Methylglyoxal suppresses human colon cancer cell lines and tumor growth in a mouse model by impairing glycolytic metabolism of cancer cells associated with down-regulation of c-Myc expression

ABSTRACT

Methylglyoxal (MG) is a highly reactive dicarbonyl compound exhibiting anti-tumor activity. The anti-tumor effects of MG have been demonstrated in some types of cancer, but its role in colon cancer and the mechanisms underlying this activity remain largely unknown. We investigated its role in human colon cancer and the underlying mechanism using human colon cancer cells and animal model. Viability, proliferation, and apoptosis were quantified in DLD-1 and SW480 colon cancer cells by using the Cell Counting Kit-8, plate colony formation assay, and flow cytometry, respectively. Cell migration and invasion were assessed by wound healing and transwell assays. Glucose consumption, lactate production, and intracellular ATP production also were assayed. The levels of c-Myc protein and mRNA were quantitated by western blot and qRT-PCR. The anti-tumor role of MG in vivo was investigated in a DLD-1 xenograft tumor model in nude mice. We demonstrated that MG inhibited viability, proliferation, migration, and invasion and induced apoptosis of DLD-1 and SW480 colon cancer cells. Treatment with MG reduced glucose consumption, lactate production, and ATP production and decreased c-Myc protein levels in these cells. Moreover, MG significantly suppressed tumor growth and c-Myc expression in vivo. Our findings suggest that MG plays an anti-tumor role in colon cancer. It inhibits cancer cell growth by altering the glycolytic pathway associated with downregulation of c-Myc protein. MG has therapeutic potential in colon cancer by interrupting cancer metabolism.

Abbreviations

AG

aminoguanidine

CCK-8

Cell Counting Kit-8

FBS

fetal bovine serum

MG

methylglyoxal

PBS

phosphate-buffered saline solution

qRT-PCR

quantitative real-time PCR

Introduction

Colon cancer is the third most common cancer and the fourth leading cause of cancer death worldwide.¹ With the increasing prevalence of risk factors such as low fiber/high fat diet, obesity, and smoking as well as delayed diagnosis and ineffective therapy, colon cancer continues to represent a major disease burden worldwide. Moreover, the incidence is increasing among people younger than 50 y.² The current management of colon cancer relies primarily on surgery (-laparoscopic colectomy) and adjuvant chemotherapy, with or without concurrent irradiation depending on the tumor location and stage.³ Nevertheless, recurrence and metastasis occur frequently in patients with colon cancer and are associated with low survival rate and poor prognosis. Thus improved treatment strategies are imperative to improving patient outcomes.

Methylglyoxal (MG) is a small dicarbonyl compound ubiquitously produced in all cells under either physiological or pathological conditions.⁴ It is formed through various metabolic routes via both enzymatic and non-enzymatic pathways, predominantly the glycolytic pathway.⁵ MG is produced in only small amounts under normal conditions, but its expression is greatly increased in hyperglycemic disorders such as diabetes.^6,7 It is catalyzed mainly by the glyoxalase system and is converted to D-lactate.⁸

MG is extremely reactive and is toxic to cell.⁹ MG exposure leads to decreased synthesis of protein, inhibition of enzymes activity, increased oxidative stress, activated signal transduction, and other cellular dysfunctions.¹⁰ MG inhibits proliferation of malignant cells by suppressing synthesis of biological macromolecules, affecting cellular metabolism and arresting the cell cycle.¹¹ MG induces apoptosis through multiple mechanisms such as accumulating reactive oxygen species and advanced glycation end products and triggering the mitochondrial apoptotic pathway.^12-14 MG also inhibits tumor metastasis, enhances the immune response, and increases sensitivity to anti-tumor agents.^15-18

c-Myc is one of the most widely expressed human proto-oncogenes and is frequently overexpressed in colon cancer.¹⁹ It has been shown to play an important role in the development and progression of this cancer.²⁰ As a regulator of transcription, it modulates tumor cell cycle, differentiation, metabolic activity, apoptosis, angiogenesis, and other biological functions.¹⁹ c-Myc is involved in regulating glycolytic genes and thus enhancing glycolysis.²¹ Its overexpression has the effect of addicting cancer cells to glucose or glutamine, without which the cells die.²² Emerging evidence suggests that cancer is a metabolic disease, and what has been learned about the Warburg effect indicates that malignant cells rely on glycolysis as their major energy source.^23,24 These findings suggest that c-Myc is involved in regulating glycolytic metabolism in colon cancer.

Although the anti-tumor effects of MG have been demonstrated in preliminary studies in some types of cancer, its role in colon cancer remains largely unknown. In the current study, we investigated the effects of MG on colon cancer cells in vitro and in vivo and the molecular mechanisms underlying these effects.

Results

MG inhibits viability and proliferation of colon cancer cells

As shown in Fig. 1A, MG reduced cancer cell viability in a time- and concentration-dependent manner. MG at a concentration of 0.25 mM at all-time points or at a concentration of 0.5 mM at 6 h and 12 h had no notable effect on DLD-1 cell viability. However, viability of these cells was significantly reduced when incubated with MG at a concentration of 1.0 mM for 24 h, and this inhibitory effect became more pronounced with increased concentration and prolonged incubation (P < 0.01 to 0.05). MG had similar effects in SW480 cells (P < 0.01 to 0.05; Fig. 1B). These MG effects were reversed by the MG scavenger aminoguanidine (AG) (P < 0.05; Fig. 1C–F).

Figure 1.

MG inhibits the viability of colon cancer cells. (A-B) MG reduced the viability of DLD-1 cells (A) and SW480 cells (B) in a time- (0, 6, 12, 24 and 48 h) and concentration- (0, 0.025, 0.5, 1.0 and 2.0 mM) dependent manner. (C-D) The effects of MG on cell viability were reversed by the MG scavenger AG (1.0 mM) in DLD-1 cells with co-incubation for 24 h (C) or 48 h (D). (E-F) Similar effects were observed in SW480 cells after co-incubation with AG for 24 h (E) or 48 h (F). * P < 0.05, ** P < 0.01 compared with control; ^δ P < 0.05 compared with 0.25 mM MG; ⁺ P < 0.05,²⁺ P < 0.01 compared with 0.5 mM MG; ^# P < 0.05 compared with 1.0 mM MG.

DLD-1 and SW480 cancer cells incubated with 0.5 mM MG formed significantly fewer colonies than their untreated counterparts (P < 0.01; Fig. 2A–B). There was almost no colony formation by cells treated with 1.0 mM MG (P < 0.01; Fig. 2A–B). Co-incubation with AG attenuated the inhibitory effects of MG on colony formation (P < 0.05; Fig. 2A–B).

Figure 2.

MG inhibits proliferation of colon cancer cells. (A-B) MG (0.5 and 1.0 mM) reduced the numbers of colonies formed by DLD-1 cells (A) and SW480 cells (B) after treated with 24 h. Co-incubation with AG (1.0 mM) for 24 h attenuated the inhibitory effects of MG on colony formation. ** P < 0.01 compared with control; ⁺ P < 0.05 compared with 0.5 mM MG; ^# P < 0.05 compared with 1.0 mM MG.

MG induces apoptosis of colon cancer cells

Incubation of DLD-1 cells with 0.5 mM or 1.0 mM MG increased cell apoptosis rate to 5.5 % and 8.9 %, respectively, compared with only 2.0 % in the control group (P < 0.01 to 0.05; Fig. 3A). Similar effects were observed in SW480 cells; MG at concentrations of 0.5 mM and 1.0 mM increased apoptosis to 6.2 % and 9.8 %, respectively, though the control group rate was 5.1 % (P < 0.05; Fig. 3B). Apoptosis rate was significantly higher in the group treated with 1.0 mM MG than in the group treated with 0.5 mM MG (P < 0.05; Fig. 3A-B). The MG-induced apoptosis was reversed by AG (P < 0.05; Fig. 3A–B).

Figure 3.

MG induces apoptosis of colon cancer cells. (A-B) Co-incubation with MG (0.5 and 1.0 mM) for 24 h increased apoptosis rates of DLD-1 cells (A) and SW480 cells (B). The MG-induced apoptosis was reversed by AG (1.0 mM) after co-treated with 24 h. * P < 0.05, ** P < 0.01 compared with control; ⁺ P < 0.05 compared with 0.5 mM MG; ^# P < 0.05 compared with 1.0 mM MG.

MG suppresses colon cancer cell migration and invasion

DLD-1 cells treated with MG exhibited a significant decrease in cell migration, as determined by wound width, compared with controls. The wound closure rates of DLD-1 cells treated with 1.0 mM MG were only 12 ± 1.9 %, 14 ± 2.9 %, and 16 ± 2.5 % at 12 h, 24 h, and 48 h, respectively, while the rate was 71 ± 9.8 % at 48 h in the control group (P < 0.01 to 0.05; Fig. 4A). Similarly, SW480 cell migration was markedly reduced by MG (Fig. 4B). Treatment with 0.5 mM or 1.0 mM MG reduced the number of cells penetrating the transwell membrane to only 66 % and 32 %, respectively, compared to the control (P < 0.01 to 0.05; Fig. 4B). Co-treatment with AG reversed the inhibitory effect of MG on cell migration (P < 0.05; Fig. 4A–B).

Figure 4.

MG suppresses colon cancer cell migration and invasion. (A) MG (0.5 and 1.0 mM) reduced DLD-1 cell migration, as determined by the wound healing assay at 0, 12, 24 and 48 h, greater wound width signifies less cell migration. (B) MG (0.5 and 1.0 mM) also reduced SW480 cell migration after treated with 24 h, as shown by number of cells that penetrated the membrane. Migrated cells were photographed at 40× magnification. (C-D) Treatment with MG (0.5 and 1.0 mM) for 24h reduced invasion of DLD-1 cells (C) and SW480 cells (D). Cells that penetrated the membrane were photographed at 100× magnification. Co-incubation with AG (1.0 mM) reversed the MG-mediated inhibition of cell migration and invasion. * P < 0.05, ** P < 0.01 compared with control; ⁺ P < 0.05,²⁺ P < 0.01 compared with 0.5 mM MG; ^# P < 0.05 compared with 1.0 mM MG.

The invasion capacity of the cancer cells was also attenuated by MG. As shown in Fig. 4C–D, the number of MG-treated cells penetrating the membrane in the matrigel transwell assay was much lower than the number of control cells penetrating the membrane. At MG doses of 0.5 mM and 1.0 mM, 62 % and 21 % of DLD-1 cells penetrated into the lower chamber, respectively, compared with untreated cells (P < 0.01 to 0.05; Fig. 4C). Similarly, treatment with the same doses of MG reduced the numbers of SW480 cells that penetrated into the lower chamber to 76 % and 28 %, respectively, compared with controls (P < 0.05; Fig. 4D). The inhibitory effect on cell invasion was much greater in cells incubated with 1.0 mM MG than in those incubated with 0.5 mM MG (P < 0.05; Fig. 4C–D). Co-incubation with AG reversed this effect (P < 0.05; Fig. 4C–D).

MG impairs glycolysis of colon cancer cells

Treatment with MG resulted in dramatic decrease in the glucose consumption and lactate production of DLD-1 cells. Incubation with 0.5 mM or 1.0 mM MG reduced glucose consumption to 85 % and 52 %, respectively, of that of controls (P < 0.01 to 0.05; Fig. 5A). The same doses of MG reduced lactate production to 70 % and 46 %, respectively, of that of controls (P < 0.01 to 0.05; Fig. 5B). The intracellular ATP production was reduced accordingly in DLD-1 cells treated with MG (P < 0.05; Fig. 5C). The higher dose of MG (1.0 mM) led to markedly lower glucose consumption, lactate production, and ATP production than the lower dose of MG (0.5 mM) (P < 0.01 to 0.05; Fig. 5A–C). These observed effects of MG were reversed by co-incubation with AG (P < 0.05; Fig. 5A–C). MG had similar effects in SW480 cells, glucose consumption, lactate production and ATP production were significantly reduced by treatment with 1.0 mM MG, and the lower dose (0.5mM) had light inhibitory effect (P < 0.01 to 0.05; Fig. 5D–F). Co-incubation with AG attenuated these MG effects (P < 0.05; Fig. 5D–F).

Figure 5.

MG impairs glycolysis of colon cancer cell lines. (A-C) Incubation with 0.5 mM or 1.0 mM MG for 24 h resulted in dramatic decreases in glucose consumption (A), lactate production (B), and ATP production (C) of DLD-1 cells. (D-F) In SW480 cells, glucose consumption (D) and lactate production (E) were significantly reduced by treatment with 1.0 mM MG, there was light inhibitory effect at 0.5 mM MG with 24 h. Intracellular ATP production of SW480 cells (F) was decreased. Co-incubation with AG (1.0 mM) prevented the observed effects of MG. * P < 0.05, ** P < 0.01 compared with control; ⁺ P < 0.05, ²⁺ P < 0.01 compared with 0.5 mM MG; ^# P < 0.05 compared with 1.0 mM MG.

MG reduces c-Myc expression in colon cancer cell lines

c-Myc protein levels were reduced by MG (Fig. 6A–B). c-Myc expression was reduced notably in DLD-1 cells treated with 0.5 mM MG (P < 0.05) but not in SW480 cells (Fig. 6A–B). The c-Myc protein levels were reduced by more than 90 % in DLD-1 cells treated with 1.0 mM MG (P < 0.01) but by only approximately 50 % in SW480 cells under the same condition (P < 0.01; Fig. 6A–B). Co-incubation of these cells with AG prevented the MG-induced inhibition of c-Myc expression (P < 0.05; Fig. 6A–B). In investigating the mechanism underlying the difference between the 2 cell lines, further analysis by qRT-PCR showed that c-Myc mRNA levels were significantly lower in DLD-1 cells than in SW480 cells (P < 0.05; Fig. 6C). This may make SW480 cells less sensitive to MG.

Figure 6.

MG reduces c-Myc expression in colon cancer cell lines. (A-B) Treatment with MG (0.5 and 1.0 mM) for 24 h significantly reduced c-Myc expression in DLD-1 cells (A) but only moderately reduced c-Myc expression in SW480 cells (B). Co-incubation of cells with AG (1.0 mM) prevented the inhibition of c-Myc expression. (C) c-Myc mRNA expression was lower in DLD-1 cells than in SW480 cells. * P < 0.05, ** P < 0.01 compared with control; ²⁺ P < 0.01 compared with 0.5 mM MG; ^# P < 0.05 compared with 1.0 mM MG.

MG reduces in vivo tumor growth

Significant inhibition of DLD-1 tumor growth was observed from treatment day 8 (day 15 after injection of DLD-1 cells) in mice treated with MG (P < 0.01 to 0.05; Fig. 7A). The administration of MG at doses of 40 mg/kg or 80 mg/kg reduced the tumor weights by 35 % and 40 %, respectively, at day 23 (P < 0.01 to 0.05; Fig. 7C). During the course of treatment, MG did not cause visible side effects or changes in mouse body weight (Fig. 7B) c-Myc protein levels in tumor tissues as detected by protein gel blot were reduced to approximately 32 % for mice treated with 40 mg/kg MG and 22 % for mice treated with 80 mg/kg MG relative to that in the tumors of the control group (P < 0.01; Fig. 7D). This result was confirmed by immunohistochemical analysis, tumor tissues with MG treatment showed much lower c-Myc expression compared with controls (Fig. 7E). There were no statistically significant differences in tumor growth, tumor weight, or c-Myc expression between the low MG dose group and high MG dose group.

Figure 7.

MG reduces in vivo tumor growth. (A) Growth of DLD-1 tumor xenografts in nude mice was significantly inhibited from treatment day 8 in mice treated with MG (40 and 80 mg/kg), the treatment lasted 16 d. (B) MG did not cause visible changes in nude mice body weight. (C) The administration of MG decreased the tumor weights. (D-E) MG reduced the levels of c-Myc protein in tumor tissue samples determined by protein gel blot (D) and immunohistochemistry (E). There were no statistically significant differences in tumor growth, tumor weight, or tumor c-Myc expression between the group that received 40 mg/kg MG and the group that received 80 mg/kg MG. * P < 0.05, ** P < 0.01, compared with control.

Discussion

Our findings demonstrate that MG inhibited viability, proliferation, migration, and invasion and induced apoptosis of DLD-1 and SW480 colon cancer cells. Moreover, MG reduced glucose consumption, lactate production, and ATP production and decreased c-Myc protein expression in these cells. MG suppressed DLD-1 tumor growth and tumoral c-Myc expression in an in vivo mouse xenograft model.

A number of in vitro and in vivo studies have shown the anti-tumor efficacy of MG, which is due mainly to its anti-proliferative and pro-apoptotic properties.^12-14,25-28 The anti-tumor role of MG may also be attributed to its effects on cell migration. ¹⁵ Our results showing that MG inhibited the viability, proliferation, migration, and invasion of DLD-1 and SW480 colon cancer cells while promoting their apoptosis are in accordance with previous results in other cancer types.^12-14,25-28 There are multiple potential mechanisms by which MG may induce tumor suppression: (1) inhibition of the synthesis of proteins and nucleic acids via irreversible modification of protein, DNA, and RNA;¹¹ (2) blocking of cell cycle G1→S progression through modulation of G1-specific cell cycle regulators (cyclin D1, cdk2, cdk4, and pRB) and formation of DNA crosslinks;^11,25 (3) impairment of cell respiration, leading to energy depletion in malignant cells by inhibiting the mitochondrial complex;²⁹ (4) desensitization of the NF-κB signaling pathway, down-regulation of anti-apoptotic proteins Bcl-2 or Bcl-XL, and up-regulation of pro-apoptotic protein Bax;¹³ (5) activation of the p38-MAPK, p42/44-MAPK signaling pathway, causing a rapid cytochrome c release and triggering the caspase cascade apoptotic pathway;^12,14 and (6) reduction of tumor migration via restoration of the transcriptional activities of wild-type p53.¹⁵ Moreover, AG abolished the MG-induced anti-tumor effects by acting as a scavenger, reacting rapidly with MG.³⁰

Cancer cells favor the aerobic glycolytic pathway and generate energy predominantly by increasing their rate of glycolysis.²⁴ MG is an intermediate metabolite of glucose.⁵ We postulate that MG induces its anti-tumor effects through disruption of cancer metabolism and glycolysis. Our results demonstrating that MG reduced glucose consumption and lactate production and impaired ATP supply in human colon cancer cells are supported by other reports in which MG has been shown to inhibit glycolysis through targeting lactate dehydrogenase (LDH) and glyceraldehyde 3-phosphate dehydrogenase (G3PDH). Notably, their inclusion of normal tissues and cells suggested that MG particularly inhibits the respiration of malignant cells but has no inhibitory effect on normal cells.^31-33

c-Myc has been reported to act as a pivotal regulator of cancer metabolism, directly regulating the glycolytic pathway.³⁴ Furthermore, MG also was demonstrated to activate the MAPK signaling pathway, which terminates in ERK-mediated c-Myc alteration.^35,36 We showed that c-Myc protein levels were decreased by MG in cancer cells, but we also found that c-Myc expression was more sensitive to MG in DLD-1 cells than in SW480 cells. This difference in c-Myc sensitivity to MG may be attributable to the significant difference in baseline c-Myc mRNA levels in the 2 cell lines. c-Myc is closely linked with altered glycolysis in cancer cells.³⁷ As a key transcription factor, c-Myc drives the expression of many glucose metabolism genes, particularly glucose transporter1 (GLUT1), hexokinase 2 (HK2), pyruvate kinase M2 (PKM2), and lactate dehydrogenase A (LDHA).³⁷ Our results indicate that the glycolytic changes caused by MG are associated with c-Myc downregulation.

We demonstrated that MG can dramatically suppress tumor growth and decrease tumor expression of c-Myc protein in a murine xenograft tumor model. These ex vivo results further confirm the in vitro findings. However, further studies are warranted to illustrate the relationship between MG and c-Myc and the exact pathways regulating that relationship. Furthermore, it would be interesting to determine the effect of MG treatment on survival rate in future study.

In conclusion, our studies demonstrate for the first time that MG plays an anti-tumor role in colon cancer. Our data indicate that MG inhibits cancer cell growth by altering the glycolytic pathway associated with down-regulation of the c-Myc protein. These findings provide evidence that MG is a potential therapeutic agent for colon cancer. Further investigations of the biological roles of MG in cancer and their molecular mechanisms should be conducted in animal models.

Materials and methods

Cell culture and regents

Human DLD-1 colon cancer cells (derived from a Dukes' stage C colon carcinoma) and SW480 cells (Dukes' stage B colon carcinoma) were purchased from the Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences. Both of the cell lines were cultured in RPMI-1640 medium (Gibco, 11875093) supplemented with 10 % fetal bovine serum (FBS; Biological Industries, 040011A), 100 IU/mL penicillin and 100 µg/mL streptomycin (Solarbio, P1400). The cells were maintained at 37°C in a 5 % CO₂ atmosphere.

MG was purchased from Sigma-Aldrich (M0252), stored at 4°C. It was diluted with phosphate-buffered saline solution (PBS) when used. MG scavenger aminoguanidine (AG) was purchased from Sigma-Aldrich (396494) too, it was dissolved in PBS and stored at 4°C.

Cell viability and proliferation assays

Cell viability was determined by using the Cell Counting Kit-8 (CCK-8; Dojindo, CK04). DLD-1 cells and SW480 cells were seeded in Corning 96-well plates (Corning, 3599) in triplicate at 1.2 × 10⁴ and 1.5 × 10⁴ cells per well, respectively, and incubated for 24 h. Cells were then treated with various concentrations of MG (0, 0.25, 0.5, or 1.0 mM), with or without AG (1.0 mM), and incubated for 6 h, 12 h, 24 h, or 48 h. After the cells were washed with PBS, CCK-8 reagent (10 µL) was added to each well. Absorbance was measured at 450 nm with a Multimode Microplate Reader Varioskan Flash (Thermo Scientific) after incubation at 37°C for 1 h.

Cell proliferation was estimated by the plate colony formation assay. DLD-1 cells and SW480 cells were seeded in Corning 6-well plates at 200 cells per well. After incubation for 24 h, cell culture medium with various concentrations of MG, with or without AG, was added to each well. Two weeks later, colonies were fixed in 4 % paraformaldehyde for 20 min and stained with crystal violet (Beyotime Institute of Biotechnology, C0121) for 20 min. The colonies were counted immediately after washing with PBS.

Cell apoptosis assay

DLD-1 and SW480 cells were seeded in Corning 6-well plates and treated with MG at various concentrations (0, 0.5, or 1.0 mM), with or without AG (1.0 mM), for 24 h. Cells were harvested and washed twice with PBS, then were stained with Annexin V-allophycocyanin (APC) and propidium iodide (PI; Annexin V-APC Cell Apoptosis Analysis kit, KeyGen Biotech, KGA1030) for 15 min according to the manufacturer′s protocol. Cell apoptosis level was quantified by a FACS-Calibur flow cytometer (BD Bioscience).

Cell migration and invasion assays

Monolayer wound healing assay was performed to mimic DLD-1 cells migration. DLD-1 cells were seeded in Corning 12-well plates at a density of 3.5 × 10⁵ cells per well. After incubation for 12 h, several wound lines were scratched to the bottom of the well using a 200-µL pipette tip. The cells were then incubated with medium containing 2 % FBS and MG (0, 0.5, or 1.0 mM), with or without AG (1.0 mM). Each wound was photographed under a Nikon Eclipse TS100 microscope at 0 h, 12 h, 24 h, and 48 h, and the wound width was determined at each time point. Wound closure was calculated as follows: percentage wound closure = 1 - (width_t/width₀) × 100 %.

Since the SW480 cells were not very adherent during culture, their migration was assessed by the transwell migration assay. SW480 cells were placed in Corning 6-well plates, and upon reaching 60-70 % confluence, medium containing various concentrations of MG (0, 0.5, or 1.0 mM), with or without AG (1.0 mM), were applied. After incubation for 24 h, cells were harvested, washed, and resuspended in serum-free RPMI-1640 medium. Cells were then seeded into 24-well transwell chambers (Corning, 3422) at a density of 8 × 10⁴ cells per well. Medium containing 10 % FBS was added in the lower chamber as a chemoattractant. After 24 h, the cells that had penetrated to the bottom of the membrane were fixed with 4 % paraformaldehyde for 20 min, stained with crystal violet for 20 min, and photographed under a Nikon Eclipse TS100 microscope at 40× magnification. The numbers of cells that penetrated through the membrane were counted in 5 random fields for each well.

To examine the effect of MG on the invasive capacity of colon cancer cells, we performed a matrigel invasion chamber assay. This method is similar to the transwell migration assay already described, with the following differences: (1) the upper transwell chamber was pre-coated with diluted matrigel (1:9 dilution; BD Biosciences, 356234); (2) DLD-1 and SW480 cells were seeded in transwell chambers at the density of 1.2 × 10⁵ and 1.5 × 10⁵ cells per well, respectively; and (3) serum level in the lower chamber was increased to 20 % and the incubation time was extended to 48 h.

Glucose metabolism, lactate production, and intracellular ATP assays

Given that aerobic glycolysis is critical to the survival and proliferation of cancer cells [23], we examined whether MG alters carbohydrate metabolism in DLD-1 and SW480 cells by determining their glucose consumption and lactate production. Cells were seeded in Corning 12-well plates at 2 × 10⁵ cells per well and treated with various concentrations of MG (0, 0.5, or 1.0 mM), with or without AG (1.0 mM). Medium and cells were collected from the wells separately after 24 h incubation. Cells were counted by a Countstar Automated Cell Counter (Inno-Alliance Biotech). Concentrations of glucose and lactate in the medium and intracellular ATP production were detected by using a Glucose Assay kit (Nanjing JianCheng Bioengineering Institute, F006), a Lactic Acid Assay kit (Nanjing JianCheng Bioengineering Institute, A019-2), and an ATP Determination kit (Invitrogen, A22066), respectively, according to the manufacturers′ protocols. The glucose and lactate concentrations in medium and intracellular ATP production were determined with adjustment of cell number to eliminate the impact of MG on proliferation. After the data were normalized on the basis of cell number, they were expressed as the percentage relative to the control.

Western blot analysis

Specific antibodies and western blot analysis were used to detect changes in c-Myc expression. Following incubation with various concentrations of MG (0, 0.5, or 1.0 mM), with or without AG (1.0 mM), for 24 h, DLD-1 or SW480 cells were subjected to lysis in radio-immunoprecipitation assay lysis (Beyotime Institute of Biotechnology, P0013B) buffer containing protease inhibitor phenylmethanesulfonyl fluoride (Beyotime Institute of Biotechnology, ST506). Protein concentrations were measured by using the Bicinchoninic Acid Protein Assay kit (Beyotime Institute of Biotechnology, P0010). Briefly, an aliquot of 60 µg denatured protein was separated by sodium dodecyl sulfate 10 % -polyacrylamide gel electrophoresis with pre-stained molecular weight markers and transferred to polyvinylidene fluoride membranes. The blots were blocked with 5 % nonfat dried milk and incubated with primary antibodies overnight at 4°C and secondary antibodies for 1 h at room temperature. The signals were detected and quantified by densitometry using Image Lab Software (Bio-Rad). The antibodies used included anti-c-Myc monoclonal antibody (1:1000 dilution; Cell Signaling Technology, 5605), anti-β-actin monoclonal antibody (1:1000 dilution; Beyotime Institute of Biotechnology, AA128), and horseradish peroxidase conjugated anti-rabbit IgG (1:1000 dilution; Beyotime Institute of Biotechnology, A0208).

Reverse transcription and real-time polymerase chain reaction

Total mRNA was extracted from tumor cells using TRIzol reagent (Invitrogen, 3101). cDNA was synthesized by using the PrimeScript RT Reagent kit (Takara Biotechnology, RR037A). The expression of c-Myc was determined on a Bio-Rad real-time PCR detection system using SYBR Green Real-time PCR Master Mix (QIAGEN, Y5-204054). The sequences of primers were as follows: c-Myc: 5′-TCCGTCCTCGGATTCTCTGCTCT-3′ (sense); 5′-GCCTCCAGCAGAAGGTGA-3′ (antisense); β-actin: 5′-GAGACCTTCAACACCCCAGC-3′ (sense); 5′-CCACAGGATTCCATACCCAA-3′ (antisense) (all, Sangon Biotech). The thermal cycling conditions were: 1 cycle at 95°C for 5 min, followed by 40 cycles at 95°C for 10 s and 60°C for 30 s. Data for comparative analysis of c-Myc expression were obtained using the ΔΔCt method with β-actin as endogenous control.

Animal studies

To establish the human colon cancer xenograft model, 5-week-old male BALB/c nude mice (Shanghai Laboratory Animal Center) each received a subcutaneous injection of DLD-1 cells (8 × 10⁶ in 200 µL culture medium) into the left hind leg. After 7 days, the animals were randomly assigned into 3 treatment groups (n ≥ 5 mice/group): (1) intraperitoneal injection of physiological saline solution (as controls); (2) MG at a dose of 40 mg/kg body weight; and (3) MG at a dose of 80 mg/kg. Treatments were given once a day, 5 d per week. Tumor formation and body weight were assessed every 4 d. Tumor volume (cm³) was recorded by measuring 2 perpendicular diameters of the tumors with the following formula: volume = (length × width²) / 2.

After 16 d of treatment (23 d after cancer cell injection), the mice were sacrificed. The subcutaneous tumors were excised immediately and weighed. Tumor tissues were sectioned and taken for subsequent protein analyses. All animal care and treatments were performed in accordance with the guidelines of the Laboratory Animal Center of Wenzhou Medical University, and all experimental procedures were approved by the university's Laboratory Animal Ethics Committee.

Immunohistochemistry

Sectioned tumor tissues were fixed in 10 % formalin, embedded in paraffin and cut into 5 µm sections. Paraffin-embedded slides were deparaffinised, rehydrated and microwaved for 30 min in antigen retrieval solution (1:10 dilution; Beyotime Institute of Biotechnology, P0088) to retrieve antigen epitope. After blocking with 3 % H₂O₂ for 8 min and 5 % normal goat serum, the sections were incubated with anti-c-Myc monoclonal antibody (1:500 dilution; Abcam, AB32072) overnight at 4°C and biotinylated goat anti-rabbit IgG (Boster, SA1028) for 1 h at 37°C. The sections were then incubated with streptavidin-biotin-peroxidase complex (Boster, A0208) for 30 min at 37°C, followed by color development with diaminobenzidine and counterstaining with hematoxylin. All sections were photographed under a microscope at 200× or 400× magnification. Brown signal at nuclear level was considered as positive staining for c-Myc. To confirm staining specificity, negative controls without primary antibody were included.

Statistical analysis

All results are expressed as the mean value ± standard deviation of at least 3 experiments. Statistical analysis was performed with SPSS software version 17.0 (Chicago). Differences between groups were compared by an analysis of variance and Student t-test. P-values less than 0.05 were considered statistically significant.

Disclosure of potential confllicts of interest

No potential conflicts of interest were disclosed.

Funding

This investigation was supported by grants from the National Natural Science Foundation of China (81170257).