Formation of a Hydroxymethylfurfural–Cysteine Adduct and Its

Article pubs.acs.org/JAFC

Cite This: J. Agric. Food Chem. 2017, 65, 9902-9908

Formation of a Hydroxymethylfurfural−Cysteine Adduct and Its Absorption and Cytotoxicity in Caco‑2 Cells Qianzhu Zhao, Yueyu Zou, Caihuan Huang, Ping Lan, Jie Zheng, and Shiyi Ou* Department of Food Science and Engineering, Jinan University, 510632 Guangzhou, Guangdong, China S Supporting Information *

ABSTRACT: Adducts of 5-hydroxymethylfurfural (HMF)-amino acids are formed during food processing and digestion; the elimination capacity of in vitro intestinal digests of biscuits, instant noodles, and potato crisps for HMF is 652, 727, and 540 μg/g, respectively. However, the safety of these adducts is unknown. In this study, an HMF−cysteine adduct named 1dicysteinethioacetal-5-hydroxymehtylfurfural (DCH), which was found to be produced in the gastrointestinal tract after HMF intake, was prepared to test its effect toward Caco-2 cells. Compared with HMF, the adduct displayed lower cytotoxicity against Caco-2 cells with an IC50 value of 31.26 mM versus 14.95 mM (HMF). The DCH did not induce cell apoptosis, whereas HMF significantly increased the apoptosis rate after incubation at concentrations of 16, 32, and 48 mM for 72 h. DCH showed an absorption rate considerably lower than that of HMF by Caco-2 cells. Lower absorption of DCH may result in lower toxicity compared with HMF against Caco-2 cells. Intracellular transformation of DCH has been observed. KEYWORDS: 5-hydroxymethylfurfural, adducts, Caco-2 cells, apoptosis

■

INTRODUCTION 5-Hydroxymethylfurfural (HMF) is an intermediate product in the Maillard reaction and caramelization. It contains a highly reactive aldehyde functional group. The reaction between sugars and amino acids in thermal food processes causes the formation of HMF, which can also be formed through acidic dehydration of carbohydrates.1,2 Cookies (0.5−74.5 mg/kg), breads (up to 410 mg/kg), coffee (300−1900 mg/kg), dried fruits (up to 2200 mg/kg), and caramel-containing foods (110−9500 mg/kg) are the most important sources of HMF.3,4 The average daily intake of HMF is estimated to be 1.6−150 mg/person (0.03−2.5 mg/kg of body weight).5 HMF has been reported to possess favorable biological effects, including antioxidative stress in humans,5 antihypoxic effect,6 in vitro antioxidant activity,7 inhibition of sickling, and lysis of red blood cells.8 However, HMF can be converted to the genotoxicant 5sulfoxymethyfurfural (SMF) by sulfotransferase, which is an enzyme distributed in almost all organs of animals and humans, especially in the liver, kidney, and colon.9 In children, HMF has been found to be effectively metabolized into SMF, which is highly electrophilic and can react with DNA and other macromolecules.10 Its mutagenic effects have been confirmed by the European Food Safety Authority.11 Thus, the adverse effect of high content of HMF in foods became a public concern. In order to find effective methods to reduce the content of HMF in foods, many reports on the formation of HMF and its influencing factors appear in the literature.12−17 Effects of amino acids on HMF formation and their influencing mechanisms have become a hot topic in recent years. Extensive research found that HMF produced in foods can react with amino acids to form various products, including adducts.18−24 These HMF−amino acid adducts could be even formed during digestion or in conditions similar to digestion.19,21,25 It was © 2017 American Chemical Society

speculated that these HMF−amino acid adducts were formed through Michael addition as one main mechanism.19,25 Free amino acids are ubiquitously distributed in foods, such as potatoes, wheat, and rice,26−28 and some amino acids, especially cysteine, glycine, and lysine, are widely used to inhibit acrylamide and HMF formation.23 Moreover, gastrointestinal digestion of protein can release large amounts of amino acids. Thus, HMF−amino acid adducts could be unavoidably produced during food processing and digestion. The toxicity and metabolism of these adducts remain unknown. In this research, the changes in HMF were investigated after simulated digestion of three foods, namely, instant noodles, potato crisps, and biscuits, and a high-purity HMF-cysteine adduct was prepared to test its effect on the growth of Caco-2 cell compared with HMF. To our knowledge, the present work is the first to investigate the toxicity of HMF adduct; the findings will help us systematically evaluate the toxicology of HMF−amino acid adducts.

■

MATERIALS AND METHODS

Reagents. 5-Hydroxymethyfurfural (HMF, 98%), L-cysteine (Cys, 99%), human salivary α-amylase (1500 U/mg), and porcine pepsin (15 000 U/mg) were purchased from Aladdin Reagents Database Inc. (Shanghai, China). Pancreatin (4000 U/mg) and bile salt were purchased from BoMei Biotechnology Co., Ltd. (Hefei, China). Trypsin−EDTA, penicillin−streptomycin, fetal bovine serum (FBS), and minimum essential medium ALPHA (α-MEM) were purchased from Thermo Fisher Scientific Inc. (Waltham, MA). Propidium iodide staining (PI) was purchased from Keygen BioTech Co., Ltd. (Nanjing, China). Received: Revised: Accepted: Published: 9902

August 24, 2017 October 19, 2017 October 23, 2017 October 23, 2017 DOI: 10.1021/acs.jafc.7b03938 J. Agric. Food Chem. 2017, 65, 9902−9908

Article

Journal of Agricultural and Food Chemistry Simulation of in Vitro Digestion. Master Kong’s fried instant noodles, Lay’s potato crisps, and Kraft soda biscuits were obtained from a local market in Guangzhou, China; the food samples were ground to pass 45-mesh sieve prior to digestion. The gastrointestinal digestion of foods was estimated using an in vitro digestion model designed by Minekus et al.,29 including simulated salivary fluid (SSF), simulated gastric fluid (SGF), and simulated intestinal fluid (SIF). Briefly, 5 g of ground food samples were placed in a conical flask, and 3.5 mL of SSF, 25 μL of CaCl2, 975 μL of distilled water, and 0.5 mL human salivary α-amylase solution (1500 U/mL) were thoroughly mixed. The mixture was incubated at 37 °C by shaking for 2 min to simulate the oral phase. Then, the oral digests were thoroughly mixed with 7.5 mL of SGF, 5 μL of CaCl2, 695 μL of distilled water, and 1.6 mL of porcine pepsin stock solution (25000 U); the pH was adjusted to 3.0 using 0.2 mL of 1 M HCl, and the mixture was incubated at 37 °C by shaking for 2 h to simulate gastric phase. After that, 16 mL of SIF, 40 μL of CaCl2, 860 μL of distilled water, 2.5 mL of bile salt solution, and 1.0 g of pancreatin solution (800 U/mL) were added and thoroughly mixed. The pH was adjusted to 7.0 using 0.6 mL of 1 M NaOH, and the mixture was incubated at 37 °C by shaking for 2 h to simulate the duodenal phase. HMF and free amino acids were determined at the end of simulated gastric and intestinal phases. In addition, 200 μg/mL HMF was added at the simulated duodenal phase to test the scavenging capacity of intestinal digestion for HMF. At the end of simulated gastric and intestinal phase, 5 mL of digests was centrifuged at 4000g for 20 min. The supernatant was extracted twice using 10 mL of petroleum to remove fat. An aliquot of 2 mL of samples in water phase was mixed with 2 mL of trichloroacetic acid (20%, w/v) and centrifuged at 10 000g for 15 min. A 1.0 mL sample (supernatant) was mixed with 1.0 mL of 0.2 mol/L sodium citrate buffer solution (pH 2.2), and then amino acids in the samples were analyzed with an L-8900 amino acid analyzer (Hitachi, Ibaraki, Japan) equipped with a visible detector. Free amino acids were separated on a sulfoacid cationic resin separation column (4.6 mm × 60 mm, 3 μm) (Hitachi Co., Ltd., Japan) at 57 °C. Each amino acid was derivatized using ninhydrin on a reaction column (4.6 mm × 40 mm) (Hitachi Co., Ltd., Japan) at 135 °C and then detected at 440 and 570 nm, respectively. The flow rates of MCI L-8500-PH-KIT buffer (Wako Pure Chemical Industries Ltd., Tokyo, Japan) and ninhydrin reagent were 0.4 and 0.35 mL/min, respectively. HMF in food and digestion mixture was extracted according to the method of Petisca et al.30 Before and after the in vitro simulation of digestion, all the samples were transferred into the centrifuge tubes with 20 mL of methanol and ultrasonic-extracted twice for 30 min. The combined extract was centrifuged at 10 000g for 20 min. Methanol and water in the extracts were removed by a rotary evaporator under vacuum at 45 °C. The residues were dissolved in 5.0 mL of deionized water and filtered through a 0.45 μm syringe filter into an autosampler vial. HMF was determined by a high-performance liquid chromatography (HPLC) method. Preparation of an Adduct of HMF with Cysteine. The adduct was prepared according to the preparation procedure of L-cysteine mercaptal of furfural by Obata and Mizutani31 with modifications. Namely, HMF (25 mmol) and cysteine (200 mmol) were dissolved in 60 mL of distilled water in a conical flask with the final pH adjusted to 3.0 using HCl (1 M) and then reacted at 90 °C for 9 h in a water bath installed with a magnetic stirrer under constant stirring at 150 rpm. The reacted mixture was cooled to 4 °C, kept for 12 h, and centrifuged at 10 000g for 20 min to remove the precipitated cysteine. The aimed product in the supernatant was separated and purified successively using macroporous resin HP-20 (from Sinopharm Chemical Reagent Co., Ltd., Shanghai, China, using distilled water as eluent), Octadecylsilyl-A-HG (from YMC Co., Ltd., Tokyo, Japan), and Sephadex LH-20 column (from GE Healthcare Co., Ltd., Uppsala, Sweden, using 5% methanol as eluent). The eluent was concentrated and lyophilized with a Scientz-10N vacuum freeze-dryer (from SCIENTZ Biotechnology Co., Ltd., Ningbo, China). The purity of the adduct was analyzed using HPLC (Shimadzu LC-20AT system, Kyoto, Japan), and the structure of the adduct, which was designated

as DCH (HMF-cysteine), was identified by HPLC/MS/MS spectrometry (4000Q-TRAP mass spectrometer, Applied Biosystem Sciex, United States) and nuclear magnetic resonance (NMR, Bruker 600 MHz Avance III NMR spectrometer). Effect of pH on Adduct Formation and HMF Elimination in HMF−Cysteine System. HMF can be eliminated by amino acids to convert to some adducts during food processing and digestion,19,32 and cysteine shows the highest activity in the elimination of HMF.19 To investigate the effect of cysteine on HMF elimination and DCH formation at different digestion conditions, a range of pH values was applied to test their reactivity. Specifically, the reaction mixture in a test tube containing 1 mL of cysteine (0.2 M in water) and 100 μL of HMF (0.4 M in water), with the pH adjusted to 1.09, 2.10, 3.06, 5.30, and 7.18 (monitored by a pH meter) using 0.1% HCl (v/v) and 0.4% NaOH (w/v). The reaction (with a final volume of 4 mL) was performed at 37 °C for 2.5 h by constant shaking in a water incubator at 150 rpm. The tubes were cooled to room temperature in an ice bath. The reaction products were diluted to 10 mL with distilled water and then filtered through a 0.45 μm syringe filter into an autosampler vial to determine the amount of HMF and DCH using HPLC. Caco-2 Cell Culture. Caco-2 cells resemble the enterocytes lining in the small intestine, which is a good model for studying the intestinal responses to environmental challenge.33 In the present research, Caco2 cells were used to evaluate the absorption of HMF and DCH and their toxicity in the small intestine. Caco-2 cells (ATCC HTB-37) from the American Type Culture Collection (Rockville, MD) were maintained in a humidified incubator at 37 °C containing 5% CO2 and 95% air, and cultivated in α-MEM with 20% of FBS, which was based on ATCC-formulated Eagle’s minimum essential medium. Cell densities were maintained as recommended, and 0.25% of the trypsin was used for detaching adherent cells from plates. Cell Cytotoxicity Assay. The cell counting kit-8 (CCK-8, from Biosharp, Hefei, China) was used to assess growth inhibition through colorimetric assay by using a water-soluble formazan dye. Caco-2 cells with a density of 2 × 104 cells/well per 100 μL of medium were seeded in 96-well plates and incubated in a humidified atmosphere of 5% CO2/95% air at 37 °C for 24 h to allow cells to sufficiently attach (more than 90%). Then, the cells were treated with different concentrations of HMF (0.5, 1, 2, 4, 8, 16, 32, 64, 128, and 256 mM) and DCH (8, 16, 32, 64, and 128 mM) at 37 °C for 24, 48, and 72 h. They were dissolved in serum-free culture medium and filtered through a 0.2 μm membrane before being added to the cell plates. After incubation, the medium containing compounds was discarded, and the well was washed by phosphate buffered saline (PBS, pH 7.4). Serum-free culture medium (100 μL) containing 10 μL of CCK-8 kit was added to each well, and the absorbance was measured at 450 nm using a microplate reader after incubation for 2 h. According to the IC50 values, cells were exposed to various concentrations of HMF (0, 4, 16, and 32 mM) or DCH (0, 16, 32, and 48 mM) for 72 h. Flow Cytometry. The cell cycle distribution and DNA fragmentation were shown by flow cytometric analysis using PI staining as recommended by Nicoletti et al.34 Caco-2 cells were seeded in 6-well plates at a density of 4 × 104 cells/well (2 mL/well). After 24 h, different concentrations of HMF (0, 4, 16, 32, 48, and 64 mM) and DCH (8, 16, 32, 64, and 128 mM) were added and incubated for 24− 72 h. The cells were collected and centrifuged at 10 000g for 5 min and thoroughly washed with PBS at room temperature. The collected cells were fixed overnight using 1 mL of 70% ethanol at −20 °C. Cells were subsequently washed and resuspended in DNA staining solution (50 μg/mL PI) at room temperature for 15 min in the dark. Analysis of cells was done using Canto II Flow cytometer (BD Biosciences, Franklin Lakes, NJ), and the data were analyzed with FlowJo V. 10 software (BD Biosciences, Franklin Lakes, NJ). Absorption of HMF and DCH by Caco-2 Cells. Cells were seeded in 12-well plates at a density of 1 × 105 cells/well (1 mL/well) and incubated in a humidified atmosphere of 5% CO2/95% air at 37 °C to allow for cell attachment. The cells were treated with 0.2 mM HMF and DCH for 4, 8, 12, and 24 h, or treated with 2 mM HMF and DCH for 0.5, 2, 4, 8, 12, and 24 h. At the end of the incubation period, 9903

DOI: 10.1021/acs.jafc.7b03938 J. Agric. Food Chem. 2017, 65, 9902−9908

Article

Journal of Agricultural and Food Chemistry the extracellular and intracellular fluids were collected according to the method of Sampath et al.35 The medium was collected, and the wells were washed once with 100 μL of PBS and collected in tube A. The cells were trypsinized (0.25%, 100 μL), washed once with 100 μL of PBS, collected in tube B, and centrifuged at 10 000g for 5 min. The supernatant was transferred into tube A. The cell pellet was washed thrice with 100 μL of PBS, and the washing solution was combined in tube A as extracellular fluid. The extracellular fluid in tube A was diluted to 1.5 mL with distilled water. For the intracellular fluid, the cell pellet in tube B was suspended in 50 μL of PBS, disrupted intermittently for 3 min by ultrasound (100 W, 2 s at a time and 18 s for intermittent time), and centrifuged at 12 000g for 20 min at 4 °C. The broken cell precipitation was washed thrice with 50 μL of PBS, and the washing solution was combined in tube B as intracellular fluid. The intracellular fluid in tube B was then diluted to 200 μL with distilled water. All samples were filtered through a 0.22 μm syringe filter into an autosampler vial for the determination of HMF, DCH, and SMF using HPLC. Determination of HMF, DCH, and SMF. Determination of HMF, DCH, and SMF36 was performed on an HPLC (Shimadzu LC20AT, Kyoto, Japan) equipped with a diode array detector and LCsolution software. The HPLC column used was an Agilent Zorbax SBAq C18 column (4.6 × 250 mm, 5 μm) (Agilent Technologies Co., Ltd.). The HPLC employed 97% ammonium formate buffer (30 mM, pH 2.44) (solvent A) and 3% acetonitrile (solvent B) as the mobile phase. The gradient program was as follows: 0−7.5 min, 0.6 mL/min; 7.5−7.6 min, 0.6−0.9 mL/min; 7.6−16.0 min, 0.9 mL/min; 16.0−16.1 min, 0.9−1.2 mL/min; 16.1−22.0 min, 1.2 mL/min. The injection volume was 5 μL, and the column temperature was set at 23 °C. HMF, DCH, and SMF were detected at 284, 240, and 277 nm, respectively, and quantified by calibration curves of external standards. The range of standard used for determination of HMF was 2.5, 10, 40, 160, 320, 640, 800, and 1280 μM; the range of standard used for determination of DCH was 6, 24, 60, 150, 300, 600, 960, and 1200 μM; and the range of standard used for determination of SMF was 0.125, 0.250, 0.500, 0.800, and 1.000 μg/mL. Statistical Analysis. All analyses were carried out in triplicate. Data were analyzed using SigmaPlot (SigmaPlot version 12.0 for windows) software and SPSS 22.0.0.1 (SPSS, Inc., Chicago, IL). Analysis of variance and Duncan’s multiple range method were used to compare the significant difference between treatments. Differences between the effects of different concentration of HMF and DCH on the viability of Caco-2 cells were compared by a one-way analysis of variance (ANOVA). S−N−K test for the population with equal variances and Tamhane’s test for that with unequal variances were employed to carry out the multiple comparisons of different treatments at p < 0.05.

Table 1. Changes in the Amount of HMF in the Digests of Instant Noodle, Potato Crisps, and Biscuit after Gastrointestinal Digestiona HMF content (μg/g) food sample

initial

gastric phase

intestinal phase

biscuit instant noodle potato crisp

5.2 ± 0.6 8.9 ± 0.1 7.7 ± 0.2

12.1 ± 0.5c 11.3 ± 0.1b 11.2 ± 0.3c

0c 2.0 ± 0.1c 0c

a Values are expressed as mean ± SD (n = 3). bSignificant differences compared with that in foods (p < 0.05). cSignificant differences compared with that in foods (p < 0.01).

further reactions with food components, such as nucleophilic amino and thiol groups of amino acids,9,19 and high pH was found to favor the reaction of HMF with cysteine, glycine, and lysine.19 Intestinal digestion of biscuits, instant noodles, and potato crisps released a number and amounts of free amino acids; the total amino acids and cysteine content amounted to 72.3, 75.7, and 36.8 mg/g, and 354, 245, 124 μg/g, respectively. The amino acids released after intestinal digestion shows high reactivity with HMF. They eliminated 652, 727, and 540 μg/g after the addition of 200 μg/mL HMF after simulated intestinal digestion system (35 mL in total volume). Adduct Identified in HMF−Cysteine Reaction System. The previous research found that cysteine showed highest reaction activity with HMF (the reaction could take place at 40 °C) and produced several possible adducts.19,24,25 To prevent the formation of side products (i.e., the Maillard reaction products), in the present study, the reaction between HMF and cysteine was conducted in acidic conditions. After successive separation on the columns of macroporous resin HP-20, Octadecylsilyl-A-HG, and Sephadex LH-20, a purified product was obtained. The molecular weight of the product was 350.3 (Figure S1). The structure of this adduct was identified by NMR. The 1H and 13C chemical shifts of purified adducts are listed in Table 2. Table 2. 1H NMR (500 MHz) and 13C NMR (125 MHz) Spectroscopic Data for Purified Adducts (Recorded in D2O) δH

no.

■

1 2 3 4 5 6 1′ 2′ 3′ 1″

RESULTS AND DISCUSSION Changes in HMF in Foods after in Vitro Gastrointestinal Digestion. Biscuits, fried instant noodles, and potato crisps, the major dietary sources of HMF, were selected as typical food samples to be used for in vitro multistep enzymatic digestion to investigate the fate of HMF during digestion. Table 1 shows that, after gastric digestion, HMF content in biscuits, instant noodles, and potato crisps increased by 132.7%, 27%, and 45.5%, respectively. However, after further digestion in simulated intestinal condition, HMF content in these foods sharply decreased; a 77.5% decrease in HMF was found in instant noodles, and no HMF was detected in potato crisps and biscuits (Table 1). The result confirmed the findings of Hamzalıoğlu and Gökmen9 in the three kinds of biscuits. They speculated that the precursors of HMF in foods could be converted to HMF under gastric (acidic) conditions.9 HMF is a multifunctional molecule containing a furan ring, a carbonyl group, and an allylic hydroxyl group that may undergo

2″ 3″

δC

5.40 (s) 6.42 (d, J = 3.2 Hz) 6.54 (d, J = 3.2 Hz) 4.58 (s) 3.20 (dd, J = 14.9, 7.6 Hz,) 3.09 (dd, J = 14.9, 7.6 Hz) 3.86 (ddd, J = 9.7, 7.6, 4.4 Hz) 3.29 (dd, J = 14.9, 4.4 Hz) 3.25 (dd, J = 14.9, 4.4 Hz) 3.86 (ddd, J = 9.7, 7.6, 4.4 Hz)

45.15 149.65 110.34 109.31 154.50 55.71 32.53 53.83 172.34 32.06 53.62 172.38

The 1H NMR spectrum showed the signals of two doublets at δH 6.54 (1H) and 6.42 (1H) with a coupling constant J = 3.2 Hz. These two signals belong to the original aromatic protons at the furan ring of HMF. The singlet at δH 4.58 (2H) indicated that the left part of HMF remains unchanged. However, the aldehyde proton signal of HMF disappeared, and a new proton at δH 5.40 (1H) was observed, which matched the dithioacetal 9904

DOI: 10.1021/acs.jafc.7b03938 J. Agric. Food Chem. 2017, 65, 9902−9908

Article

Journal of Agricultural and Food Chemistry

mechanism in simultaneous decrease in HMF and DCH needs to be further investigated. Because DCH can be formed at a wider pH range, it was chosen as the represented adduct with absorption and effect on the growth of Caco-2 cells tested by comparison with its parental contaminant, HMF. Effects of DCH and HMF on the Viability and Apoptosis of Caco-2 Cells. CCK-8 assay showed that Caco-2 cells exposed to HMF, and the adduct DCH undergoes decrease in cell viability in a dose- and time-dependent manner (Figure 3). At the concentration of 8 mM incubated for 24 h,

signal. In addition, two sets of cysteine protons were found at δH 3.86 (2H), 3.29 (1H), 3.25 (1H), 3.20 (1H), and 3.09 (1H), which suggested that this dithioacetal adduct was formed clearly by nucleophilic addition of two thiols onto the aldehyde of HMF. This proposed structure was then confirmed by 13C NMR spectrum, in which 12 carbons were observed, including the dithioacetal carbon at δC 45.15 as well as two cysteine carbon signals at δC 172.38, 172.34, 53.83, 53.62, 32.53, and 32.06. On the basis of the obtained NMR data, we assigned it as DCH. Its structure is shown in Figure 1. The molecular formula

Figure 1. Structure of the identified adduct, 1-dicysteinethioacetal-5hydroxymehtylfurfural (DCH).

is C12H18N2O6S2, with a theoretical m/z value of 350.40. This result was consistent with those derived from mass spectrometry, which gave [M + H]+• at 351.3 and a [M − H]+• at 349.3. Effect of Cysteine on HMF Elimination and DCH Formation at Different pH Conditions. Figure 2 shows that Figure 3. Effect of different concentrations of HMF and DCH on the viability of Caco-2 cells. Different letters indicate significant differences (p < 0.05) between different treatments at each incubation time (24, 48, and 72 h).

both HMF and DCH showed significant injury on the Caco-2 cell. After the concentration increased to 16 mM, the cell viability significantly decreased; only 65% and 57% remained viable when the concentration increased to 128 mM (Figure 3). With the elongation of culture time, DCH showed less toxicity than HMF against Caco-2 cell at all tested concentrations (Figure 3). The calculated IC50 values of HMF and DCH against Caco-2 cells were 14.95 ± 1.09 and 31.26 ± 1.50 mM, respectively, after 72 h of incubation, indicating a decrease in the toxicity of HMF against Caco-2 cell after formation of the adduct with cysteine. The effects of HMF and DCH on apoptosis were investigated by flow cytometry method. After incubation with HMF at concentrations of 16, 32, and 48 mM for 72 h, apoptosis peaks were observed at the sub-G0 phase, and the apoptosis rates were 11.6%, 14.7%, and 12.9%, respectively, which were considerably higher than that of the control (4.1%). The rate of cells in G0/G1 phase was reduced to 75.3%, 64.2%, and 58.2% at concentrations of 16, 32, and 48 mM, respectively, compared with 80% in the control. The rate of the cells in the S phase was greatly influenced only by HMF at 48 mM (Figure 4). However, after incubation with the same concentrations of DCH for 72 h, apoptosis peaks did not differ greatly compared with the control (Figure 4). All of these results proved that the formation of the adduct DCH between HMF and cysteine lowered the toxicity of HMF.

Figure 2. Effect of different pH values on HMF elimination and DCH formation in a HMF−cysteine reaction system. Error bars express the standard deviation.

HMF can react with cysteine at all tested pH values. The higher the pH value, the higher the amount of HMF eliminated. However, lower pH favors DCH formation; 3.2 times more DCH was formed at pH 1.09 compared with that at pH 7.18 (Figure 2). Higher reactivity of HMF with cysteine at higher pH indicated that other products were formed between HMF and cysteine, as previously reported.19 Another possibility is that DCH can further react with cysteine at higher pH. DCH contains two molecules of cysteine and one molecule of HMF; thus, it probably could release one molecule of cysteine to form HMF-cysteine adduct or further react with another molecule of cysteine through furan ring to form HMF-3cysteine adduct, or react with another molecule of HMF through the amino group of cysteine to form 2HMF-2cysteine. The underlying 9905

DOI: 10.1021/acs.jafc.7b03938 J. Agric. Food Chem. 2017, 65, 9902−9908

Article

Journal of Agricultural and Food Chemistry

Figure 4. Cells treated for 72 h with different doses of HMF and DCH.

Table 3. Extracellular and Intracellular Levels of HMF and DCH after Incubation of Caco-2 Cells with HMF and DCH at Different Timesa HMF content (μmol/well) treatment

incubation time (h)

extracellular

0.2 mM HMF

4 8 12 24 4 8 12 24

0.18 ± 0.01 0.17 ± 0.01 0.13 ± 0.01 0.12 ± 0.01 1.77 ± 0.03 1.78 ± 0.02 1.58 ± 0.15 1.47 ± 0.05 (×10−2) 0.45 ± 0.03 0.86 ± 0.32 0.57 ± 0.03 0.59 ± 0.02 (×10−2) 5.33 ± 1.69 4.99 ± 1.91 4.52 ± 1.27 6.94 ± 0.36

2 mM HMF

0.2 mM DCH

4 8 12 24

2 mM DCH 4 8 12 24 a

DCH content (μmol/well)

intracellular (×10−3)

extracellular

intracellular (×10−3)

1.23 1.14 0.86 1.04 7.05 5.65 6.47 7.18

± ± ± ± ± ± ± ±

0.13 0.26 0.04 0.16 0.42 1.50 1.79 3.68

− − − − − − − −

− − − − − − − −

0.70 0.68 0.69 0.71

± ± ± ±

0.01 0.01 0.01 0.01

0.19 0.18 0.17 0.16

± ± ± ±

0.01 0.01 0.01 0.01

4.50 4.47 4.56 4.55

± ± ± ±

0.04 0.01 0.07 0.03

0.89 0.91 0.86 1.03

± ± ± ±

0.02 0.03 0.07 0.15

1.81 1.80 1.83 1.84

± ± ± ±

0.05 0.05 0.04 0.05

6.38 7.60 6.72 7.83

± ± ± ±

0.17 0.25 0.38 1.23

Values are expressed as means ± SD (n = 3).

considerably less than HMF. As shown in Table 3, 80% and 92% of added DCH were left in the extracellular fluids after incubation of Caco-2 cells with 0.2 and 2 mM DCH for 24 h. Moreover, the amount of HMF transformed from DCH in extracellular and intracellular fluids accounted for 2.95% and 0.36% of the added DCH, respectively, at 0.2 mM, and 3.49% and 0.05% of the added DCH at 2 mM (Table 3). These results suggested that 82.95% (80% + 2.95%) and 95.49% (92% + 3.49%) of added DCH have not been absorbed by Caco-cells; only 17.05% and 4.51% of DCH have been absorbed by Cacocells after incubation of 0.2 and 2 mM DCH for 24 h, respectively. The absorption rate of DCH sharply lowered compared with that of HMF (40% and 36.5%). The intracellular DCH accounts for 2.30% and 0.39%, which are considerably lower than the absorbed rates (17.05% and

However, the toxicity of DCH at high concentrations should not be disregarded. Absorption of HMF and DCH in Caco-2 Cells. The extracellular HMF at two levels of HMF treatments decreased as the incubation time prolonged; 60% and 73.5% of added HMF were left in the extracellular fluids after incubation of Caco-2 cells with 0.2 and 2 mM HMF for 24 h (Table 3), indicating that 40% and 36.5% of the added HMF has been absorbed by Caco-2 cells incubated with two concentrations of HMF. However, the intracellular HMF accounts for only 0.52% and 0.36% after incubation of Caco-2 cells with 0.2 and 2 mM HMF for 24 h, respectively (Table 3). The possible reason for the great difference between the absorbed amount of HMF and intracellular amount of HMF is that the absorbed HMF has been metabolized in Caco-2 cells.37 Absorbed DCH was 9906

DOI: 10.1021/acs.jafc.7b03938 J. Agric. Food Chem. 2017, 65, 9902−9908

Article

Journal of Agricultural and Food Chemistry

cysteine; DCH, 1-dicysteinethioacetal-5-hydroxymehtylfurfural; EDTA, ethylene diamine tetraacetic acid; EMEM, Eagle’s minimum essential medium; FBS, fetal bovine serum; HMF, 5hydroxymethylfurfural; HPLC, high-performance liquid chromatography; IC50, half maximal inhibitory concentration; LC, liquid chromatography; α-MEM, minimum essential medium alpha; MS/MS, tandem mass spectrometry; NMR, nuclear magnetic resonance; PBS, phosphate buffered saline; PI, propidium iodide; Pro, proline; SD, standard deviation; SGF, simulated gastric fluid; SIF, simulated intestinal fluid; SMF, 5sulfoxymethyfurfural; SSF, simulated salivary fluid

4.51%) of the added DCH after incubation of Caco-2 cells with 0.2 and 2 mM DCH for 24 h (Table 3), respectively. This suggested that DCH can be possibly metabolized to other compounds in Caco-2 cells. Because cytochrome P450 has been discovered in Caco-2 cells38 and reported to be responsible for transformation of furfuryl mercaptan (a compound similar to DCH) to furfuryl methyl sulfoxide or furoic acid,39 the possible metabolic pathway of DCH is to form hydroxymethylfurfuryl methyl sulfoxide or hydroxymethyl-2furoic acid. As shown in Table 3, a trace amount of HMF was also detected in the intracellular fluids, which is possibly transformed from HMF. Because of the high stability of thioacetals, they are often used to protect carbonyl compounds in organic synthesis. However, its deprotection to the parent carbonyl compounds is not easy.40 The deprotection can be undertaken only through the addition of oxidizing reagent (i.e., oxone, benzyltriphenylphosphonium peroxymonosulfate, etc.) or catalysis by Bronsted acids.40,41 These transformation conditions do not occur in Caco-2 cells; whether some enzymes, such as oxidases, hydrolases are responsible for the transformation of DCH to HMF still needs to be investigated. SMF has not been detected both in extracellular and intracellular fluids after incubation with HMF and DCH, possibly because of lack of sulfotransferase activity in Caco-2 cells or because of the limits of our detection method for SMF. The adducts formed between HMF and amino acids are produced in thermally processed foods or in the digestive tract after intake of HMF along with proteins. As known, HMF has been reported as a risky compound through metabolism especially having toxicity at high concentration, but there was no literature about metabolism or toxicity of HMF−amino acid adducts up to now. The present research proved that the adduct DCH was less absorbed in and less toxic against Caco-2 cells. However, DCH and other kinds of adducts that formed between HMF and amino acids still need to be fully evaluated, including their absorption in the different sections in the gastrointestinal tract, their pharmaceutical kinetics, and their effects on tissues and organs.

■

■

(1) Capuano, E.; Fogliano, V. Acrylamide and 5-hydroxymethylfurfural (HMF): A review on metabolism, toxicity, occurrence in food and mitigation strategies. LWT-Food Sci. Technol. 2011, 44 (4), 793− 810. (2) Ames, J. M. The Maillard reaction. In Biochemistry of Food Proteins, 1st ed.; Hudson, B. J. F., Ed.; Springer: Boston, MA, 1992; pp 99−153. (3) Murkovic, M.; Pichler, N. Analysis of 5-hydroxymethylfurfual in coffee, dried fruits and urine. Mol. Nutr. Food Res. 2006, 50 (9), 842− 846. (4) Ameur, L.; Mathieu, O.; Lalanne, V.; Trystram, G.; Birlouezaragon, I. Comparison of the effects of sucrose and hexose on furfural formation and browning in cookies baked at different temperatures. Food Chem. 2007, 101 (4), 1407−1416. (5) Mousavi, R.; Alizadeh, M.; Saleh-Ghadimi, S. Consumption of 5hydroxymethylfurfural-rich dried fruits is associated with reduction in urinary excretion of 8-hydroxy-2′-deoxyguanosine: a randomized clinical trial. Eur. Food Res. Technol. 2016, 242 (5), 677−684. (6) Zhao, L.; Chen, J. P.; Su, J. Y.; Li, L.; Hu, S. Q.; Li, B.; Zhang, X.; Xu, Z. B.; Chen, T. F. In vitro antioxidant and antiproliferative activities of 5-hydroxymethylfurfural. J. Agric. Food Chem. 2013, 61 (44), 10604−10611. (7) Li, M. M.; Wu, L. Y.; Zhao, T.; Xiong, L.; Huang, X.; Liu, Z. H.; Fan, X. L.; Xiao, C. R.; Gao, Y.; Ma, Y. B.; et al. The protective role of 5-HMF against hypoxic injury. Cell Stress Chaperones 2011, 16 (3), 267−273. (8) Janzowski, C.; Glaab, V.; Samimi, E.; Schlatter, J.; Eisenbrand, G. 5-Hydroxymethylfurfural: assessment of mutagenicity, DNA-damaging potential and reactivity towards cellular glutathione. Food Chem. Toxicol. 2000, 38 (9), 801−809. (9) Monien, B. H.; Frank, H.; Seidel, A.; Glatt, H. Conversion of the common food constituent 5-hydroxymethylfurfural into a mutagenic and carcinogenic sulfuric acid ester in the mouse in vivo. Chem. Res. Toxicol. 2009, 22 (6), 1123−1128. (10) Pastoriza de la Cueva, S. P.; Á lvarez, J.; Végvári, Á .; MontillaGómez, J.; Cruz-López, O.; Delgado-Andrade, C.; Rufián-Henares, J. A. Relationship between HMF intake and SMF formation in vivo: An animal and human study. Mol. Nutr. Food Res. 2017, 61 (3), 1600773. (11) Kavousi, P.; Mirhosseini, H.; Ghazali, H.; Ariffin, A. A. Formation and reduction of 5-hydroxymethylfurfural at frying temperature in model system as a function of amino acid and sugar composition. Food Chem. 2015, 182, 164−170. (12) Gökmen, V.; Açar, Ö . Ç .; Serpen, A.; Morales, F. J. Effect of leavening agents and sugars on the formation of hydroxymethylfurfural in cookies during baking. Eur. Food Res. Technol. 2008, 226 (5), 1031− 1037. (13) Kocadağlı, T.; Gökmen, V. Multiresponse kinetic modelling of Maillard reaction and caramelisation in a heated glucose/wheat flour system. Food Chem. 2016, 211, 892−902. (14) Mogol, B. A.; Gökmen, V. Effect of chitosan on the formation of acrylamide and hydroxymethylfurfural in model, biscuit and crust systems. Food Funct. 2016, 7 (8), 3431−3436. (15) Kocadağlı, T.; Gökmen, V. Effect of sodium chloride on αdicarbonyl compound and 5-hydroxymethyl-2-furfural formations from

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jafc.7b03938. HPLC, MS, MS/MS, 1H NMR, and 13C NMR data (PDF)

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +86-13640210646. Fax: +8620-85226630. ORCID

Shiyi Ou: 0000-0002-6779-0858 Funding

The present work was funded by the National Nature Science Foundation of China (31671957). Notes

The authors declare no competing financial interest.

■

ABBREVIATIONS USED ANOVA, analysis of variance; ATCC, American type culture collection; Car, carnosine; CCK-8, cell counting kit-8; Cys, 9907

DOI: 10.1021/acs.jafc.7b03938 J. Agric. Food Chem. 2017, 65, 9902−9908

Article

Journal of Agricultural and Food Chemistry

(32) Gökmen, V.; Açar, Ö . Ç .; Serpen, A.; Morales, F. J. Effect of leavening agents and sugars on the formation of hydroxymethylfurfural in cookies during baking. Eur. Food Res. Technol. 2008, 226 (5), 1031− 1037. (33) Moyes, S. M.; Morris, J. F.; Carr, K. E. Culture conditions and treatments affect Caco-2 characteristics and particle uptake. Int. J. Pharm. (Amsterdam, Neth.) 2010, 387 (1−2), 7−18. (34) Nicoletti, I.; Migliorati, G.; Pagliacci, M.; Grignani, F.; Riccardi, C. A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. J. Immunol. Methods 1991, 139 (2), 271−279. (35) Sampath, C.; Zhu, Y.; Sang, S.; Ahmedna, M. Bioactive compounds isolated from apple, tea, and ginger protect against dicarbonyl induced stress in cultured human retinal epithelial cells. Phytomedicine 2016, 23 (2), 200−213. (36) Lu, Y. L.; Zhou, N. L.; Liao, S. Y.; Su, N.; He, D. X.; Tian, Q. Q.; Chen, B.; Yao, S. Z. Detection of adulteration of anti-hypertension dietary supplements and traditional Chinese medicines with synthetic drugs using LC/MS. Food Addit. Contam., Part A 2010, 27 (7), 893− 902. (37) Delgado-Andrade, C.; Seiquer, I.; Morales, F. J. Estimation of hydroxymethylfurfural availability in breakfast cereals. Studies in Caco2 cells. Food Chem. Toxicol. 2008, 46 (5), 1600−1607. (38) Hochman, J. H.; Chiba, M.; Yamazaki, M.; Tang, C. Y.; Lin, J. H. P-glycoprotein-Mediated Efflux of Indinavir Metabolites in Caco-2 Cells Expressing Cytochrome P450 3A4. J. Pharmacol. Exp. Ther. 2001, 298 (1), 323−330. (39) Lake, B. G.; Price, R. J.; Walters, D. G.; Phillips, J. C.; Young, P. J.; Adams, T. B. Studies on the metabolism of the thiofurans furfuryl mercaptan and 2-methyl-3-furanthiol in rat liver. Food Chem. Toxicol. 2003, 41 (12), 1761−1770. (40) Hajipour, A. R.; Mallakpour, S. E.; Mohammadpoor-Baltork, I.; Adibi, H. Solid State Deprotection of Acetals and Thioacetals Usin Benzyltriphenylphosphonium Peroxymonosulfate. Molecules 2002, 7 (8), 674−680. (41) Satchell, D. P. N.; Satchell, R. S. Mechanisms of hydrolysis of thioacetals. Chem. Soc. Rev. 1990, 19, 55−81.

glucose under caramelization conditions: a multiresponse kinetic modeling approach. J. Agric. Food Chem. 2016, 64 (32), 6333−6342. (16) Kocadağlı, T.; Gö kmen, V. Effects of sodium chloride, potassium chloride, and calcium chloride on the formation of αdicarbonyl compounds and furfurals and the development of browning in cookies during baking. J. Agric. Food Chem. 2016, 64 (41), 7838− 7848. (17) Rannou, C.; Laroque, D.; Renault, E.; Prost, C.; Sérot, T. Mitigation strategies of acrylamide, furans, heterocyclic amines and browning during the Maillard reaction in foods. Food Res. Int. 2016, 90, 154−176. (18) Cai, Y.; Zhang, Z. H.; Jiang, S. S.; Yu, M.; Huang, C. H.; Qiu, R. X.; Zou, Y. Y.; Zhang, Q. R.; Ou, S. Y.; Zhou, H.; et al. Chlorogenic acid increased acrylamide formation through promotion of HMF formation and 3-aminopropionamide deamination. J. Hazard. Mater. 2014, 268, 1−5. (19) Zou, Y. Y.; Pei, K. H.; Peng, X. C.; Bai, W. B.; Huang, C. H.; Ou, S. Y. Possible adducts formed between hydroxymethylfurfural and selected amino acids, and their release in simulated gastric model. Int. J. Food Sci. Technol. 2016, 51 (4), 1002−1009. (20) Gökmen, V.; Kocadağlı, T.; Göncüoğlu, N.; Mogol, B. A. Model studies on the role of 5-hydroxymethyl-2-furfural in acrylamide formation from asparagine. Food Chem. 2012, 132 (1), 168−174. (21) Hamzalıoğlu, A.; Gökmen, V. Investigations on the reactions of α-dicarbonyl compounds with amino acids and proteins during in vitro digestion of biscuits. Food Funct. 2016, 7 (6), 2544−2550. (22) Zhang, Z. H.; Zou, Y. Y.; Wu, T. G.; Huang, C. H.; Pei, K. H.; Zhang, G. W.; Lin, X. H.; Bai, W. B.; Ou, S. Y. Chlorogenic acid increased 5-hydroxymethylfurfural formation when heating fructose alone or with aspartic acid at two pH levels. Food Chem. 2016, 190, 832−835. (23) Zou, Y. Y.; Huang, C. H.; Pei, K. H.; Cai, Y.; Zhang, G. W.; Hu, C. Y.; Ou, S. Y. Cysteine alone or in combination with glycine simultaneously reduced the contents of acrylamide and hydroxymethylfurfural. LWT-Food Sci. Technol. 2015, 63 (1), 275−280. (24) Hamzalıoğlu, A.; Gö kmen, V. Investigation and kinetic evaluation of the reactions of hydroxymethylfurfural with amino and thiol groups of amino acids. Food Chem. 2018, 240 (1), 354−360. (25) Hamzalıoğlu, A.; Gökmen, V. Formation and elimination reactions of 5-hydroxymethylfurfural during in vitro digestion of biscuits. Food Res. Int. 2017, 99 (1), 308−314. (26) Elmore, J. S.; Koutsidis, G.; Dodson, A. T.; Mottram, D. S.; Wedzicha, B. L. Measurement of acrylamide and its precursors in potato, wheat, and rye model systems. J. Agric. Food Chem. 2005, 53 (4), 1286−1293. (27) Mustafa, A.; Åman, P.; Andersson, R.; Kamal-Eldin, A. Analysis of free amino acids in cereal products. Food Chem. 2007, 105 (1), 317−324. (28) Saikusa, T.; Horino, T.; Mori, Y. Distribution of free amino acids in the rice kernel and kernel fractions and the effect of water soaking on the distribution. J. Agric. Food Chem. 1994, 42 (5), 1122− 1125. (29) Minekus, M.; Alminger, M.; Alvito, P.; Ballance, S.; Bohn, T.; Bourlieu, C.; Carriere, F.; Boutrou, R.; Corredig, M.; Dupont, D.; Dufour, C.; Egger, L.; Golding, M.; Karakaya, S.; Kirkhus, B.; Le Feunteun, S.; Lesmes, U.; Macierzanka, A.; Mackie, A.; Marze, S.; McClements, D. J.; Menard, O.; Recio, I.; Santos, C. N.; Singh, R. P.; Vegarud, G. E.; Wickham, M. S.; Weitschies, W.; Brodkorb, A. A standardised static in vitro digestion method suitable for food - an international consensus. Food Funct. 2014, 5 (6), 1113−1124. (30) Petisca, C.; Henriques, A. R.; Pérez-Palacios, T.; Pinho, O.; Ferreira, I. M. P. L. V. O. Assessment of hydroxymethylfurfural and furfural in commercial bakery products. J. Food Compos. Anal. 2014, 33 (1), 20−25. (31) Obata, Y.; Mizutani, J. Studies on the degradation mechanisms of proteins and their derivatives in the foods. Part IV the decompositions of several mercaptals and mercaptoles of cysteine in aqueous solutions. Bull. Agric. Chem. Soc. Jpn. 1960, 24 (6), 562−564. 9908

DOI: 10.1021/acs.jafc.7b03938 J. Agric. Food Chem. 2017, 65, 9902−9908