Potential Protective Effect of l-Cysteine against the Toxicity of

Jul 23, 2014 - Biology Department, Jacksonville State University, Jacksonville, Alabama 36265, ... profiling in risk assessment, Jackson et al.28 asse...
0 downloads 0 Views 5MB Size
Article pubs.acs.org/JAFC

Potential Protective Effect of L‑Cysteine against the Toxicity of Acrylamide and Furan in Exposed Xenopus laevis Embryos: An Interaction Study John Russell Williams,† James R. Rayburn,*,† George R. Cline,† Roger Sauterer,† and Mendel Friedman§ †

Biology Department, Jacksonville State University, Jacksonville, Alabama 36265, United States Produce Safety and Microbiology Research, Western Regional Research Center, ARS, U.S. Department of Agriculture, Albany, California 94710, United States

§

ABSTRACT: The embryo toxicities of two food-processing-induced toxic compounds, acrylamide and furan, with and without added L-cysteine were examined individually and in mixtures using the frog embryo teratogenesis assay-Xenopus (FETAX). The following measures of developmental toxicity were used: (a) 96 h LC50, the median concentration causing 50% embryo lethality; (b) 96 h EC50, the median concentration causing 50% malformations of the surviving embryos; and (c) teratogenic index (96 h LC50/96 h EC50), an estimate of teratogenic risk. Calculations of toxic units (TU) were used to assess possible antagonism, synergism, or response addition of several mixtures. The evaluated compounds demonstrated counterintuitive effects. Furan had lower than expected toxicity in Xenopus embryos and, unlike acrylamide, does not seem to be teratogenic. However, the short duration of the tests may not show the full effects of furan if it is truly primarily genotoxic and carcinogenic. L-Cysteine showed unexpected properties in the delay of hatching of the embryos. The results from the interaction studies between combination of two or three components (acrylamide plus L-cysteine; furan plus L-cysteine; acrylamide plus furan; acrylamide plus furan and Lcysteine) show that furan and acrylamide seem to have less than response addition at 1:1 toxic unit ratio in lethality. Acrylamide and L-cysteine show severe antagonism even at low 19 acrylamide/1 L-cysteine TU ratios. Data from the mixture of acrylamide, furan, and L-cysteine show a slight antagonism, less than would have been expected from binary mixture exposures. Bioalkylation mechanisms and their prevention are discussed. There is a need to study the toxicological properties of mixtures of acrylamide and furan concurrently formed in heat-processed food. KEYWORDS: acrylamide, furan, Xenopus laevis, embryo, FETAX, L-cysteine, L-cysteine adducts, teratogenicity, interactions, additive response, antagonism, synergism, isobole diagram, mechanisms, binary mixtures, ternary combinations, toxic units, dietary significance, food safety



INTRODUCTION Reports that potentially toxic acrylamide and furan are present in foods (bread, coffee) formed during their processing (baking, frying, grilling) under conditions that also induce the formation of Maillard browning products,1−13 heightened interest in the chemistry, biochemistry, and safety of these reactive molecules. In vivo, acrylamide is reported to act as a carcinogen, neurotoxin, and reproductive toxin,14−19 whereas the cited references indicate that furan is reported to induce hepatocarcinogenicity in rodents, so it is considered to be a presumptive carcinogen. We previously reviewed the chemistry of formation and distribution of acrylamide in a large number of widely consumed foods.16 The highest amounts (in μg/kg) were present in cooked wheat cereals (738), potato chips (466), graham crackers (459), canned olives (414), French fries (413), roasted almonds (320), toast (213), and popcorn (180). Mariotti et al.20 found that ascorbic acid (vitamin C), which is highly susceptible to heat-induced degradation,21,22 enhanced the formation of furan during frying and baking of starchy food and that the furan levels linearly correlated with the degree of nonenzymatic browning of baked wheat flour samples. With respect to processing, model studies with glucose and free amino acids showed that roasting produced 25−100 times greater amounts of furan than that produced by pressure cooking.23 A © 2014 American Chemical Society

survey of furan content in 40 heat-processed foods showed that the highest amounts (in μg/g) were present in vegetable beef soup (0.0880), baked beans (0.0842), freshly brewed coffee (0.0517), soy sauce (0.0511), and freshly brewed decaffeinated coffee (0.0418). Surprisingly, the values for instant coffee (0.0015) and instant decaffeinated coffees (0.0045) were quite low.5 On the basis of the data, the author estimates a mean exposure to furan for the average U.S. consumer of ∼0.2 μg/kg/ day. A Brazilian study reported that furan levels (in μg/kg) in 14 cereal products ranged from 5.0 (cake) to 105.3 (wafer) and were quite high in roasted ground coffee, ranging from 448.9 (instant) to 4164.3 (strong).9 Consumption studies in Belgium show that infants have a higher risk of dietary furan ingestion compared with children and adults.24−27 On the basis of the results of a hepatic global gene expression profiling in risk assessment, Jackson et al.28 assembled a simplified adverse outcome for risk assessment that indicates chronic exposure to >2 mg/kg body weight favors chronic regenerative proliferation, suggesting how furan causes cancer. Received: Revised: Accepted: Published: 7927

March 20, 2014 July 9, 2014 July 14, 2014 July 23, 2014 dx.doi.org/10.1021/jf5013743 | J. Agric. Food Chem. 2014, 62, 7927−7938

Journal of Agricultural and Food Chemistry

Article

Table 1. 96 h LC50, 96 h EC50 (Malformation), Teratogenic Index (TI; LC50/EC50), Mortality−Malformation No Observable Effects Concentration (NOEC), Lowest Observable Effect Level (LOEC), and Minimum Concentration To Inhibit Growth (MCIG) for Acrylamide, Furan, and L-Cysteinea mortality clutch

total embryos

LC50 (mmol/L)

EC50 (mmol/L)

April 1, 2013

280

April 22, 2013

280

June 10, 2013

280

Aug 2, 201313

280

1.879 (1.692−2.099) 2.136 (2.011−2.263) 1.466 (1.306−1.641) 2.080 (1.843−2.381) 1.890

0.702 (0.625−0.792) 1.161 (1.007−1.356) 0.780 (0.674−0.917) 1.134 (0.992−1.314) 0.944

53.94 (49.51−59.36) 38.81 (36.11−41.70) 37.92 (34.60−41.50) 43.56

44.41 (39.13−51.82) 27.30 (24.25−31.14) 47.20 (41.04−56.60) 39.64

average Aug 6, 2012

400

Aug 13, 2012

400

Sept 20, 2012

400

average

malformation

NOEC (mmol/L)

LOEC

NOEC (mmol/L)

LOEC

MCIG (mmol/L)

2.0

2.5

1.0

2.0

0.5

1.8

2.0

2.5

1.0

2.0

1

1.9

2.0

2.5

0.5

1.0

0.5

1.8

2.0

2.5

0.5

1.0

0.5

TI Acrylamide 2.7

2.0 Furan 1.2

20

30

20

30

30

1.4

30

40

20

30

30

0.8

40

50

20

30

30

54

90

27

54

27

42

64

85

64

42

27

33

33

1.1 L-Cysteine

Aug 6, 2012

760

Aug 13, 2012

760

Sept 10, 2012

760

average a

42.40 (37.04−49.83) 49.01 (44.49−55.15) 34.14 (30.08−39.48) 41.85

39.71 (33.96−50.04) 40.46 (34.49−51.13) 35.68 (33.06−41.17) 38.62

1.1 1.2

n.c.

1.0

33

1.1

Numbers in parentheses are the 95% Fieller bounds. Numbers for acrylamide are from interaction experiments.



On the basis of the available data, Moro et al.10 concluded that both genotoxicity and chronic cytotoxicity seem to contribute to furan-induced tumor formation.8,29−31 Reported efforts to reduce or prevent the formation of furan in foods include the use of polyphenolic compounds and plant extracts32 and modification of food-processing conditions.33,34 In a previous study, we reported that the amino acid L-cysteine and the tripeptide glutathione protected frog embryos against acrylamide-induced malformation (birth defects) of fetuses and that N-acetyl-L-cysteine protected against acrylamide-induced mortality but not against malformations.35 In related studies, we also reported that other food ingredients protected embryos against potato glycoalkaloid-induced teratogenesis.36−38 Because acrylamide and furan can be formed concurrently during heat processing of numerous foods, the objectives of this study were (a) to define possible cumulative adverse and beneficial effects using the frog embryo teratogenesis assayXenopus (FETAX) of acrylamide, furan, and L-cysteine individually and the combinations of L-cysteine and acrylamide, acrylamide and furan, furan and L-cysteine, and acrylamide, furan, and L-cysteine; (b) to analyze mathematically and graphically the interactions of these compounds in the ex vivo environment that might optimize the protection by L-cysteine against adverse effects; and (c) to depict the effect of exposure to the food ingredients on the appearance of the frog embryos.

MATERIALS AND METHODS

Materials. The following compounds were obtained from Sigma (St. Louis, MO, USA): L-cysteine (catalog no. C7352); acrylamide (catalog no. A3553); human chorionic gonadotropin (catalog no. CG5); MS-222 (catalog no. E10521); calcium sulfate (catalog no. C-3771); magnesium sulfate (catalog no. M-7506); sodium bicarbonate (catalog no. S8875); potassium chloride (catalog no. P-5405); calcium chloride (catalog no. C-1016); and sodium chloride (catalog no. S7653). Furan was obtained from Aldrich-Sigma (St. Louis, MO, USA; catalog no. 185922). Animal Care and Husbandry. Xenopus frogs were purchased from Xenopus I, Inc. (Dexter, MI, USA) and housed in a glass aquaria recirculation system with 2−4 frogs per 10 gallons of dechlorinated tap water. Human chorionic gonadotropin (250−600 IU) was injected in the dorsal lymph sac of both male and female frogs to induce breeding. Breeding pairs were placed in false-bottom breeding chambers, and embryos were collected the next morning. The jelly coat was removed by swirling the embryos for 1−3 min in a 2% L-cysteine solution (pH 8.1). Embryos were then rinsed and placed into sorting dishes. Embryos were double sorted into test dishes of 20 embryos per 8 mL of test solution in plastic disposable 60 × 15 mm Petri dishes. The negative control was FETAX solution (0.625 g/L NaCl, 0.086 g/L NaHCO3, 0.03 g/L KCl, 0.015 g/L CaCl2, 0.006 g/L CaSO4·2H2O, and 0.075 g/L MgSO4 in distilled ASTM type I water).39 Frogs were not bred more than once every 2 months. FETAX Assay Procedures. FETAX was used to determine acrylamide toxicity and teratogenesis according to the American Society for Testing Materials (ASTM) Guide for FETAX and our earlier studies with teratogenic glycoalkaloids.40−47 Stock solutions of acrylamide and of test compounds were made in FETAX solution. Appropriate dilutions were made to achieve final concentrations. Each day of the 4 day test, new solutions were placed into 60 mm covered plastic Petri dishes with 7928

dx.doi.org/10.1021/jf5013743 | J. Agric. Food Chem. 2014, 62, 7927−7938

Journal of Agricultural and Food Chemistry

Article

various concentrations of test compounds dissolved in FETAX solution. At 24, 48, and 72 h, all solutions were prepared and replaced in the dishes. Dead embryos were removed, and live embryos counted. The embryos were cultured in the dark at 24 °C. At 96 h, surviving (stage 46) embryos were identified to verify embryos had reached the proper stage of development. Stage 46 embryos possess hind-limb buds and tightly coiled guts but do not yet feed. At the end of 96 h, the mortality was determined and embryos were anesthetized with MS-222. Malformed survivors, dead embryos, and the developmental stage were observed under a dissecting microscope and recorded according to the Atlas of Malformations.48 Embryos were photographed with an Olympus Stylus 720sw on macro setting for growth measurements, and detailed photographs were taken of selected embryos using Pro Reg digital camera attached to a Nikon dissection microscope. As a measure of growth, head−tail length was measured by following body contour using a computer equipped with digitizing software Image Pro Plus. Malformations were determined by observing embryos under a dissecting microscope for abnormal survivors, developmental stage, and dead embryos. Xenopus laevis embryos are transparent and allow for easy determination of gut, heart, head, axial, and other internal and external malformations. The malformations are quantitatively scored on a standard score sheet and tallied. The data were recorded according to American Society of Testing Material (ASTM) guidelines.40 The total number of abnormal tadpoles divided by the total number of living tadpoles multiplied by 100 is the percentage malformed from each dish. The 96 h LC50 (concentration causing 50% lethality) and 96 h EC50 (concentration inducing malformation or gross malformations in 50% of surviving embryos) were calculated using probit analysis from Systat 13 software. The teratogenic index (TI) was calculated as the 96 h LC50 divided by the 96 h EC50 (malformation). The minimum concentration to inhibit growth (MCIG) was calculated using ANOVA with the post hoc Bonferroni test using Systat 13 software. Individual Toxicity Experimental Design and Analysis (Acrylamide, Furan, and L-Cysteine). For concentration−response studies of toxic effects on embryos, four negative controls (4 replicates of 20 embryos in 60 × 15 mm Petri dishes) and various test concentrations (2 replicates of 20 embryos in 60 × 15 mm Petri dishes each) of chemical ranging 0.5 to 80 mM were tested. Each experiment was repeated three times. Each experiment used between 280 and 480 embryos. Previously tested acrylamide34 had the fewest concentrations, as detailed experiments have previously been carried out as indicated by the total number of embryos for each test (Table 1). The furan and L-cysteine data were independent definitive assays as these chemicals had not been tested in the FETAX assay. The acrylamide experiments were the acrylamide positive controls for the interaction experiments. All data were analyzed according to FETAX methods above. Interaction Experiments of Binary and Ternary Mixtures. On the basis of data from the individual experiments above, binary and trinary mixtures were calculated such that estimated ratios for mixtures could be created. For acrylamide/L-cysteine, the ratios attempted were 0TU:1TU, 1TU:0TU, 1TU:1TU, 1TU:3TU, 3TU:1TU, and 19TU:1TU. For furan/acrylamide and furan/L-cysteine, the ratios used were 0TU:1TU, 1TU:0TU, and 1TU:1TU. For acrylamide/furan/ L-cysteine, the ratios used were 0TU:0TU:1TU, 0TU:1TU:0TU, 1TU:0TU:0TU, and 10TU:4TU:1TU. Once appropriate mixtures were made, they were tested with the standard FETAX to determine the 96 h LC50, EC50, and TI for each of the interaction ratios. For each clutch of embryos single-chemical toxicity measurements were taken to determine the clutch-specific 96 h LC50 for each individual experiment. Each mixture study was repeated two times. Toxic Unit Conversion. A toxic unit (TU) is defined as the 96 h LC50 = 1 TU for mortality. Toxic unit conversion allowed for the two compounds to be normalized to toxicity and, thus, a correlation could be studied for mixtures. A hypothetical example calculation is shown below:

96 h LC50 for a mixture containing a ratio of 1 mmol/L X : 2 mmol/L Y is (i.e., 1 TU:1 TU) 0.5 mmol of X : 1 mmol of Y

TU of X = [0.5 mmol/L (amount of X in mixture)]/ [ 1 mmol/L (LC50 of X )] = 0.5 TU of Y = [1 mmol/L (amount of Y in mixture)] /[2 mmol/L (LC50 of Y)] = 0.5 The toxic units add (0.5 + 0.5 =1) and would indicate concentration additive response. The TU values for the each component of the mixture were plotted on an isobole diagram to demonstrate the interaction of the two compounds. Generally, such diagrams are divided into three major regions relating to the types of interactions possible: synergism, additive effects, and antagonism. The area of the graph below the diagonal line connecting 1 TU of axis X to 1 TU of axis Y shows increasing synergism as the points approach the origin. Concentration addition is indicated by the diagonal line connecting 1 TU of axis X to 1 TU of axis Y. Points found between the lines connecting 1 TU to 1 TU and within the squares formed by these two points from their respective axis show response additive interactions. Points outside this box indicate antagonistic interactions.



RESULTS Controls. Each experimental trial contained 4 replicates of 20 embryos as controls. These 80 embryos represented the genetic effects of the clutch absent exposure to the compound of study. A 96 h control mortality >10% represents a poor-quality clutch. All control groups had