Comparisons between Orange- and Green- fleshed Non-netted and Orange-fleshed Netted Muskmelons: Antioxidant Changes following Different Harvest and Storage Periods

The consumption of netted muskmelons (Cucunt.is izzelo L. Reticulatus group) has raised health concerns due to pathogenic bacteria attaching to sites on the netted rind inaccessible to sanitation. The purpose of this study was to compare I) the enzymic and nonenzymic antioxidant capacity between representative cultivars of netted muskmelon and both greenand orange-fleshed honey dew muskmelons during storage for 17 days and 2) levels of non-nutrient phytochcniicals between these genotypes in consideration of ultimately substituting netted orange-fleshed with nonnetted orange-fleshed muskmelon. Netted muskmelon ('Cruiser'), green-fleshed ('Honey Brew'), and orange-fleshed ('Orange Dew') muskmelons were harvested in Texas at the beginning (21 May) and at the end (11 June) of the production season in 2004. Fruit were analyzed immediately (day 0) or stored simulating retail conditions for 7 or 14 clays at 7°C and 95% ± 2% relative humidity plus 3 days at 21 °C. Both 'Orange Dew' and 'Honey Brew' non-netted cultivars evinced similar and less lipid peroxidation, and hence postharvest senescence, during the 17-day storage period than the netted muskmelon 'Cruiser'. In comparison with 'Cruiser', 'Orange Dew' generally exhibited higher concentrations of 13-carotene and phenolics and, with few exceptions, higher activities of the antioxidant enzymes ascorbate peroxidase (AsPX), monodehydroascorbate reductase (I\1l)HAR), dehydroascorbate reductase (DHAR), catalase (CAT), guaiacol peroxidase (PDX), and superoxide dismutase (SOD). Higher AsPX and SOD activities in both 'Orange Dew' and 'Honey Brew' appear to confer a greater resistance to lipid l)erOxidation in these muskmelon genotypes than to the netted 'Cruiser'. 'Orange Dew' also appears to be a healthier 100(1 choice not onl y due to its lack of a netted rind which could potentially harbour human illness-related pathogens, but also that it is superior to both 'Cruiser' and 'Honey Brew' in overall beta-carotene and phenolic levels. Among the different fruit and vegetables associated with food borne illness, netted muskmelon has consistently tested positive for Salmonella Lignieres (Castillo et al., 2004). Shigella Chatellani & Dawson (U.S. Food and Drug Administration. 2003), and Escherichia coli 0157:H7 (Del Rosario and Beuchat. 1995), and has been associated with large outbreaks of human illness (more than 25.000 individual cases since 1990) in the United States and Canada (Castillo et al.. 2004), and in 2001 and 2002 two cases of human death (Canadian Food Inspection Agency, 2003). Survival of human illness pathogens on netted muskmelon following surface pasteurization (Annous et al.. 2004; Ukuku, 2004) is attributed to inaccessible sites on the rind such as the netting (Beuchat and Scouten, 2004). Replacing netted, orange-fleshed muskmelon with a suitably phytonutrient Received for publication 16 May 2005. Accepted for publication 16 Aug. 2005. Atlantic Food and Horticulture Research Centre Contribution No. 2303. This research was funded in part by Agriculture and Agri-Food Canada (DM1-I.) and the USDA-ARS under CRIS No. 6204-43000-014-ODD to G.E.L. The valuable technical assistance of Robert D. Meyer (USDA-ARS. Weslaco. Texas). Vivian E. Willrncts and Michele L. Elliot (both Agriculture and Agri-Food Canada. Kentville. N.S.) is gratefully acknowledged. The authors wish to thank Drs. John DeLong and Charles Forney, Atlantic Food and Horticulture Research Centre. Agriculture and A gri-Food Canada. Kentville. N.S., for critical review of this manuscript. Use of company or product names by the USDA or Agriculture and Agri-Food Canada does not imply approval or recommendation of the product to the exclusion of others that may be suitable. 'To whom reprint requests should be addressed. Email: [email protected] dense non-netted, orange-flesh genotype such as orange-fleshed honey dew (C. inc/a L. Inodorus group) would greatly reduce the incidence of food borne illness associated with netted fruit. Non-netted, orange-fleshed honey dew fruit, such as 'Orange Dew', are commercially grown in limited quantities in the U.S. However, little is known regarding the human health-promoting phytochemicals or the antioxidant capacity of these genotypes compared to netted muskmelon cultivars. Moreover, enzymatic antioxidant capacity in orange-fleshed honeydew muskmelon associated with postharvest storage life has not been addressed. The current postharvest shelf-life of atypical green-fleshed honey dew muskmelon is 3-4 weeks (Edwards and Blennerhassett, 1990) and 10 to 14 d for netted muskmelon (Lester, 1988). It is unknown if non-netted orange-fleshed honey dew muskmelon cultivars have a postharvest storage life similar to those of green-fleshed honey dew or orange-fleshed netted muskmelons. Additionally. it is unknown what effect the environment plays on antioxidant capacities in melon genotype fruit harvested at the beginning vs. the end of the production season. Oxidative stress has been associated with postharvest quality losses of fresh fruit and vegetables (Hodges et al.. 2004). Active oxygen species (AOS) such as superoxide (0,) and the hydroxyl radical ('OH) play significant roles in lipid peroxidation, polysaccharide cleavage, and in both nucleic acid and protein degradation (Hodges, 2003). Past work has shown that capacities of certain antioxidants are important in maintaining postharvest quality 110 J. Aii:s. Soc. Hosr. Sci. 131(1 ):l 10-117. 2006. in spinach (Spinacia oleracea L.) (Hodges et al., 2001; Hodges and Forney, 2003), apples (Malus xdornestica Borkh.) (Barden and Bramlage, 1994), pears (Pvrus coinmunis L.) (Vanoli et al.,1995), pepper fruit (Capsicum annuum L.) (Jiménez et al., 2003), broccoli (Brassica oleracea L. Ital ica ( Troup) (Toivonen and Sweeney, 1998; Yamauchi and Kusabe. 2001), potato (Solanuin tuberoswn L.) seed tubers, (Zabrouskov et al., 2002), and nonnetted muskmelon (Lacan and Baccou, 1998). In addition, human epidemiological evidence has linked intake of fruit and vegetables with lower incidences of cancer, cardiovascular disease, immune system decline, and certain neurological disorders (Huang et al.. 2005), all of which have been associated with oxidative damage (Hodges and Kalt. 2003); it is the antioxidative capacity in fruit and vegetables that has been identified as one of the major mechanisms contributing to human health maintenance (Hodges and Kalt, 2003; Olsson et al., 2004; Vinson et al.. 2001). Enzy mc and nonenzyni ic antioxidants represent an important mechanism in AOS scavenging (for reviews, see Blokhina et al., 2003: Luric, 2003). Superoxide dismutase [SOD (EC 1.15.1.1)] catalyzes the dismutation of Oto I-LU2. Both catalase [CAT (EC 1.11.1.6)] and ascorbate peroxidase IAsPX (EC 1.11.1.11)1 eliminate H 2O, with ascorbate(vitamin C) being oxidized during the latter reaction catalyzed by AsPX. Ascorbate is re-reduced directly through activities of inonodehydroascorbate reductase MDHAR (EC 1.6.5.4)] and dehydroascorbate reductase I DHAR (EC 1.8.5.4)1 and indirectly through activity of glutathione reductase [GR (EC 1.6.4.2) 1Peroxidases [PDX (EC 1.11. 1.7)] also convert H,02 to H 20, though they can form 01 and H2 O, in a complex reaction involving NADH oxidation. Not only can ascorbate reduce H 2 0, in the AsPX-catalyzed reaction described above, but this compound can also react directly with the AOS 0,-, 'OH and singlet oxygen ( 1 0,) (Hodges and Forney 2003), and may play a role in the regeneration of a-tocopherol (Horemans et al., 2000). One of the primary antioxidant functions of 13-carotene, a lipid soluble antioxidant precursor to vitamin A, is to quench 1 0, (Knox and Dodge. 1985). The phenolic class of phytochemicals, which include flavonoids. tannins, hydroxycinnamate derivatives, and lignin, also possess antioxidative properties due to their high reactivity as hydrogen or electron donors, their ability to chelate proxidant transition metal ions such as Fe 1° and Cu 3 , and their free radical chainbreaking function (Blokhina et al.. 2003; Hodges and Kalt. 2003; Proestos et al., 2005). Studies have suggested that fruit extracts and/or isolated phenolics are associated with anti-cancer, antiinflammatory, card ioprotective, and neuroprotective properties (Hodges and Kalt, 2003). The purpose of this study was to compare 'Orange Dew', a representative orange-fleshed honey dew melon, with 'Cruiser,' a netted orange-fleshed muskmelon, for 1) enzymatic and nonenzymatic (human nutritional) antioxidant capacities. 2) stability of antioxidant capacity over the commercial harvesting period, and 3) postharvest senescence (following 17 d of simulated commercial/retail storage). For comparative purposes, fruit of the green-fleshed honey dew muskmelon 'Honey Brew' were also included in this study. Materials and Methods PLANT MATERIAL. Fully abscised fruit of orange-fleshed, netted muskmelon ('Cruiser'), orange-fleshed. non-netted honey dew ('Orange Dew'), and green-fleshed, non-netted honey dew ('Honey Brew') melons, free of defects, were hand-harvested from Starr Produce Co. (Rio Grande City, Texas). All fruit were collected from the held by 0900 ii R, at the beginning of the commercial harvest season (21 May) and 3 weeks later at the end of the 2004 season (11 June). Fruit were washed in 0.02% sodium hypochlorite for 30 s, rinsed in tap water, and then randomized into lots for storage treatment. Fruit were assessed the day of harvest or after storage for 7 or 14 d at 7 °C and 95% ± 2% RH plus 3 d at 21 °C to simulate commercial storage plus retail display temperatures. All fruit were chilled to 4 °C, washed with distilled water, the epidermis removed with a vegetable peeler, and the polar-ends (totaling two-thirds ofthc fruit) removed and discarded. Wedges of the remaining equatorial -region mesocarp tissue, devoid of seeds and integument tissue, were pureed in a food processor (Quick 'N Easy: Black and Decker. Towson, Md.) using 3to 5-s pulses. Tissue samples were assayed fresh for enzymes. frozen (liquid nitrogen, then stored at —80 °C) for compounds, and lyophilized (following freezing in liquid nitrogen) for total antioxidants. METABOLITE ASSSAYS. Ascorbic acid and dehydroascorbic acid were extracted from frozen tissue and determined according to the procedure of Hodges et al. (2001) and reported in this study as total ascorbate. 13-carotene was extracted under low light conditions from lyophilized tissue (0.020 g) using ice-cold heptane (1.0 mL) plusO.5 mLinternal standard [trans Apo-8'-carotene (40 tg' mL-'): Sigma Chemical Co., St. Louis] according to the modified procedure of Koch and Goldman (2004). The internal standard stock solution was made by dissolving Apo-8-carotene in 1.0 mL methanol then bringing to volume (250 mL) with heptane before storage in the dark at —20° C. This sample-internal standard mixture was vortexed for 2 mm. then centrifuged at 3000 g for It) miii at 0 °C. One milliliter of the supernatant was removed and 1.5 ml, fresh cold heptane was added to the pellet, vortexed for 2 mm then centrifuged as above. A second 1.0 mL of the supernatant was removed and 1.0 mL cold heptane was added to the pellet. vortexed for 2 min then centrifuged as above. A third 1.0 mL of the supernatant was removed, the supernatants were pooled and a I .0-mL aliquot was passed through a 0.2-mm nylon Millex-LCR 13 filter (Millipore Corp.. Bedford, Mass.) and stored in the dark at —20 °C until high-performance liquid chromatograph (HPLC) determination. Twenty microliters of each extract was injected into an HPLC system (Agilent Technologies. New Castle. Del.) equipped with a Discovery C18, 5-!tm column (150 x 4.6 min) and a C 18 guard column (20 x 4.0 mm; Supelco. Bellefonte, Pa.). 13-carotene was separated in a mobile phase of 100% methanol at a flow rate of 2.0 inL'rnin-' and detected at 454 urn. Malondialdehyde content was determined on 2.0 g of fresh tissue using the TBARS procedure of Hodges et al. (1999). Phenolic levels in 2.0 g of fresh tissue were assayed using the Folin-Ciocalteu method (Singleton and Rossi. 1965). Results are expressed as gallic acid equivalents (mg-g1 FW). Protein content was determined using the Bio-Rad assay (Bio-Rad Lab. Hercules, Calif.) based on the method of Bradford (1976). ENZYME ASSAYS. Activities of AsPX. CAT, DHAR. GR . MDHAR, PDX, and SOD were analyzed in 15 g of fresh tissue as described in Lester et al. (2004). TOTAL ANTIOXIDANT ASSAY. Both lipophilic and hydrophilic antioxidants were analyzed using randomly methylated l3-cyclodextrin (RMCD) as a solubility enhancer, 2.2 '-azobis (2-amidinopropane) dihydrochloride (AAPH) as a peroxyl generator and 6hydroxy-2.5,7.8 -tetra methylchroman-2-carboxylic acid (Trolox) as a standard according to Prior et al. (2003). J. AMER. Soc. HORT. Sci. 131(l):110-1 17. 2006. 111 STATISTICAL ANALYSES. Analysis of variance (ANOVA) using the general linear model procedures of SAS (SAS Institute. Cary. NC.) was conducted as a three-factor design with cultivar, harvest time, and storage duration as the main factors (Table I). Duncan's multiple range tests (P :5 0.05) were performed to evaluate the significance of differences between dependent variable means. Data are the average of live single-fruit replications per harvest. The enzyme and MDA means for cultivar, harvest time, and storage duration were extracted from the ANOVA and submitted to a principal component analysis (PCA) using GenStat (release 8.!: VSN International Ltd., Hemel Hempstead, England). The PCA loadings produced three major components accounting for 71% of the enzyme and MDA variation by treatments.