Volume 64, No. 3, 1999—JOURNAL OF FOOD SCIENCE 457 CHEMISTRY/BIOCHEMISTRY JOURNAL OF FOOD SCIENCE CHEMISTRY/BIOCHEMISTRY Hydroxyl Scavenging Activity of Glucose, Fructose, and Ribose-Lysine Model Maillard Products A. N. Wijewickreme, Z. Krejpcio, and D. D. Kitts Authors Wijewickreme and Kitts are with Food Science, Univ. of British Colum- bia, 6650 N.W. Marine Drive, Vancouver, BC, Canada. Author Krejpcio is with the Dept. of Food Hygiene & Human Nutrition, Agriculture Univ., Poznan, Poland. Address inquiries to Dr. D. D. Kitts. ABSTRACT Nondialyzable model Maillard reaction products (MRPs) were synthesized by reacting L-lysine with D-glucose, D-fructose, and D-ribose under different conditions of reaction time, tem- perature, pH, and water activity (a w ). Five experiments com- mon to all three models yielding greater than 0.5g of MRPs were assessed for antioxidant activity. All MRPs exhibited detectable, but variable, non site-specific hydroxyl radical ( • OH) scavenging activity (30–90%) in the deoxyribose as- say. MRPs derived from a ribose-lysine study showed the highest • OH scavenging activity (80–90%) in both deoxyri- bose and DNA nicking assays. All MRPs at higher concen- trations (0.2 mg/mL) decreased lipid peroxidation in linoleic acid emulsions. Relative effectiveness of different MRPs to scavenge free radicals can vary with reaction conditions and substrates. Key Words: Maillard products, free radicals, hydroxyl radi- cal, nondialyzable, non site-specific INTRODUCTION THE EFFECTS OF ACTIVE OXYGEN AND FREE RADICALS IN PRODUCING tissue damage in human disorders is becoming increasingly well rec- ognized (Halliwell et al., 1993). Active oxygen as superoxide (O 2 •- , hydrogen peroxide (H 2 O 2 ), or hydroxyl radical ( • OH) is a by-product of normal metabolism. It represents a potential toxic hazard to various biological molecules leading to tissue damage, cell injury or death (Yen and Chen, 1995; Yuan and Kitts, 1997). Formation of -OH or highly oxidizing species from H 2 O 2 have been associated with transi- tion metal ions such as iron and copper and reducing agents (at low concentrations) such as O 2 •- and ascorbate (Aruoma et al.,1987). The damage induced by free radicals is often prevented by scavengers. Some of these are glucose, mannitol, formate, thiourea, butan-1-ol, ethanol, plant phenols, -tocopherol, and -carotene (Byers and Peri, 1992; Guo et al., 1997). Maillard reaction products (MRPs) are well known to exhibit antioxidant activities in both model lipid (McGookin and Augustin, 1991; Wijewickreme and Kitts, 1997) and food (Bedinghaus and Ockerman, 1995; Wijewickreme and Kitts, 1998a; b) systems. Al- though many antioxidant activity measurements were associated with determining peroxyl radical scavenging activity of MRPs, very few studies have reported the affinity of different model MRPs to specif- ically scavenge • OH in vitro. Our objective was to assess the non site-specific • OH scavenging activity of 3 model MRPs (glucose- lysine, fructose-lysine, ribose-lysine) by employing both chemical (deoxyribose) and biological (DNA nicking) in vitro assays. Fur- ther, the effectiveness of MRPs to control lipid peroxidation in an aqueous medium was compared with a nonaqueous lipid (linoleic acid emulsion) system. MATERIALS & METHODS Materials All chemicals and reagents were of highest purity available. D- glucose, D-fructose, D-ribose, L-lysine, 2-butylatedhydroxy anisole (BHA), trichloroacetic acid (TCA), trizma base, ethidium bromide, bromophenol blue, xylene cyanole FF, ficoll, electrophoresis grade agarose, di-sodium ethylenediaminetetraacetic acid (2Na.EDTA), fer- ric chloride, ferrous sulphate, deoxyribose, L-ascorbic acid, 2-thiobar- bituric acid (2-TBA), sodium acetate, chelex 100, and pBR322 plas- mid DNA from Escherichia coli strain RRI were purchased from Sigma Chem. Co. (St. Louis, MO). Sodium dihydrogen orthophos- phate, di-sodium hydrogen orthophosphate, and sodium hydroxide were purchased from BDH Chem. Co. (Detroit, MI). Hydrogen per- oxide (H 2 O 2 ) was obtained from Fisher Scientific Co. (Fair Lawn, NJ). Deionized double distilled water and all buffers were further purified by treating with chelex-100. All glassware was submerged overnight in 2N HCl and rinsed thoroughly with deionized, distilled water before use. Model MRPs MRPs were synthesized according to the method described by Wijewickreme et al. (1997). L-lysine (800 mM) was heated with the following sugars: D-glucose, D-fructose, or D-ribose (800 mM) in a hot air oven (Blue M, Blue Island, IL) set at a prescribed temperature for a designated time period. Following heating, round bottom flasks containing brown solutions were rapidly cooled on ice, dialyzed (6-8 kD) against 20-30 changes of deionized distilled water for 7 days, and the non-dialyzable fraction was lyophilized. The prepared lyophilized powders were termed as Glu-Lys, Fru-Lys, and Rib-Lys MRPs. Five sets of conditions common to all 3 models yielding >0.5g of MRPs were assessed for antioxidant activity. Except Rib-Lys MRPs, the other MRPs of Glu-Lys and Fru-Lys reactions were identical to those studied in earlier reports (Wijewickreme and Kitts, 1997). Non site-specific -OH assay Non site-specific -OH radical scavenging activity of MRPs was measured according to the method given by Halliwell et al. (1987). Solutions of FeCl 3 and ascorbate were prepared in deaerated water immediately before use. Final reaction solution (1 mL) consisted of aliquots (0–200 L) of MRPs (0.2 mg/mL), FeCl 3 (100 mol), EDTA (100 mol), H 2 O 2 (1mmol), deoxyribose (3.6 mmol), and L-ascorbic acid (100 mol) in potassium phosphate buffer. The reaction mixture was incubated for 1h at 37°C and heated further in a boiling water bath for 15 min following the addition of 1 mL each of TCA (10%) and 2- TBA (0.5% 2-TBA in 0.025M NaOH containing 0.02% BHA). After cooling, color development was measured at 532 nm. All readings were corrected for any interference from brown color of the MRPs. DNA nicking assay All experiments were conducted in potassium phosphate buffer