2 Materials and methods
2.1 Fish Oil and Flavors
Fish oil was manufactured by DSM Nutritional Products (Dartmouth, NS, Canada) and donated by Nature’s Way of Canada Ltd. The fresh base oil contained approximately 18% EPA and 12% DHA and was stabilized with 3 mg g-1 natural mixed tocopherols. Flavoring agents were sourced from FONA International (Geneva, IL, USA). The flavors consisted of medium-chain triacylglycerols and natural flavor compounds. They were not stabilized with any additional antioxidants.p -Anisidine crystals and solvents used in this study were purchased from Sigma Aldrich (Oakville, ON, Canada)
2.2 Screening Flavored Fish Oils
The fourteen flavors screened were apple, blackcurrant, bubblegum, chocolate-vanilla, citrus punch, coconut, cranberry, grapefruit-tangerine, lemon, mandarin, mango, meat, orange, and peanut butter. Fourteen 4 mL screw top vials were loaded with accurately weighed fish oil (about 2.5 g per vial). As a common industrial practice, 2% (w/w) flavor was added to each vial. All vials were purged with nitrogen, capped, and vortexed for 30 seconds. pAV of these flavored oils and the unflavored base oil were measured in duplicate, following AOCS Official Method Cd 18-90 (AOCS, 2011).
2.3 Identifying Major Flavor Compounds Using GC-MS
After determining that chocolate-vanilla and lemon flavors had the largest influence on the measured pAV, GC-MS was used to identify major flavor components in these flavors. A Thermo Trace 1310 GC was coupled with a Thermo ISQ 7000 single quadrupole mass spectrometer equipped with an autosampler. Flavors were diluted 100 times in hexane with 50 ppm internal standard (tribenzylamine). Half a microliter of each diluted flavor was injected in split mode with a split ratio of 150:1 at 250 °C. Separation of compounds was achieved on a Zebron ZB-5 capillary column (5% phenyl, 95% dimethylpolysiloxane; 5 m GUARDIAN; 30 m length, 0.25 mm I.D., 0.25 µm film thickness). The oven temperature was: 1) held at 60 °C for 2 min, 2) increased to 80 °C at a rate of 7 °C min-1, 3) increased again to 150 °C at a rate of 3 °C min-1, 4) further increased to 260 °C at 7 °C min-1, 5) increased to 300 °C at 3 °C min-1, and 6) held for 20 min. Total run time was 78 min. This long temperature program ensured large molecules, including the triacylglycerol carrier oil, had eluted from the column at the end of each run. The MS was operated in TIC mode, with a scanning range from 60 to 600 amu and dwell time of 0.2 sec. Transfer line and ion source were both set at 300 °C. MS data were processed by Chromeleon Chromatography Studio. Peaks with area larger than 1.0 % of internal standard were tentatively identified through the NIST library search and confirmed using an MS processing method with three confirmation ions that had been developed and employed for routine in-house analysis of flavors. The retention indices of all the identified compounds were compared to further confirm the peaks’ identities.
2.4 Oxidation of Chocolate-Vanilla and Lemon Flavored Fish Oils
The following experiment and subsequent analyses were performed twice, once with chocolate-vanilla flavor and once with lemon flavor. Accurately weighed aliquots of fish oil (~3.5 g) were added to fifty-one 4 mL screw top vials. To eighteen of these vials, 2% (w/w) flavor was added. All vials were then purged with nitrogen, capped, and vortexed for 30 seconds. Following vortexing, three vials without flavor and three with flavor added were set aside to serve as unoxidized baseline samples. The remainder of the samples were placed uncapped in a 40 ± 2°C oven for oxidation. Every four days for a period of 20 days, nine oil samples (two triplicate sets of unflavored oils, and one triplicate set of oil to which flavor had been added prior to incubation) were taken out of the oven and cooled to room temperature. To one set of triplicate unflavored oil samples, 2% (w/w) flavor was added, thus creating three streams of oxidized oil: unflavored oil (UFO), oil to which flavor had been added before oxidation (FBO), and oil to which flavor was added after oxidation (FAO). All three triplicate sets of oil samples were then purged with nitrogen, capped, and vortexed for 30 seconds. pAV of oils was measured following the protocol described above.
2.5 Statistical Analyses
The following equation was used to model (SPSS, IBM, New York) the pAV growth data for each treatment: