Research and Reviews: Meat

Special Circular 172-99

Effects of Anka Rice, Nitrite, and Phosphate on Warmed-Over Flavor in Roast Beef

H. W. Ockerman¹ and J.H. Cheng
The Ohio State University Department of Animal Sciences

Abstract

Warmed over flavor in cooked beef is primarily due to the oxidation of lipids, particularly phospholipids located in the lean tissue membranes. Anka rice did not inhibit lipid oxidation (thiobarbituric acid values) during surface curing but may have masked rancidity in cooked and cooled beef. But anka rice plus nitrite and phosphate (A+N+P) improved sensory evaluation and seemed to have a synergistic effect on flavor during refrigerated storage and proved to be the most desirable treatment for re-cooked beef.

Introduction

Warmed-over flavor (WOF) of cooked beef is primarily due to oxidation of lipids, particularly phospholipids located in lean tissue membranes. Many researchers have indicated that chelators help inhibit WOF in cooked beef. In the Orient, Chinese roasted pork is a very popular warmed- over food that uses deep dark-red anka rice as a coloring agent and to increase flavor.

The purposes of this study were to compare the effects of anka rice (A), phosphate (P), and nitrite (N) alone and in combination on WOF in roast beef; to compare the effects of these ingredients added in a curing solution only applied to the surface; and the effects on specific characteristics in roast beef during refrigerated storage.

Materials and Methods

Experimental Design

The experimental design is shown in Figure 1. All treatment samples after cooking were stored at 4°C until measured for chemical analysis, oxidation, and physical characteristics at 0, 2, 4, 7, and 10 days of refrigerated storage after reheating, and microbial analysis was conducted on the same days without reheating. Sensory evaluation was tested at 0, 2, 4, and 7 days of refrigerated storage after reheating.

Figure 1. Experimental Design

Chemical Analysis

Moisture analysis was performed by the oven drying method (Ockerman, 1985). Fat analysis was determined by the AOAC method of ether extraction (Ockerman, 1985). The pH values were measured on 1 to 10 diluted samples using a Corning pH meter (Ockerman, 1985).

Thiobarbituric Acid (TBA) Value

The samples of all treatments were separated into nonpeeled (whole tissue) and peeled (center tissue only) samples before the modified extraction/filtration TBA method (Pensel, 1990) was conducted.

Physical Measurements

Shear value was measured by the Warner-Bratzler shear instrument (Ockerman, 1985). The modified centrifuge technique (Tsai and Ockerman, 1981) was used to evaluate water-holding capacity (WHC).

Sensory Analysis

The sensory panelists were asked to score warmed-over flavor, warmed-over aroma, roast beef flavor, juiciness, and tenderness, utilizing a 9-point scale.

Total Plate Count (TPC) Method

TPC procedure (Speck, 1984) was used to measure microbial growth of samples, which were not reheated after refrigerated storage on the day of test.

Statistical Analysis

Data was analyzed by the Statistical Analysis System (SAS) to compute analysis of variance (ANOVA) and correlation coefficients (SAS, 1994). Duncan’s multiple range test from SAS was used for comparison of treatment means and across times (SAS, 1994).

Results and Discussion

The Peeled and Nonpeeled TBA Values

The major tendency for peeled TBA values was a steady significant increase over time for all treatments (Table 1). The peeled TBA values of the fresh treatment were significantly higher than all others at four and seven days. At 10 days, the fresh and the control treatment had significantly higher TBA values than all other treatments. Of the other treatments, N and A+N treatments had the lowest peeled TBA values at 10 days. From two to 10 days, treatments that contained either P or N had significantly lower peeled TBA values when compared to the control treatment. When samples containing these two additives were compared at the levels used, it would also suggest that N added to treatments gives more antioxidative protection than P. However, the antioxidant activity of A treatment as measured by TBA values resulted in significantly higher TBA values when compared to both P and N treatments over these times (seven to 10 days) suggesting only minimal antioxidant activity. N and A+N treatments maintained significantly lower peeled TBA values than other treatments – especially at seven and 10 days of storage.

For nonpeeled TBA values A+P and A+N+P treatments did not increase significantly during this time frame (two to 10 days). At zero day, the nonpeeled TBA value of the fresh treatment was lower than the control treatment, and the reason is probably due to the catalytic activity of some ingredients (e.g., salt) in the surface curing which increased lipid oxidation. In general, all treatments increased in TBA over storage times as would be expected. Again, fresh treatment received the highest value from two through 10 days of storage and this difference was significant from all treatments except the control treatment at two days. The next highest TBA values over time were for control and A treatments, and these were significantly higher than other treatments. However, at 10 days, the control treatment was even higher than the A treatment. Treatments with antioxidants (N and P but not A) had lower nonpeeled TBA values than the fresh and the control treatments from day two to 10. When N was added to treatments instead of P, the result would suggest that N had more antioxidative properties to decrease nonpeeled TBA values. From days seven and 10, N and A+N treatments received the lowest TBA values. At 10 days of refrigerated storage, the nonpeeled TBA value of anka rice treatment was closed to 1 mg/kg, and fresh and control treatments were higher than 1.0.

The correlation of peeled and nonpeeled TBA value was r = 0.97. The external surface has more extensive exposure to oxygen but also has more contact with additives, some of which should have antioxidative properties. In general, nonpeeled samples had higher TBA values, particularly late in storage, when compared to peeled samples, suggesting that oxygen content had a stronger influence on oxidation than antioxidative additives.

Treatments with N contained in surface curing had lower peeled and nonpeeled TBA values than those with P, and N treatment surprisingly was even lower in TBA values than the combination of these two ingredients. Antioxidants did not stop the increase of peeled and nonpeeled TBA values completely, but slowed down lipid oxidation. Overall, there was no synergistic effect of A, N, and P on TBA values in roast beef. In both peeled and nonpeeled TBA values, N and A were proven to be the most significant desirable treatments from a TBA standpoint.

Warmed-Over Aroma (WOA)

Lower WOA scores indicated more intense warmed-over aroma. At two days, the fresh, control, and A treatments had lower WOA scores compared to other treatments. The results suggested that A used alone did not inhibit WOA at two days, but N and P did retard the development of WOA. This agrees with the A treatment in nonpeeled samples, which also had similar TBA values as the control treatment, and other antioxidant treatments had lower TBA values compared to control treatment. From two to seven days, A used alone did not inhibit the development of WOA. This result also agreed with measurement of nonpeeled TBA values. At four days, WOA score of fresh treatment was lower than that of control treatment. The control treatment had the same score as other treatments except for N and A+N+P treatments. Treatments with antioxidants, except for the A treatment, had higher WOA scores than control treatment at seven days. Both P and N treatments decreased WOA at seven days. These results suggested that P and N inhibit the development of WOA in some cases, but that A did not.

WOA scores of all treatments decreased over refrigerated time; therefore, all antioxidative ingredients used in surface curing did not stop lipid oxidation but retarded its development. P, N, N+P, A+P, A+N, and A+N+P had more desirable WOA scores at seven days than did fresh, control, and A treatments. Only N and A+N+P treatments maintained higher WOA scores compared to control treatment from two to seven days. It would appear from both TBA values and warmed-over aroma results that N and P are retarding WOA development, and A may slightly mask it in some treatments.

Warmed-Over Flavor (WOF)

Lower WOF scores indicated more intense warmed-over flavor. Treatments with antioxidants (not including A treatment) had significantly more desirable WOF scores than the control treatment, which had values similar to fresh and A treatments at two days. A alone added in surface curing did not inhibit the development of WOF during short storage (two days) when compared to the control. At the fourth day, only N and A+N+P treatments had significantly more desirable WOF scores when compared to the control, which had a higher score than the fresh treatment. Some ingredients used in the control treatment retarded or masked (flavoring ingredients) the development of WOF compared to fresh treatment at four days of refrigerated storage. At seven days of storage, all treatments containing nitrite and tripolyphosphate inhibited WOF development, but A did not. These results agreed with nonpeeled TBA values. All WOF scores decreased among treatments over time. Although N and P added in surface curing had more desirable WOF scores most days, they only retarded the development of WOF.

Other Characteristics

For Warner-Bratzler shear value, the fresh treatment had the significantly highest shear value over time. Some ingredients in the control treatment decreased shear value more than others. P at best had only a moderate effect on shear value due to surface curing. For water-holding capacity, A and N, in general, did not increase water-holding capacity in roast beef over refrigerated storage time. However, all the treatments with A, N, and P maintained one of the highest WHC values over time. There was a nonsignificant treatment-time interaction for roast-beef flavor, juiciness, and tenderness scores by sensory evaluation. All sensory scores of these three attributes significantly decreased during refrigerated storage of roast beef.

Microbial Growth (TPC)

N and A added in surface curing retarded microbial growth, but phosphate alone did not prevent it; however, it still inhibited microbes compared to the control treatment.

Conclusions

A did not inhibit lipid oxidation of roast beef as measured by TBA values but A+N+P often resulted in improved sensory evaluation (WOA and WOF) during refrigerated storage. This would suggest that under the conditions of this research A may have a flavor-masking effect on oxidation. Some characteristics of roast beef (WHC, shear value, juiciness, tenderness) also were not improved by A used in surface curing over time.

Sodium tripolyphosphate added in surface curing retarded lipid oxidation as compared to the control treatment. Due to surface-curing samples with added sodium tripolyphosphate, the pH, moisture, and yield were not significantly affected. However, objective parameters such as WHC and shear value or sensory evaluation such as juiciness and tenderness scores, were slightly improved by this alkaline phosphate in roast beef during refrigerated storage.

In several important characteristics, there seemed to be a synergistic effect among A, N, and P when used in surface curing and refrigerated storage. When comparing all evaluations, it currently would seem that the A+N+P is the most desirable treatment in retarding the development of warmed-over flavor and in retarding overall palatability of roast beef during refrigerated storage.

References

Ockerman, H. W. 1985. Quality Control of Post-Mortem Muscle Tissue. Department of Animal Sciences, The Ohio State University, Columbus, Ohio.

Pensel, N. A. 1990. Influence of experimental conditions on porcine muscle and its effect on oxidation. Thesis. The Ohio State University, Columbus, Ohio.

SAS. 1994. Statistical Analysis System. SAS Institute, Inc. Cary, N.C.

Speck, M. L. 1984. Compendium of Methods for the Microbiological Examination of Foods. 2nd Ed. American Health Association, Inc., Washington, D.C.

Tsai, T. C. and Ockerman, H. W. 1981. J. Food. Sci. 46:697.

¹ For more information, contact at: The Ohio State University 15 Animal Science, 2029 Fyffe Road, Columbus, OH 43210, (614) 292-4317, Fax (614) 292-2929; email:ockerman.2@osu.edu