Effect of Natural and Synthetic Antioxidants on the Oxidative Stability of Cooked, Frozen Pork Patties
ABSTRACT: The effect of grape seed extract (GS; 0.02%), oleoresin rosemary (OR; 0.02%), water-soluble oregano extract (WS; 0.02%), propyl gallate (PG; 0.02% of fat), butylated hydroxyanisole (BHA; 0.02% of fat), and butylated hydroxytoluene (BHT; 0.02% of fat) on the oxidative and color stability of precooked pork patties stored at −18 °C for up to 6 mo were determined. Pork lean and trim were ground and mixed (30% fat). Antioxidants mixed with salt (2%) were added. Patties were formed, cooked to 71 °C, over wrapped in PVC, and stored at −18 °C. Lipid oxidation was determined using thiobarbituric acid-reactive substances (TBARS) and descriptive sensory evaluation. Color was determined instrumentally and visually. Samples were evaluated after 0, 1, 2, 3, 4, 5, and 6 mo of frozen storage. Based upon TBARS values, PG (0.21 mg MDA/kg) and GS extract (0.23) had more antioxidant activity over the storage period than did WS, OR, BHA, and BHT. GS had no effect on a* or b* values. Grape seed extract (0.02%) has the potential to inhibit oxidative rancidity as well as current synthetic antioxidants.
Consumers spend less time preparing foods at home and more money on convenience products such as precooked entrees than ever before (Hollingsworth 1994; Dumagan and Hackett 1995). Annual U.S. per capita consumption of pork is 51 pounds, with 62% of products sold in a processed form (Davis and Lin 2005). These meat products are ground, smoked, cured, or seasoned at the factory level, then distributed while frozen (Davis and Lin 2005).
However, precooked ground meat products are very susceptible to lipid oxidation, which can lead to undesirable flavor and odor changes during storage, because of their high fat content (20% to 30%) and the processes to which they are subjected. Lipid oxidation in meat products is often synonymous with warmed-over flavor (WOF), which can be described as the disappearance of fresh meat flavor and the development of stale, rancid flavors in precooked stored meats (Kanner 1994). In the initial stages of lipid oxidation, products have a cardboard flavor. As oxidation progresses, other flavors, such as painty, rancid, and oxidized, develop (St. Angelo and others 1990). Secondary compounds such as hexanal, pentanal, heptanal, and octanal produced from the oxidation of polyunsaturated fatty acids are responsible for the presence of warmed-over-flavors (stale, wet cardboard, painty, grassy, rancid) associated with cooked, stored pork (Campo and others 2006; Rojas and Brewer 2007).
Susceptibility of muscle tissue to lipid oxidation is related to the degree of lipid unsaturation, muscle, animal diet, additives such as salt, cooking method, manner of storage, and pH of the muscle (Kanner 1994; Rhee and Ziprin 2001).
As the demand for table-ready, reheatable entree items has increased, so have WOF problem-solving efforts. Lipid oxidation can be reduced by oxygen-restrictive packaging and various additives such as antioxidants currently used in the food industry. BHA, BHT, and propyl gallate are synthetic phenolic antioxidants that have been approved for use by the USDA since 1955 (Biswas and others 2004; USDA 2006). The use of synthetic antioxidants is less desirable due to current governmental recommendations and consumer preferences, so there is growing interest in finding alternative solutions.
Polyphenolic compound-containing plant materials such as rosemary, sage, α-tocopherol, pine bark extract, green tea, coffee, grape skin, aloe vera, fenugreek, ginseng, mustard, and soya protein can retard the development of WOF in meat products (Barbut and others 1985; St. Angelo and others 1990; Stoick and others 1991; Chen and Waimaleongora 1999; McCarthy and others 2001; Ahn and others 2002; Nissen and others 2004; Han and Rhee 2005). Many anthocyanidins/anthocyanins, flavonoids, and phenolic acids have antioxidative activities similar to BHA and BHT (Fukumoto and Mazza 2000). Some components found in rosemary oleoresin, such as rosmanol, rosmarinic acid, carnosic acid, and carnosol, may be up to 4 times as effective as BHA and equal to BHT (Houlihan and others 1984).
Grape seed extract contains a significant amount of antioxidative polyphenols as well (Ahn and others 2002; Mielnik and others 2006; Rojas and Brewer 2007). In order for natural antioxidants to be accepted by the food industry, they must work as well as synthetics currently in use. In addition, natural antioxidants must not adversely affect color, odor, or flavor of the meat product.
The objective of this study was to compare the effectiveness of several natural to synthetic antioxidants in the prevention of lipid oxidation in cooked, ground pork patties during frozen storage.
Materials and Methods
Based on information provided by the suppliers, grape seed extract (Gravinol Super™, Kikkoman, Tokyo, Japan) contained 98% total flavanols (89% proanthocyanidins), and water-soluble oregano extract (Origanox™ WS, RAD Natural Technologies Ltd., Barrington Chemical Corp., N.Y., U.S.A.) contained at least 22% total phenolics (7% rosmarinic acid and other hydroxycinamic compounds). Oleoresin rosemary was extracted in canola oil (Herbalox® Seasoning HT-25, Kalsec Inc., Kalamazoo, Mich., U.S.A.). Butylated hydroxytoluene (Sigma, Madrid, Spain), butylated hydroxyanisole (Sigma, St. Louis, Mo., U.S.A), propyl gallate (Sigma, Copenhagen, Denmark), and sodium chloride (NaCl; U.S. Salt, Watkins Glen, N.Y., U.S.A) were food grade.
Fresh pork lean (20 kg) and trim (10 kg) were obtained from the Univ. of Illinois Meat Science Laboratory. Lean and trim were weighed out to produce a target fat content of 30%, coarsely ground separately at 4 °C through a 1.3-cm plate (4152 Meat Grinder, Hobart Corp., Troy, Ohio, U.S.A.), then reground through a 0.32-cm plate (4346 Series Mixer/Grinder, Hobart Corp.). Ground meat mixture was weighed into 21 to 1.2 kg portions, placed into a Kitchen Aid mixer and homogenized for 2 min. Antioxidants (0.02% grape seed extract, water-soluble oregano extract, and oleoresin rosemary based on total weight of finished product and 0.01% BHA, BHT, and propyl gallate based upon total weight of trim in the finished product) were dispersed in salt (2% on finished product weight basis) as a carrier and added to meat mixture. Meat mixture was homogenized again for 1 min. Patties (7.5-cm dia, 0.8-cm thick) weighing approximately 50 g (raw) were cut then cooked on a flat griddle to an internal temperature of 71 °C (AMSA 1995). Internal temperature was monitored with a Digi-Sense® Scanning Thermometer (Model 920000-00, Cole Parmer Instrument Co., Barrington, Ill. U.S.A) using copper-Constantin thermocouple wires (Type T, Omega Engineering, Inc., Stamford, Conn., U.S.A.). Patties were cooled to 22 °C, packaged (6 patties/tray) on foam trays (15 × 20 cm), over wrapped with commercial polyvinyl chloride film (oxygen transmission rate = 880 cm3/m2/24 h), and held frozen (−20 °C) in the dark for up to 6 mo. Patties were allowed to thaw at 4 °C for 24 h prior to analyses.
Instrumental color and pH
Each cooked patty was divided horizontally into 2 portions and instrumental color was determined on the internal surface. The spectral curve was determined over the 400 to 700 nm range using a HunterLab (LabScan II, Hunter Associates Laboratory Inc., Reston, Va., U.S.A.), calibrated to white and black standards. CIE L*, a*, and b* values were calculated using Illuminant A, and the 2° standard observer (CIE 1978). Chroma ([a*2+b*2]1/2) and hue angle (arctan [b*/a*]) were also calculated. The pH was determined directly on patties using a standardized MPI pH meter (Meat Probes Inc., Topeka, Kans., U.S.A.).
An 8-member descriptive panel composed of Univ. of Illinois at Urbana, Champaign, undergraduate and graduate students and staff experienced in meat product evaluation was trained during six 1-h sessions to evaluate flavor, odor, and visual color of patties. During the training, a panel leader facilitated discussions of product characteristics. Using cooked patties, refrigerated and stored for 14 d, descriptive flavor and odor terms were selected by group consensus (Table 1). A 15-cm semi structured line scale (0 = none, 15 = extreme) was used to characterize the flavors and odors with an anchor value assigned for each respective standard. Visual green, tan, and gray colors were evaluated using a 15-cm semi structured line scale with standards developed using paint chips (Table 2).
Table 1—. Descriptive odor and flavor terms and standards for cooked, frozen pork patties.
| Fresh cooked pork||Cooked fresh pork loin (71 °C)+ 2% NaCl|| 7|
| Rancid||Linseed oil|| 7|
| Sweaty||Dry cat food dispersed in distilled H20|| 6|
| Grassy||Hexanalb (0.0005 ppm)|| 7|
| Herbal||Origanox™ WS (100ppm)|| 8|
| Sulphur||Hard boiled eggs|| 7|
| Fresh cooked pork||Cooked fresh pork loina+ 2% NaCl||10|
| Rancid||Cooked ground pork patty stored at 4°C for 21 d||15|
| Wet cardboard||Cardboard pieces wetted with distilled H20|| 4|
| Salty||Noniodized salt in distilled H20 (8000 ppm)|| 7|
Table 2—. Visual color standards for cooked, frozen pork patties.
| 6007-3B fennel splasha||1 ||86.14||−6.59||19.47|
| 96252 brass tackb||7.5||68.34||−3.38||27.38|
| 96253 cellini goldb||14 ||53.97||−1.27||31.82|
| 00YY61/121 coral clichéc||1 ||89.23||−2.90||13.21|
| 90YR48/183 soft wheatc||7.5||82.28||−1.05||16.78|
| 90YR38/239 rolling hillsc||14 ||76.00|| 2.86||22.41|
| 6005-1A asiagoa||1 ||72.23||−3.97||15.62|
| 6005-1B oatlands subtle taupea||7.5||64.42||−4.61||16.32|
| 6005-1C smoked oystera||14 ||53.12||−3.54||16.45|
All 7 samples labeled with 3-digit random numbers were evaluated during each session. Samples were prepared and presented to panelists as described by the American Meat Science Association (AMSA 1995). For odor and flavor evaluation, 3 patties were uniformly cut into 4 pie-shaped wedges then cut in half again. Two pieces were placed in each covered glass petri plate (60 × 15 mm; Fisher Scientific, Pittsburgh, Pa., U.S.A.), and reheated (Isotemp oven 516G; Fisher Scientific) at 60 °C for 30 min prior to presentation to the panel. References were prepared daily. Roasted coffee beans (odor), distilled water, Granny Smith apple slices, and unsalted crackers were presented for panelists to clear the palate between samples. Visual color evaluations were conducted after odor and flavor analyses under warm white fluorescent light (approximately 1200 lux) against the white background provided by the paper plates.
Thiobarbituric acid-reactive substances
Thiobarbituric acid-reactive substances (TBARS) were determined using the extraction method of Witte and others (1970) as modified by Miller (1998). Meat sample (5 g) was placed in a porcelain mortar (pour lip dia = 90 mm) with 1-mL BHT solution (0.2 mg/mL), and 45.5 mL of extraction solution containing 10% trichloroacetic acid in 0.02M phosphoric acid (TCA/H3PO4), brought to 50 mL. The mixture was ground with a pestle for 30 s then filtered through Whatman nr 1 filter paper. One additional sample was spiked with 12 mL of 10 μM 1,1,3,3,-tetraethoxypropane (TEP). Three 5-mL aliquots of filtrate from each sample were transferred into separate screw-cap test tubs (15 × 200 mm). To 1 aliquot, 5-mL deionized water was added (sample blank); to the 2nd and 3rd aliquots, 5-mL 0.02 M 2-thiobarbituric acid was added (test samples). Tubes were covered with parafilm, capped, inverted 3 times to mix, and held in the dark for 20 h at 22 °C. Absorbance was determined spectrophotometrically (Beckman Coulter DU® 640, Inc., Fullerton, Calif., U.S.A.) at 530 nm. Absorbance of sample blanks was subtracted from those of test samples. Concentration of malondialdehyde (MDA) was calculated from a standard curve using solutions of TEP (0 to 10 ng MDA/mL) based on the equation obtained from the curve (y= 0.0819x+ 0.0054; R2= 0.99). MDA recovery was computed using the TEP-spiked samples. TBARS were calculated, making the appropriate correction, and expressed as mg malondialdehyde/kg sample. BHT crystalline, tricholororacetic acid crystal reagent, and O-phosphoric acid were reagent grade (Fisher Scientific); TEP and 2-thiobarbituric acid were also reagent grade (Sigma-Aldrich).
Designed as a randomized complete block, this experiment included 7 antioxidant treatments (control, GS, OR, WS, PG, BHA, and BHT) and 7 storage periods (0, 1, 2, 3, 4, 5, and 6 mo). The entire experiment was duplicated (7 treatments by 7 storage times). Analyses, including TBARS, instrumental color, and pH, were conducted in duplicate on each of 2 replicate samples. Sensory evaluation was conducted in duplicate by 8 trained panelists. Data were analyzed treating antioxidant and storage period as independent variables using the PROC MIXED procedure (SAS 2002). TBARS, instrumental color, and pH main effects and interactions were treated as dependent variables and considered significant at P < 0.05. For sensory data, panelists were considered repeated measures. Least squares means for significant effects were separated using probability of difference, adjusted with the Bonferroni procedure for multiple comparisons. Pearson correlation coefficients were calculated between TBARS, off-odor, and off-flavor.
Results and Discussion
Effect of antioxidants on pH and color of cooked, frozen pork patties
No significant antioxidant by time interactions or antioxidant main effect occurred for pH. When data were pooled over antioxidant treatment, pH values for all treatments increased from 6.3 at 0 mo to 6.8 at 3 mo to 7.0 at 6 mo (data not shown).
No significant antioxidant by time interactions occurred for b* or L* values. When data were pooled over antioxidant treatment, storage time had significant but inconsistent effects on both b* and L* values; b* value ranged from 11 to 15 over the 6-mo storage period, and L* value ranged from 59.5 to 63.5 (data not shown). When pooled over storage time, no significant difference existed in b* value (range = 11.5 to 12.9; data not shown). When data were pooled over storage time, L* values (lightness) differed due to antioxidant treatment (Table 3). Samples containing GS were significantly darker and than the control; however, these differences were small (Table 3). This color difference could be due to the addition of GS extract. Rojas and Brewer (2007), however, found that GS (0.02%) had no effect on L* values of pork during refrigerated storage.
Table 3—. Effect of antioxidants on instrumental and visual color of cooked, frozen pork patties.
| L* value||61.61bc||59.95a||61.45bc||62.35bc||60.73ab||62.34c||62.11bc||0.34|
| a* value||2.94||3.03||2.96||3.03||2.69||2.89||2.61||0.10|
| b* value||12.37ab||11.50a||12.44ab||12.80b||11.96ab||12.83b||12.87b||0.24|
| Hue angle||76.05ab||74.94a||76.44ab||76.32ab||77.02b||76.97ab||77.99b||0.48|
| Saturation Index||21.36||18.71||20.03||20.22||19.12||20.20||20.33||0.71|
A significant antioxidant by storage time interaction existed for a* value (Table 4). Initially, red color (a* value) of samples did not differ due to antioxidants. Differences during the remaining 6 mo of storage were significant but followed no consistent trends and ranged from 1.8 to 4.0, less than a 3 unit difference. When averaged over storage time, antioxidant had no effect on a* value (range = 2.6 to 3.0; Table 3). Grape seed extract has a red coloration, which has the potential to increase redness of pork patties. Ahn and others (2007) reported that adding grape seed at 1.0% to beef patties, cooked and stored for 9 d resulted in redder patties than controls. In the present study, samples containing PG and BHT had the highest hue angles indicating they diverged the most from true red (Table 3). However, the range for hue angle (in degrees) was from 76.05 to 77.99, which may be of little significance with respect to visual appearance.
Table 4—. Effect of antioxidants and storage time on instrumental a* values of cooked, frozen pork patties.
No antioxidant by storage time interactions existed for visual color characteristics. Antioxidants had no effect on visual green or tan color (Table 3). GS-containing patties were darker gray than those OR; however, they did not differ from other treatments. Gray color of GS-containing samples was similar to the control; however, it was higher than OR- and BHT-containing samples by about 1.5 units on a 15-point line scale. Visual green, tan, and gray colors increased very slightly over time (data not shown). Visual gray color was significantly correlated with L* and a* values (correlation coefficients =−0.92 and −0.91, respectively; Table 5) indicating samples that were more visually gray (darker), had lower L* and a* values (samples were darker and less red).
Table 5—. Correlations coefficientsa between instrumental and visual color of cooked, frozen pork patties.
|a* value|| 0.37|| 0.28||−0.91b|
|b* value||−0.64||−0.51|| 0.38|
Effect of antioxidants on odor and flavor of cooked, frozen pork patties
Pork odor remained constant through the 4th month of storage becoming significantly lower after the 6th month (Table 6). Rancid odor began to increase between the 1st and 2nd months of storage, increased through the 5th month then decreased at the 6th month. Pork flavor was highest at 0 mo of storage; it was significantly lower after 6 mo of storage. Cardboard and rancid flavors were lowest at 0 mo of storage. Cardboard flavor was significantly higher after 4, 5, and 6 mo of storage. Rancid flavor was significantly higher after 3 mo, and remained higher throughout the storage period.
Table 6—. Effect of frozen storage time on odor and flavor of cooked pork patties.
GS-containing samples had lower rancid odor scores than controls and BHT-containing samples (Table 7). GS-containing samples had lower grassy odor scores than controls and did not differ from other antioxidant treatments. GS and PG-containing samples had lower cardboard and rancid flavor scores than controls and WS-containing samples. GS-containing samples also had lower rancid flavor scores than OR, BHA, and BHT-containing samples.
Table 7—. Effect of antioxidants on odor, flavor, and TBARS of cooked, frozen pork patties.
Effect of antioxidants on TBARS of cooked, frozen pork patties
All antioxidants reduced TBARS compared to the control. GS- and PG-containing samples had the lowest TBARS values. Other than control samples, WS- and BHT-containing samples had the highest TBARS. A significant antioxidant by storage time interaction existed for TBARS (P < 0.05; Table 8). TBARS values increased the most during the 1st month of storage for all antioxidant treatments. They continued to increase through the 3rd month of storage then tended to become stable or decrease slightly. Immediately after cooking but before storage (0 mo), control samples had TBARS > 0.4 mg/kg while GS-, PG-, and OR-containing samples had TBARS < 0.1 mg/kg indicating that the latter group of antioxidants protected samples from oxidation induced by cooking. Between months 0 and 1, TBARS of control samples nearly tripled. TBARS of OR-, WS-, and BHA-containing samples also increased dramatically. Samples containing PG, GS, and BHA had lower TBARS values (0.26, 0.31, 0.46, 0.63, respectively) than did controls, and samples containing WS and OR (0.92, 0.81, and 0.75, respectively).
Table 8—. Effect of natural and synthetic antioxidants and storage time on TBARS values (mg mda/kg sample) of cooked, frozen pork patties.
Grape seed extract controlled lipid oxidation in salted, cooked pork patties during the 6 mo frozen storage period. These results are similar to those of Rojas and Brewer (2007), who reported that grape seed extract added prior to cooking was an effective antioxidant since it prevented TBARS formation, and alleviated the prooxidant effects of NaCl, while having no effect on the moisture or pH of the product.
Ahn and others (2002) reported adding grape seed extract to beef patties resulted in 57% lower TBARS values when compared to controls after 3 d of refrigerated storage. The grape seed extract used in this study contained 98% flavanols, which was higher than that (75% polyphenols) in the extract used by Ahn and others (2002). Brannan and Mah (2007) found the addition of 0.1% grape seed extract to ground, salted chicken thigh meat stored for 9 d at 4 °C was an effective antioxidant, preventing TBARS formation, alleviating the prooxidant effects of NaCl, while having no effect on the moisture content or pH of the product.
Relationship between TBARS, odor, and flavor
TBARS were significantly correlated (P < 0.05) with grassy and rancid odors (correlation coefficients = 0.92 and 0.97, respectively; Table 9). Grassy odor is a reliable indicator of WOF in meat products (Nissen and others 2004; Nam and others 2007; Rojas and Brewer 2007). Rancid odor has also been correlated with the development of WOF in meat products (Campo and others 2006). These correlations indicate that TBARS can be used as an indicator of the off-odors commonly associated with WOF in cooked, frozen pork patties.
Table 9—. Pearson correlation coefficients between odor, flavor, and TBARS of cooked, frozen pork patties.
TBARS were significantly correlated with cardboard and rancid flavors (correlation coefficients = 0.91 and 0.91, respectively; Table 9). These results are similar to those of Rhee and others (2005). Cardboard flavor was correlated with grassy and rancid odors (correlation coefficients = 0.86 and 0.90, respectively). In addition, rancid flavor was correlated with grassy and rancid odors (correlation coefficients = 0.93 and 0.89, respectively).
Grape seed extract is a by-product of grape juice and wine processing which contains high concentrations of proanthocyanidins. These proanthocyanidins have powerful antioxidative activity due to their ability to scavenge free radicals, chelate metals, and perform synergistically with other antioxidants (Lu and Foo 1999). Other natural antioxidants used in this study have been evaluated at various concentrations when added to meat products. Oregano extract has been shown to be less effective at 0.02% than at 0.1% (Sánchez-Escalante and others 2003). Rosemary extracts in meat products are effective between 0.02% and 1% (St. Angelo and others 1990; McCarthy and others 2001; Ahn and others 2002; Nissen and others 2004). The rosemary extract and oregano extract added in this study were effective at preventing lipid oxidation, but not as effective as grape seed extract and propyl gallate.
With respect to synthetic antioxidants, McCarthy and others (2001) demonstrated that adding 0.01% of a BHA/BHT combination prevented lipid oxidation in cooked pork patties over a 9-d storage period. However, Jayasingh and Cornforth (2003) found BHT (0.01% of fat) was ineffective in cooked pork patties stored at −20 °C for 6 mo. Wang and others (1999) reported that propyl gallate prevented lipid oxidation in refrigerated beef patties.
Based on TBARS values, rancid and grassy odors, and cardboard flavors, grape seed extract, a natural antioxidant, was as effective as propyl gallate at preventing oxidation in cooked pork patties. Grape seed extract did not affect instrumentally measured red color (a* value) or yellow color (b* value). The L* value of samples containing GS was lower than those of controls indicating that grape seed extract may have caused some darkening in pork patties. However, sensory panelists were unable to perceive this change visually. Grape seed extract has the potential to be as powerful an antioxidant as synthetic antioxidants that are currently used in cooked, frozen meat products without adversely affecting meat color.