fizz, Ym1, and CD206 were normalized to the housekeeping gene hypoxanthine phosphoribosyl transferase (25). Each sample was determined at least in triplicates. Primers used in the present study for the amplification were summarized in Table 1.

The peritoneal macrophages were cultured in DMEM medium with 10% FCS in 12-well plates. Cells were treated with IL-4 and the indicated reagents for the indicated time depending on the experiment purpose. After stimulation, the cells were washed once in cold PBS, lysed in RIPA buffer (50 mM Tris–HCl pH 7.4, 1% NP-40, 0.25% Na-deoxycholate, 150 mM NaCl, 1 mM EDTA pH7.4) with protease and phosphatase inhibitor cocktails (Sigma) for 10 min on a rocker at 4°C. Cells were scraped, centrifuged at 12,000 rpm for 10 min at 4°C, and the supernatants were mixed with 5× protein-loading buffer (26). Protein concentration is determined using a BCA assay. Proteins samples were analyzed on SDS-PAGE and transferred onto polyvinylidene difluoride (PVDF) membrane (Millipore). PVDF membrane was blocked with TBST (100 mM Tris–HCl pH7.5, 150 mM NaCl, 0.05% Tween20) with 5% non-fat dried milk for 1 h and then incubated with primary antibodies overnight on a shaker at 4°C. The appropriate HRP-coupled secondary antibody was used and detected through chemiluminescence (Millipore) (27). β-Actin was used as a protein-loading control. PVDF membranes were stripped, washed in TBST, and immunostained with the other antibody.

Glycolysis of macrophages was detected by measuring the detritiation of [3-³H]-glucose. In brief, the assay was initiated by adding 1μCi [3-³H]-glucose (Perkin Elmer) for 2 h. Medium was transferred to microcentrifuge tubes containing 50 µL 5 N HCL. The microcentrifuge tubes were then put in 20 mL scintillation vials containing 0.5 mL water and the vials capped and sealed. ³H₂O was separated from un-metabolized [3-³H]-glucose by evaporation diffusion for 24 h at room temperature.

A total of 2 × 10⁶ cells per well were seeded in 12-well plates for 2 days. The medium was collected, and the glucose and lactate levels were examined immediately. Glucose and lactate were measured spectrophotometrically using an Olympus AU5400. The glucose consumption and lactate production were normalized to cell numbers.

B16-F1 melanoma (B16) cells (1 × 10⁶, in 200 µL PBS) were injected subcutaneously into the flanks of mice. 2-DG treatment was once per day from 9 days (1,000 mg/kg) (28). Male C57BL/6 strains of mice (6–8 weeks) were used. There was no systematic means of randomization of mice, and the experiment was carried out blindly throughout. The mice were sacrificed on day 19. Tumors were resected and transferred to 5 mL PBS on ice. Tumor weight was measured on a scale. The tumors were then processed for flow cytometry sorting on the same day or fixed in formalin for immunohistochemistry.

OVA-induced acute allergic inflammation was elicited as previously described by i.p. sensitization with 20 mg OVA mixed with 2 mg alum at days 1 and 14 (22, 29). From day 21, mice followed by nebulizer-delivered airway challenges with 1% OVA for 5 consecutive days. The mice were pretreated 2 days before aerosolized and intraperitoneally injected 2-DG (1,000 mg/mL) continuous till the sacrifice. Mice were assessed 24 h after the last OVA challenge. Cells from bronchoalveolar lavage fluid (BALF) were stained with anti-mSiglecF, anti-mCD11c, anti-mCD206, anti-mCD11b, and anti-mCD45 mAbs and analyzed by a flow cytometry. The expressions of M2 markers including Arg, Fizz, Ym1, and CD206 in these cells were detected by real-time PCR. Lung tissues were fixed with formaldehyde and embedded in paraffin. The thin sections of the embedded tissues were stained with hematoxylin and eosin (H&E) stain. The evaluation of inflammatory infiltrate was done microscopically.

Bronchoalveolar lavage fluid was performed according to the method of Haque et al. (30). In brief, the lungs were lavaged in situ three times with 1.5 mL cold PBS. Recovered BALF was immediately cooled to 4°C and centrifuged at 1,700 rpm for 5 min (27). The BALF cells were stained for CD45⁺CD11c⁺SiglecF⁺ and sorted by a MoFlo XDP Cell Sorter (Beckman). The purification of alveolar macrophages was up to 98%.

All data are presented as the mean ± SD. Statistical significance was determined by two-tailed Student’s t-test or by one-way ANOVA. Data were analyzed using one-way ANOVA followed by post hoc comparisons using the Tukey’s multiple or two-way ANOVA statistical analysis followed by a post hoc test. A p value less than 0.05 was considered as statistically significant.

It has been well appreciated that M1-polarized macrophages increased glycolysis to quickly trigger microbicidal activity. However, how glycolytic activity is regulated in M2 macrophage polarization was poorly understood. Thus, we compared the glucose consumption and lactate production in M0 and M2 macrophages using freshly isolated peritoneal macrophages from naïve B6 mice. As shown in Figure 1, M2 macrophages had an enhanced glycolysis (p < 0.01, Figures 1A,B). The glycolytic uptake of M2 macrophages was measured by the generation of ³H-labeled glucose from [3-³H]-2-DG. Our results showed that M2 macrophages contained much higher glycolytic uptake than M0 macrophages (p < 0.01, Figure 1C). Glucose utilization depends on a chain of reactions catalyzed by multiple key enzymes, eventually leading to the generation of lactate and net production of two ATP molecules as the energy source. Real-time PCR analysis revealed that IL-4-stimulated M2 macrophages markedly increased these genes encoding glycolysis-related molecules, including the transporter Glut1 (Slc2a1), hexokinase II (HK II) and aldolases (Aldoc) (p < 0.01, Figure 1D). These data collectively indicate strong upregulation of glucose metabolism during IL-4-induced M2 macrophage polarization. To directly test the importance of the metabolic reprogramming in M2 macrophage activation, we induced M2 polarization in the presence or absence of 2-DG. While IL-4 stimulation for 48 h, macrophages were treated with appropriate doses of 2-DG, and their mRNA expression profiles were analyzed over a time course ranging from 12 to 36 h using real-time PCR. We found that the mRNA expression of Arg, Fizz, and Ym1 reduced even 2-DG treatment was employed as later as 36 h after IL-4 induction (p < 0.01, Figure 1E). Similarly, the IL-4-induced mRNA expression of Arg, Fizz, and Ym1 was substantially diminished as the concentrations of 2-DG increasing, indicating that 2-DG inhibits M2 polarization in a dose-dependent manner (p < 0.01, Figure 1F). Additionally, 2-DG treatment resulted in decreased production and activity of Arg by IL-4-induced M2 macrophages (p < 0.01, Figures 1F,G). Furthermore, 2-DG severely reduced the expression of CD206, a cell surface marker for M2 macrophages, on IL-4-treated macrophages as determined by flow cytometry assays (p < 0.01, Figure 1I; Figure S1 in Supplementary Material). Together, these results suggest that glycolysis is necessary and significant for macrophage polarization and low doses of 2-DG treatment blocks M2 macrophage polarization in a time- and dose-dependent manner.

In order to exclude the possibility that the decreased M2 polarization caused by 2-DG is simply caused by the increased cell death, we detected the cell death ratio after macrophages were treated with as high as 1 mM 2-DG in vitro. We observed 2-DG treatment did not remarkably alter the cell activity of M2 macrophages (Figure 2A). Furthermore, M2 macrophages treated with or without 1 mM 2-DG were harvested and stained with Annexin V and PI. Flow cytometry less than 7% of 2-DG-treated M2 macrophages showed Annexin V⁺PI⁻, and approximately 1% of the cells were Annexin V⁺PI⁺, which is similar to 2-DG-untreated macrophages (Figure 2B). Moreover, the level of LDH released from 2-DG-treated macrophages was similar with M0 and IL-4-treated macrophages (Figure S2 in Supplementary Material). These results excludes the possibility that cell death contributes to the poor M2 macrophage polarization caused by the treatment with low doses of 2-DG.

The glucose consumption and lactate production were also significantly inhibited by 2-DG treatment during IL-4-induced M2 polarization using the in vitro M-CSF-induced BMDMs, as observed in freshly isolated peritoneal macrophages (p < 0.01, Figure 2C). The cellular requirements for energy can be supplied by the enzymatically regulated glycolysis of glucose into glucose-6-phosphate, accompanied by the release of ATP; an endpoint of this process is the release of pyruvate from cells which is the enzyme substrate for TCA cycle. Glucose process depends on a chain of reactions catalyzed by multiple important enzymes. Thus, we used Western blots to detect the expressions of HK II and PDH in M2 macrophages after 2-DG treatment. Consistent with the decreased glucose consumption, 2-DG treatment significantly decreased the HK II and PDH protein expressions in M2 macrophages (p < 0.01, Figure 2D). These results showed that 2-DG treatment could significantly impact the glucose metabolism states of M2 macrophages.

Hif-1α is a transcription factor crucial for the induction of a M2-like macrophage polarization and controls the expression of genes encoding rate-limiting components of the glycolytic pathway (31). Additionally, the functions of Hif-1α are most well defined in relation to modifying glycolysis (32, 33). To assess whether Hif-1α regulation is involved in the poor M2 macrophage polarization caused by 2-DG, we therefore detected the Hif-1α expression in macrophages treated with 2-DG. The expression of Hif-1α was increased in IL-4-induced M2 macrophages. However, 2-DG treatment significantly inhibited the IL-4-induced Hif-1α mRNA expression in macrophages as detected by real-time PCR (p < 0.01, Figure 3A; Figure S3 in Supplementary Material). Nicely consistent with the mRNA expression, we detected less expression of Hif-1α in macrophages treated with IL-4 and 2-DG than those in macrophages treated with IL-4 alone by Western blots (Figure 3B). It is reported that Bay 87-2243 is a Hif-1α inhibitor (34). Bay 87-2243 treatment significantly decreased the expressions of M2 makers like Arg, Ym1, and CD206 in IL-4-induced M2 macrophages as assayed by real-time PCR (p < 0.01, Figure 3C). The Arg protein expression and activity were also decreased after Bay 87-2243 treatment as detected by Western blots (Figures 3D,E). In addition, as the treatment of Hif-1α activator, IOX2 (35, 36), for 48 h could significantly reverse the 2-DG-impaired expressions of M2 makers including Arg, Ym1, and CD206 molecules as determined by real-time PCR (p < 0.01, Figures 3F–H). Similarly, IOX2 treatment rescued the decreased activity of Arg in M2 macrophages under 2-DG treatment (p < 0.01, Figure 3I). As a suppressor of tumorigenesis and inflammation, AMPK activation opposes most of the metabolic alterations that occur in proliferating cells (37). Furthermore, it has been shown that 2-DG administration had effects on AMPK activation (38). Thus, we detected the activation of AMPK during M2 macrophage induction in the presence and absence of 2-DG. The phosphorylation of AMPK in macrophages was decreased after IL-4 treatment. However, 2-DG treatment rescued this alteration as determined by Western blots (Figure 4A; Figure S4 in Supplementary Material). It is reported that compound C inhibits AMPK activation (39, 40). Compound C treatment significantly rescued the 2-DG treatment-decreased expressions of Arg, Fizz, and Ym1 molecules during M2 macrophage induction as detected by quantitative PCR (p < 0.01, Figures 4B–D). In consistent with mRNA levels, we found that the inhibition of AMPK under compound C treatment could upregulate expression of Hif-1α and arginase in the presence of 2-DG as determined by Western blots (Figure 4E). 2-DG treatment significantly decreased the percentage of CD206⁺ cells in F4/80⁺CD11b⁺ macrophages in the presence of IL-4 (p < 0.01, Figure 4F). However, compound C treatment could significantly rescue the decreased percentage of CD206⁺ cells in F4/80⁺CD11b⁺ macrophages in the presence of IL-4 in the presence of 2-DG as determined by flow cytometry (p < 0.01, Figure 4F). Therefore, the deficiency of AMPK activation induced Hif-1α expression and then enhanced M2 macrophages polarization.

Chitin, a polymerized sugar, is a structural component of helminths, arthropods and fungi (41). Chitin administration recruits M2 macrophages to the administration location, which are very important for the following recruitment of eosinophils (42, 43). Indeed, the intraperitoneal administration of chitin significantly recruited more cells, including F4/80⁺CD11b⁺ macrophages and Siglec-F⁺CD11b⁺ eosinophils to the peritoneal cavity after 48 h (Figures 5A–D). However, 2-DG treatment decreased the total cell number of peritoneal cells, and importantly, the cell numbers of the recruited Siglec-F⁺CD11b⁺ eosinophils after chitin treatment (Figures 5A,C,D). Consistent with this observation, Chitin-elicited peritoneal macrophages expressed higher levels of Arg, Fizz and Ym1, the hall-markers of M2 macrophages (44), than control peritoneal macrophages isolated from chitin-untreated mice (p < 0.01, Figure 5E). However, 2-DG treatment considerably reduced the expression of genes encoding Arg, Fizz, Ym1, CD206, and MCP-1 in chitin-induced peritoneal macrophages compared with those of only chitin-elicited mice (p < 0.01, Figure 5E). These results suggest that 2-DG prevents M2 macrophage polarization in response to chitin administration in vivo.

Tumor-associated macrophages (TAMs) are macrophages residing in the tumor microenvironments (3). TAMs (defined as CD45⁺CD11b⁺CD64⁺F4/80⁺ cells) are recognized as M2-polarized macrophages abundantly expressing M2-specific markers such as Arg1, CD206 and high levels of vascular endothelial growth factor (Vegf) (31, 45). To determine aerobic glycolysis on TAMs, we subcutaneously injected 1 × 10⁶ B16 cells, intraperitoneally injected 2-DG from day 9 and harvested on day 19. In the tumor-bearing mouse model, 2-DG significantly inhibited the tumor growth as evidenced by the smaller tumor size and the less tumor weight compared with control tumors (p < 0.05, Figures 6A–C). Histological analysis of tumor sections revealed larger necrotic area of tumor in control mice than those in 2-DG treated mice (Figure 6D). Moreover, the sorted TAMs by flow cytometry showed severely reduction of Arg, Fizz, CD206 and Vegf expressions after 2-DG treatment (p < 0.05, Figure 6E). In addition, we assessed the direct effect of 2-DG on cell growth of B16 tumor cells in vitro. As expected, high doses of 2-DG treatment significantly inhibited B16 tumor cell growth in vitro, but low doses of 2-DG (1 mM) failed to do so (Figure S5 in Supplementary Material), indicating that 2-DG can directly inhibit tumor cell growth. These data collectively suggest that 2-DG treatment impacts the polarization of M2-like TAMs in tumor mass in vivo while 2-DG has the ability to direct inhibit tumor cell proliferation.

In an OVA-induced allergic airway inflammatory mouse model, in which M2 macrophages play a critical role. To address whether 2-DG acts on allergic airway inflammation, we treated OVA-sensitized mice with 2-DG 2 days before the first OVA challenge for 5 consecutive days (Figure 7A). 2-DG injection decreased the cell number in BALF (Figure 7B) and decreased the pathogenesis in lungs as evidenced by H&E staining (Figure 7C; Figure S6 in Supplementary Material). The percentages and cell numbers of eosinophils in the BALF were significantly lower in 2-DG-treated asthma mice than untreated asthma mice (p < 0.05, Figures 7D,E; Figure S7 in Supplementary Material), whereas identical cell numbers of the infiltrated macrophages in both mice were observed (Figures 7F,G). Consistent with the previous reports (22), alveolar macrophages of OVA-challenged mice expressed significantly higher CD206 molecules than those of OVA-unchallenged control mice (Figure 7H; Figure S8 in Supplementary Material). However, macrophages from 2-DG-treated allergic airway inflammatory mice showed lower CD206 expression than only OVA-challenged mice (Figure 7H). Moreover, the freshly isolated alveolar macrophages sorted from BALF cells from 2-DG-treated allergic airway inflammatory mice expressed significantly less Arg, Fizz, and CD206 molecules as detected by quantitative PCR (p < 0.05, Figure 7I; Figure S9 in Supplementary Material). Thus, 2-DG treatment significantly blocks M2 macrophage polarization and prevent the pathogenesis of allergic airway inflammation in mice.