RESEARCH ARTICLE

Antitumor effect of fructose 1,6-bisphosphateand its mechanism in hepatocellular carcinoma cells

Yin-Xiang Lu & Xi-Can Yu & Mei-Ying Zhu

Received: 16 August 2013 /Accepted: 17 September 2013 /Published online: 1 October 2013# International Society of Oncology and BioMarkers (ISOBM) 2013

Abstract We aimed to investigate the antitumor effect andmechanism of fructose 1,6-bisphosphate (F1,6BP) in a hepa-tocellular carcinoma cell line. HepG2 cells were treated withdifferent concentrations of F1,6BP alone or in combinationwith antioxidant N -acetyl-L -cysteine (NAC) or catalase(CAT), and cell proliferation assays were performed. Nuclearmorphology was observed by fluorescence microscopy afterHoechst staining, and apoptosis was measured with flow cy-tometry. Changes in reactive oxygen species (ROS) levels inHepG2 cells were detected by 2′,7′-dichlorodihydrofluoresceindiacetate (DCFH-DA) staining. A colorimetric assay wasadopted to determine the percentage of oxidized glutathionein these cells. CAT and glutathione peroxidase (GSH-Px)mRNA expression levels in HepG2 cells were measured byreal-time fluorescence quantitative PCR. HepG2 cell prolifer-ation was significantly inhibited by F1,6BP, accompanied byan increase in intracellular ROS levels and oxidized glutathi-one. Upregulated apoptosis and characteristic nuclear morpho-logical changes were observed, and the expression of CATandGSH-Px mRNA was increased after F1,6BP treatment. Theantitumor effect of F1,6BP was significantly decreased afterpretreatment with NAC and CAT in HepG2 cells. In conclu-sion, F1,6BP can induce the apoptosis of HepG2 cells. Themechanism involved may be associated with the generation ofROS, especially the production of H2O2.

Keywords Fructose 1,6-bisphosphate . Hepatocellularcarcinoma . ROS . Apoptosis

Introduction

Hepatocellular carcinoma (HCC) is one of the most commoncancers worldwide. The incidence is increasing year by yearnot only in Asia but also in the USA, Europe, and Japan as aresult of hepatitis virus infection, alcoholic liver disease, and/or genetic problems. Approximately 70 % of patients arediagnosed with advanced stage disease, and its prognosis isvery poor [1–3]. Curative therapeutic options are limited toearly stage disease and mainly consist of resection ororthotopic liver transplantation if cirrhosis is evident [4].Thus, new therapeutic molecular targets and treatment areurgently required for hepatocellular carcinoma.

In recent years, a series of powerful experimental studieshave supported the targeting of the glycolytic pathway as agood tumor therapeutic strategy [5, 6]. Fructose 1,6-bisphosphate (F1,6BP) is a sugar used in metabolism-relatedmedicine that can shift the metabolism of glucose from glycol-ysis to the pentose phosphate pathway to increase the level ofreduced glutathione [7]. However, its antitumor effects havenot been previously investigated. In this study, we observed theeffect of F1,6BP treatment on hepatocellular carcinoma cellproliferation, apoptosis, and reactive oxygen species (ROS)levels and explored its mechanism of action.

Material and methods

Cell culture

The hepatocellular carcinoma cell line HepG2 was purchasedfrom ATCC (Manassas, VA, USA) and cultured in Dulbecco's

Y.<X. Lu (*) :X.<C. Yu :M.<Y. ZhuCancer Center, Xinchang People’s Hospital, Zhejiang, Chinae-mail: YinXiang_Lu@126.com

Y.-X. Lue-mail: luyx@sxmail.net

Y.<X. Lu :X.<C. Yu :M.<Y. ZhuClinical Laboratory, Xinchang People’s Hospital, Zhejiang, China

Tumor Biol. (2014) 35:1679–1685DOI 10.1007/s13277-013-1231-z

modified Eagle's medium (DMEM; Gibco, USA) supplement-ed with 10 % heat-inactivated fetal bovine serum (FBS; Gibco,USA) at 37 °C in a 5 % CO2 atmosphere.

Effect of F1,6BP on HepG2 cell proliferation

CellTiter 96 AQueous Non-Radioactive Cell Proliferation As-says was performed to test cell proliferation ability. Briefly,HepG2 cells were harvested during the logarithmic growthphase, seeded into 96-well flat-bottom plates at a density of5×104 cell/mL in 100 μL/well DMEM containing 10 % FBSand incubated for 24 h. Medium was replaced with 100 μL/well of fresh medium containing varying concentrations of F1,6BP (0, 2.5, 5, and 10 mmol/L),and plates were incubated at37 °C at 5 % CO2 atmosphere for 0, 1, 2, and 3 days. Subse-quently, 20 μL CellTiter 96 AQueous Non-Radioactive CellProliferation Assay Reagent (Promega, USA) was added toeach well. After 1–4 h of additional incubation, the absorptionvalues at 490 nm were determined with an automatic enzyme-linked immunosorbent assay plate reader (BMG POLARstarOmega, BMG Labtech, Germany).

Analysis of antioxidant effect on the proliferative abilityof HepG2 cells treated with F1,6BP

HepG2 cells were pretreatedwithN-acetyl-L-cysteine (NAC) orcatalase (CAT) (2.5 mmol/L and 100 U/mL, respectively) anti-oxidants for 2 h and then cultured in growth medium containingF1,6BP (0, 2.5, 5, and 10mmol/L) for 24 h. HepG2 proliferationwas measured with the CellTiter 96 AQueous Non-RadioactiveCell Proliferation Assay method, as described above.

Determination of ROS levels

2′,7′-Dichlorodihydrofluorescein diacetate (DCFH-DA;Invitrogen, USA) is a dye that becomes fluorescent after oxi-dation and is used to determine intracellular ROS levels.HepG2 cells were harvested in the logarithmic phase andseeded into 24-well flat-bottom plates at a density of 5×104

cell/well and then cultured in growth medium containing 0, 2.5,5, or 10 mmol/L F1,6BP for 48 h. The supernatant was thendiscarded and replaced with 200 μL of 10 μmol/L DCFH-DAdiluted in growth medium, and cells were cultured in the darkfor 1 h. Plates were then washed three times with growthmedium, and fluorescence was measured with a microplatereader at an excitation wavelength of 488 nm and an emissionwavelength of 525 nm. The experiment was performedindependently for three times.

Determination of glutathione levels

HepG2 cells were seeded into six-well plates at a density of 1×105/well, and 0, 2.5, 5, or 10 mmol/L F1,6BP was added,

followed by incubation for 48 h. Cells were then harvested andwashed for three times in phosphate buffer saline (PBS), andprotein removal agent from the Glutathione/Oxidized Glutathi-one Detection Kit (Biyuntian Biotechnology Research Institute,Jiangsu, China) was added. Samples then underwent three cyclesof rapid freeze-thawing in liquid nitrogen and a water bath set at37 °C, before centrifugation at 10,000×g for 10 min at 4 °C andincubation in a 4 °C ice bath for 5 min. Working reagent wasadded to 10 μL supernatant and incubated for 25 min at roomtemperature, and then, the OD value was measured at 412 nmusing amicroplate reader, and the ratio of oxidized glutathione tototal glutathione was analyzed.

Analysis of CAT and glutathione peroxidase mRNA levels

HepG2 cells were treated for 48 h with 0, 2.5, 5, or 10 mmol/LF1,6BP, and total RNAwas extracted with TRIzol (Invitrogen).Then, 500 ng of RNAwas reversely transcribed to cDNA, andCATand glutathione peroxidase (GSH-Px) mRNA levels wereexamined with real-time quantitative PCR kits (Takara, Japan).The primers used in this experiment were synthesized by ShengGong Company (Shanghai, China) and are listed in Table 1.Data were analyzed by designating the mRNA number of thetarget genes in the no medicine control group as 1 and bycalculating the relative transcript level of the target genes inthe experimental group using the 2−ΔΔCT method.

Nuclear morphological changes observed by fluorescencemicroscopy

HepG2 cells were seeded into six-well plates at a density of 4×103cells/well. Cells were then preincubated with NAC(2.5 mmol/L) or CAT (100 U/mL) for 2 h under normalconditions, and then, 5 mmol/L F1,6BP was added. Afterincubation for 24 h, the cells were washed twice in PBS andfixed with 1 mL/well of 4 % paraformaldehyde at 4 °C for10 min. Plates were washed with PBS three times, and cellswere stained with Hoechst 33342 in the dark for 10 min atroom temperature, before three further washes with PBS. Cellswere then immediately observed under an inverted fluores-cence microscope; live cell nuclei showed dispersion anduniform fluorescence, while dead cells were not stained withHoechst. Following apoptosis, the nuclei underwent significantmorphological changes, and blue fluorescent-stained compactparticulates could be seen in the nucleus or cytoplasm.

Flow cytometry analysis of apoptosis induced by F1,6BP

HepG2 cells were seeded into six-well plates at a density of 3×105 cells/well and incubated for 24 h. Cells were pretreated withNAC (2.5 mmol/L) or CAT (100 U/mL) for 2 h under normalconditions, and then, F1,6BP (5 mmol/L) was added and incu-bated for 24 h. Cells were harvested after digestion with

1680 Tumor Biol. (2014) 35:1679–1685

pancreatic enzymes and washed three times with PBS and thendouble-stained with phycoerythrin-conjugated annexin V and7-AAD according to the manufacturer's instructions (BD Bio-science, San Jose, CA, USA) and analyzed using flow cytom-etry. Viable cells were labeled as Annexin V (−) and 7-AAD (−)and located in the left lower quadrant. Early apoptosis cellswere labeled as annexin V (+) and 7-AAD (−) and located inthe upper left quadrant. Dead cells, which were labeled asannexin V (−) and 7-AAD (+), were located in the lower rightquadrant. Late apoptotic and necrotic cells, which were labeledas annexin V (+) and 7-AAD (+), were located in the rightupper quadrant. Twenty thousand events were collected foreach sample using the Becton Dickinson FACScan (BD Bio-science). Data were analyzed using the FlowJo software (TreeStar, San Carlos, CA, USA).

Statistical analysis

Statistical analysis was performed using the SPSS 17.0software. Significance differences between groups weredetermined by Student's t test. Results were expressed asmean ± SEM. A P value of less than 0.05 was consideredstatistically significant (P <0.05).

Results

F1,6BP inhibited the proliferation of HepG2 cells

Results shown in Fig. 1 and Table 2 indicate that differentconcentrations of F1,6BP could inhibit the proliferation of

HepG2 cells at 24, 48, or 72 h in a time- and dose-dependent manner (P <0.01).

F1,6BP induced ROS production in HepG2 cells

ROS levels in HepG2 cells changed markedly after the cellswere treated with F1,6BP and were dose dependent. Themean fluorescence intensities of ROS were 65.84±4.37 and79.92±5.47 at 5 or 10 mmol/L F1,6BP treatment, respec-tively, and were significantly increased compared with thecontrol (39.18±2.19) (P <0.01; Fig. 2).

F1,6BP upregulated the percentage of oxidized glutathione

The ratio of oxidized glutathione to total glutathione wasupregulated in a dose-dependent manner in HepG2 cellsfollowing treatment with F1,6BP for 48 h. The ratios in thecells treated with 5 and 10 mmol/L F1,6BP were 10.3±0.74and 13.02±0.61 %, respectively, which were significantlyhigher than those of the control group (6.1±0.31 %)(P <0.01; Fig. 3).

Antioxidants weaken the anti-HepG2 effects of F1,6BP

Data showed that the cell proliferation rate was lower inthe F1,6BP-treated group than the groups pretreatedwith NAC (nonspecifically eliminate ROS) or CAT(specifically eliminate H2O2) for 2 h. These resultsrevealed that the inhibitory effects of F1,6BP on HepG2cell proliferation was closely related with ROS levelsand was mediated by the upregulation of the H2O2

levels in the cells (Fig. 4).

Table 1 The primer sequences of CAT and GSH-Px

Gene Forward primer Reverse primer Product length (bp) Annealing temperature (°C)

CAT TGAGGTTGAACAGATAGC CACAGGTATATGAAGATAATTG 145 53.1

GSH-Px CCAGTCGGTGTATGCCTTC CAGAGGGACGCCACATTC 108 58.3

β-actin CAGGGCGTGATGGTGGGCA CAAACATCATCTGGGTCATTCTC 253 56.0

**

**

*

*

*

*

*

Fig. 1 Inhibition rates of different concentrations of F1,6BP inHepG2 cells (n =3). *P <0.01, compared with negative control group

Table 2 The inhibition rates of different concentrations of F1,6BP onHepG2 cells (n =3)

The concentration ofF1,6BP (mmol/L)

The inhibition rate (%)

24 h 48 h 72 h

0 0 0 0

2.5 29.77±2.31** 41.83±3.61** 54.27±2.69**

5 46.21±3.47** 66.72±5.13** 77.41±5.14**

10 69.89±4.12** 82.14±4.22** 87.88±4.80**

**P<0.01, compared with negative control group

Tumor Biol. (2014) 35:1679–1685 1681

Effect of F1,6BP on CAT and GSH-Px mRNA levels

The data showed that mRNA levels of CAT and GSH-Px inHepG2 cells were upregulated after the cells were treated withF1,6BP for 48 h (P <0.01). Notably, CAT mRNA levels werethree times higher than the control group (Fig. 5).

Effect of F1,6BP on cell apoptosis

As shown in Fig. 6a, the normal cell nuclei of HepG2 cellswere uniform, round, or oval. After F1,6BP treatment (Fig. 6b),more apoptotic cells were evident. Nuclear morphologicalchange characteristics of apoptosis such as karyorrhexis spher-ical particles, chromatin condensation, particle shape distribu-tion, pyknosis, and lobulated nuclear fragmentation were ob-served. Fewer apoptotic cells in the groups pretreated withNAC or CAT were recorded, compared with those treatedwithF1,6BP alone (Fig. 6c, d). This was confirmed by flowcytometry analysis. The apoptotic rate of the cells treated withF1,6BP alone was 41.0±2.15 % (Fig. 7b), which was higherthan that of HepG2 cells cultured under normal conditions(7.12±1.04 %; Fig. 7a) or with NAC (11.3±1.82 %) and

CAT (11.5±1.19 %) pretreatment (Fig. 7c, d). Thus, antioxi-dants such as NAC and CATcan relieve the apoptosis-inducingeffects of F1,6BP.

Discussion

The production of ROS is an inevitable consequence of aerobicmetabolism. In healthy cells, ROS generated during respirationare retained by the mitochondria and reduced by a defensivesystem that includes nonenzymatic and enzymatic antioxidantsand protective enzymes such as superoxide dismutase, catalase,and glutathione peroxidase. However, some stimulators canresult in the accumulation of ROS in the cytoplasm and leadto an oxidative stress burden on the cell. Oxidative stress hasbeen implicated as a factor in many diseases including cancer[8]. Catalase (CAT) and glutathione peroxidase (GSH-Px) en-zymes are widely used as markers of oxidative stress [9]. ROSinclude free radicals, such as the superoxide anion (O2

−), hy-droxyl radicals (OH), and the nonradical hydrogen peroxide(H2O2). Normally, ROS levels are in a dynamic equilibriumthrough enzymatic and nonenzymatic antioxidant defense sys-tems. Previous studies revealed that the ROS levels in tumorcells are higher compared with normal cells, and the antioxidant

**

**

*

Fig. 2 Effect of varying concentrations of F1,6BP on ROS levels inHepG2 cells. Cells were treated for 48 h (n=3). *P <0.05, **P <0.01

**

**

*

Fig. 3 The percentage of oxidized glutathione in total glutathione inHepG2 cells which were treated with different concentrations of F1,6BP for 48 h, n =3. *P <0.05; **P <0.01, compared with 0 mmol/LF1,6BP-treated group

**

****

**

****

Fig. 4 The effects of different types of antioxidants on the prolifera-tion ability of HepG2 which were treated with F1,6BP, n =3. **P <0.01, compared with F1,6BP-treated alone

**

**

**

**

**

**

Fig. 5 The effects of F1,6BP on the expression levels of CAT andGSH-Px mRNA in HepG2 cell, n =3. **P <0.01, compared with0 mmol/L F1,6BP-treated group

1682 Tumor Biol. (2014) 35:1679–1685

Fig. 6 The morphology of thecell nucleus in hepaticcarcinoma cell HepG2, whichwas treated with F1,6BP aloneor together with NAC or CAT,was observed underfluorescence microscope afterHoechst staining (×400). a Thenormal cell nuclei of HepG2cells. b The cell nuclei ofHepG2 cells after F1,6BPtreatment. c , d The cell nucleiof HepG2 cells in the groupspretreated with NAC or CAT

Fig. 7 Analysis of apoptosisrate with flow cytometry aftercells treated with 5 mmol/L F1,6BP alone or together withNAC or CAT for 24 h. a Theflow cytometry results of normalHepG2 cells. b The apoptoticrate of HepG2 cells after F1,6BPtreatment. c , d The flowcytometry analysis of HepG2cells in the groups pretreatedwith NAC or CAT

Tumor Biol. (2014) 35:1679–1685 1683

system is relatively weak [10–12]. High levels of ROS in tumorcells promote the survival, development, and migration oftumor cells through the induction of DNA damage, leading tothe mutation of oncogenes, activation of inflammation factors,and stabilization of hypoxia-inducible factors. Conversely, tu-mor cells must limit the high levels of ROS to a certainthreshold; otherwise, further increases in ROS levels will in-duce tumor cell apoptosis and necrosis [13, 14]. This studyshowed that the products of glucose metabolism function asROS-scavenging agents in addition to their involvement inenergy metabolism and maintaining balanced redox. Severalresearchers speculated that the enhancement of sugar metabo-lism levels in tumor cells compensates for high ROS levels, toensure these ROS can be kept at levels that are advantageous totumor survival [15].

F1,6BP is an important intermediate in the process of glu-cose metabolism. It can competitively inhibit the glucose up-take of tumor cells. F1,6BP increases the flux of glucose intothe pentose phosphate pathway, and it has anti-inflammatory,endotoxemia-preventative, macrophage-activating, and liverinjury functions [7, 16] and has been widely used as a thera-peutic agent for various harmful conditions in a variety oftissues [17]. This study discovered that F1,6BP could inhibitthe proliferative ability of hepatocellular carcinoma HepG2cells and could induce apoptosis in parallel with the upregula-tion of ROS levels. Glutathione is important for clearing ROSfrom cells, and it can protect enzymes with sulfhydryl toprevent destruction by oxidation. When ROS levels are upreg-ulated, reduced glutathione is oxidized and changed into oxi-dized glutathione, and the ratio of oxidized glutathione to totalglutathione also rises. In this study, following treatment ofHepG2 cells with F1,6BP for 48 h, the ratio of oxidizedglutathione to total glutathione was upregulated, and thischange correlated with ROS levels. The results revealed thatthe balance of redox state in the cell was destroyed, leading tothe apoptosis of HepG2 cells.

H2O2 is an important component of ROS, and it canproduce OH in mitochondria and can trigger cell damage[18]. The generation and degradation of H2O2 is regulatedby many types of enzymes, such as CAT and GSH-Px. In thisstudy, CATand GSH-Px mRNA levels were upregulated afterHepG2 cells were treated with F1,6BP for 48 h. The highexpression of antioxidant enzymes is the result of feedbackprotection by the cells themselves, which is induced by highROS levels. Because CAT is a specific scavenging agent ofH2O2, and GSH-Px can eliminate H2O2, we deduce that theupregulation of H2O2 is the key reason for F1,6BP inhibitionof HepG2 cell proliferation. To further understand therelationship between the antitumor function of F1,6BP andROS, ROS-scavenging agents NAC or CATwere used togeth-er with F1,6BP, and their effect on the antitumor function ofF1,6BP was observed. The results revealed that the prolifera-tion rates of HepG2 cells were higher following NAC or CAT

pretreatment than those treated with F1,6BP alone, and theCAT-treated group exhibited the most dramatic increase.Moreover, the apoptotic rate of HepG2 cells was downregu-lated by NAC or CAT. These results confirmed that F1,6BPexerted its antitumor function by inducing the production ofH2O2 in HepG2 cells.

In conclusion, this study confirmed that F1,6BP couldinhibit HepG2 cell proliferation by inducing ROS produc-tion, particularly H2O2, leading to apoptosis. This studyprovides an experimental basis for the clinical applicationof drugs targeting sugar metabolism.

Conflicts of interest None

References

1. Han DJ, Kim JB, Park SY, et al. Growth inhibition of hepatocel-lular carcinoma Huh7 cells by Lactobacillus casei extract. YonseiMed J. 2013;54:1186–93.

2. Jihye C, Jinsil S, Lee IJ, et al. Feasibility of sorafenib combinedwith local radiotherapy in advanced hepatocellular carcinoma.Yonsei Med J. 2013;54(5):1178–85.

3. Park KW, Park JW, Choi JI, et al. Survival analysis of 904patients with hepatocellular carcinoma in a hepatitis B virus-endemic area. J Gastroenterol Hepatol. 2008;23:467–73.

4. Nel I, Baba HA, Ertle J, et al. Individual profiling of circulatingtumor cell composition and therapeutic outcome in patients withhepatocellular carcinoma. Transl Oncol. 2013;6(4):420–8.

5. De Raedt T, Walton Z, Yecies JL, et al. Exploiting cancer cellvulnerabilities to develop a combination therapy for ras-driventumors. Cancer Cell. 2011;20(3):400–13.

6. Mauro C, Leow SC, Anso E, et al. NF-κB controls energyhomeostasis and metabolic adaptation by up-regulating mitochon-drial respiration. Nat Cell Biol. 2011;13(10):1272–9.

7. Lian XY, Khan FA, Stringer JL. Fructose-1,6-bisphosphate hasanticonvulsant activity in models of acute seizures in adult rats. JNeurosci. 2007;27(44):12007–11.

8. Gourlay CW, Ayscough KR. Identification of an upstream regu-latory pathway controlling actin-mediated apoptosis in yeast. JCell Sci. 2005;118(Pt 10):2119–32.

9. Debnath T, Park SR, da Kim H, et al. Anti-oxidant and anti-inflammatory activities of inonotus obliquus and germinatedbrown rice extracts. Molecules. 2013;18(8):9293–304.

10. Schumacker PT. Reactive oxygen species in cancer cells: liveby the sword, die by the sword. Cancer Cell. 2006;10(3):175–6.

11. Trachootham D, Zhou Y, Zhang H, et al. Selective killing ofoncogenically transformed cells through a ROS-mediated mecha-nism by beta-phenylethyl isothiocyanate. Cancer Cell. 2006;10(3):241–52.

12. Chen W, Zhao Z, Li L, et al. Hispolon induces apoptosis in humangastric cancer cells through a ROS-mediated mitochondrial path-way. Free Radio Biol Med. 2008;45(1):60–72.

13. Kim AD, Kang KA, Kim HS, et al. A ginseng metabolite, com-pound K, induces artophagy and apoptosis via generation ofreactive oxygen species and activation of JNK in human coloncancer cells. Cell Death Dis. 2013;4:e750.

14. Rasul A, Di J, Millimouno FM, et al. Reactive oxygen speciesmediate isoalantolactone-induced apoptosis in human prostatecancer cells. Molecules. 2013;18(8):9382–96.

1684 Tumor Biol. (2014) 35:1679–1685

15. Perera RM, Bardeesy N. Cancer: when antioxidants are bad.Nature. 2011;475(7354):43–4.

16. Cuesta E, Boada J, Calafell R, et al. Fructose 1,6-bisphosphateprevented endotoxemia, macrophage activation, and liver injuryinduced by D-galactosamine in rats. Crit Care Med. 2006;34(3):807–14.

17. Alva N, Cruz D, Sanchez S, et al. Nitric oxide as a mediator offructose 1,6-bisphosphate protection in galactosamine-inducedhepatotoxicity in rats. Nitric Oxide. 2013;28:17–23.

18. Burlacu A, Jinga V, Gafencu AV, et al. Severity of oxidative stressgenerates different mechanisms of endothelial cell death. CellTissue Res. 2001;306(3):409–16.

Tumor Biol. (2014) 35:1679–1685 1685