Background

Although evaluating the internal color of a ground beef patty is not a reliable method to determine if a patty is properly cooked to assure that a safe product is consumed, consumers are still influenced by the internal color of the cooked patty (van Laack et al., 1996). Ground beef patties that are cut soon after being removed from an oven or grill may appear to be fully cooked (brown-tan) internally although internal temperatures are far below that which ensures a microbiologically safe product (Hague et al., 1994). This phenomenon is well documented and is commonly identified as premature browning (Hague et al, 1994). On the other hand, information related to color reversion is very limited. Color reversion can be described as the reappearance of a red color some time after cutting open a cooked ground beef patty. A more thorough understanding of color reversion may also provide new insight into mechanisms associated with premature browning. Fresh (uncooked) meat color is largely affected by the amount and chemical state of the meat pigments (myoglobin, hemoglobin, cytochrome c). A considerable amount of information is known about the factors that affect the chemical state of myoglobin. In a well-bled animal, myogobin would be the primary pigment of interest. There are three chemical states of myoglobin. Deoxymyoglobin (DMb) is the natural color found in the interior of unexposed intact beef often described as dark purplish-red in color. Oxymyoglobin (OMb) is the bright cherry red color. Metmyoglobin (MMb) is an undesirable brown, oxidized color. Myoglobin is composed of a protein portion (globin) and a non-protein portion (heme ring). The oxidative state of the iron in the heme ring dictates what if any ligands (atoms, chemicals) can bind and the color of the pigment. MMb is the only chemical state of myoglobin that the heme-iron is oxidized (ferric iron, Fe3+). MMb cannot bind with oxygen. The heme iron must be in the reduced state (ferrous, Fe2+) in order for oxygen to bind. Almost immediately after exposing the internal surface of beef to air, the meat will have all three of the chemical states of myoglobin present. DMb (dark purplish-red) will be the most visible initially. Upon exposure to air, low levels of tissue oxygen promote the formation of MMb (brown). In fresh beef, the formation of the brown MMb pigment may not be visible because fresh beef has very active enzymatic reducing systems that limit the heme-iron from becoming oxidized. With time oxygen penetration becomes significant enough such that OMb represents the predominant visible pigment. In order for this to happen, OMb that forms on the outer layer, must yield a sufficiently thick layer such that the next inner layer (MMb) and innermost layer (DMb) are not visible. This is one of the key factors to producing beef that has a desirable bloom. At low levels of oxygen, MMb formation is promoted.  

For color reversion (red to reappear) to occur in a cooked patty, the protein portion of the pigment must undergo limited heat denaturation. Hunt et al. (1999) and Warren et al (1996) found that the chemical state of myoglobin in the uncooked ground beef patty influences the percent denatured myoglobin in the same patty post cooking. When deoxymyoglobin was the predominant myoglobin form in the patty less denaturation occurred than when oxymyoglobin and metmyoblobin (MMb) were present. Upon cooling the chemical structure of the myoglobin must be able to sufficiently return to its native, functional state and the heme iron needs to be in the reduced state so that either OMb (bright red) or DMb (red) can form.  

Other factors that can affect oxygenation of fresh beef include: pH, temperature, time, meat age, bacteria, MMb-reductase, and meat respiration rate (Madhavi and Carpenter, 1993; Moiseev and Cornforth, 1999). The pH of the meat is a critical factor because it influences the activity of reducing enzymes and oxygen consumption rate.  

Madhavi and Carpenter (1993) demonstrated that oxygen consumption was higher in ground beef than intact muscles due to greater tissue disruption. Increased oxygen consumption was also associated with decreased color stability in uncooked product. Ryan et al. (2006) found that myoglobin denaturation increased with increasing post-cook holding time, indicating the importance of time and temperature in the denaturation of myoglobin. They also found that 63.9% of myoglobin must be denatured to produce a medium doneness (a*=15.5). To achieve a well-done status (a*=12.4), 80.3% of the myoglobin must be denatured in cooked ground beef patties. Color reversion has been documented in pork and poultry products. Ghorpade and Cornforth (1993) found color reversion occurred in pork roasts cooked to 65° C after holding for 15 minutes. Trout (1989) found that higher pH protects myoglobin from denaturation. Osborn et al. (2003) found that metmyoglobin can be induced in heated beef muscle extracts and reduced to form OMb through an NADH-metmyoglobin reducing system.  

Since color reversion has been observed in pork and poultry products, we hypothesize that cooked ground beef patties will display a return of redness following cooking. In addition, we hypothesize that fresh meat holding time and freezing will affect color reversion in cooked ground beef patties. 

Methodology 

Three samples (11 kg each) of cow beef trimmings (7-10 d postmortem) were obtained from three separate production lots at a commercial processing facility. Two samples (4.5 kg, 7-10 d postmortem) from two different lots of coarse ground 90/10 beef trimmings were obtained from an additional commercial meat processor. Each sample served as meat for an individual replication. The pH was measured on each replication quantity using a 10 g sample homogenized in 90 ml of distilled, deionized water. The average pH was 5.95 with a range of 5.66 to 6.14.  

The cow beef trimmings were coarse ground through a 1.9-cm (¾ in.) plate upon arrival at the University of Wisconsin Meat Science Laboratory. Each 11 kg batch of coarse ground cow beef trimmings was blended and 4.5 kg were retained for use in this project. The retained 4.5 kg were vacuum packaged and stored in an ice slurry at 0 C for three days to allow the myoglobin to reach its native state. Following the three-day hold period, 1.3 kg of the coarse ground product were vacuum packaged and frozen in a –25 C freezer 60 days. The remaining 3.2 kg of each sample was ground once through a 0.32-cm (1/8 in.) plate. A 1.3-kg sample of this product was placed on a Styrofoam tray, wrapped in oxygen-permeable PVC film and refrigerated at 0 to 2 C in the dark for seven days. All unfrozen product (0d and 7d storage) was reground once through a 0.32-cm plate prior to patty formation to promote equal dispersion of metmyoglobin, oxymyoglobin, and deoxymyoglobin throughout each sample. Each coarse ground frozen treatment was thawed for 5 days in a 0 to 2 C dark cooler before regrinding. The thawed samples were then ground through a 0.32-cm plate twice prior to patty formation to ensure equal dispersion of temperature as well as myoglobin chemical states. A manual patty press was used to form three 113.5 g patties that measured 11.5 cm x 12.5 cm x 0.8 cm for each treatment (0d storage, 7d unfrozen storage, and 60d frozen storage. The patties were cooked at an average rate of 6.18 C per minute (2.93 standard deviation) on a liquefied propane gas grill (Char Broil Model 463350905, Char Broil Co., Columbus, GA 31904). The patties were turned once at an internal temperature of 43.3 C (110 F) and removed from the grill at an internal temperature of 62.8 C (145 F). The temperature rise was monitored using a DigiSense Digital Thermometer. Once the endpoint temperature (69.4 C; 157 F) was achieved, each patty was wrapped in aluminum foil and allowed to cool at room temperature to an internal temperature of 54.4 C (130 F). The patties were then cut longitudinally with an electric fillet knife. Initial cooked color measurements were taken on the freshly cut surface of one half of each patty. The halves used for the initial color measurements were then held at room temperature for 10 minutes. Color measurement was repeated following a 10-minute hold time at room temperature.  

Instrumental Color Measurement  

Color was determined using a chroma meter (Model CR200, 8 mm aperture, Minolta Camera Co., Ltd., Osaka, Japan) and a scanning reflectance spectrophotometer (Model UV-2401 PC, Shimadzu Corp., Kyoto, Japan). Color instruments were standardized against a white (No. 20933026 CIE L* 97.91, a* -0.70, b* +2.44) calibration plate. The chroma meter was used to determine CIE L* a* b* values on the cut surface of the ground beef patties. Each chroma meter measurement was repeated three times per patty at 0 and 10 minutes post cooking (time effect, an independent variable). CIE L* estimates lightness, a* redness, and b* yellowness. Using the spectrophotometer, the chemical states of myoglobin were estimated by the following reflectance wavelength combinations: deoxymyoglobin (DMb, %R474 nm / %R525 nm), metmyoglobin (MMb, %R572 nm / %R525 nm), and oxymyoglobin (OMb, %R610 nm / %R 525 nm) as recommended by AMSA (1991). Each spectrophotometer measurement was performed once at 0 and 10 minutes post cutting per patty.  

Statistical analysis  

All experiments were replicated five times. Data was statistically analyzed using PROC MIXED Model of SAS (2002). The experiments involving unfrozen meat (stored 7 d, meat age experiment) and frozen meat were independently analyzed against unfrozen (0 d stored) ground beef. The unfrozen patty data were analyzed as a split-plot design with day (0 vs 7) as the main effect and time of exposure to air after cutting (time, 0 vs 10 min.) as the split plot effect. The frozen patty data were compared to the unfrozen patty (0 d) results also in a split-plot design. The main effect was treatment (unfrozen-no storage vs frozen and stored, 60 d). Time of exposure to air (0 vs 10 min) represented the split plot effect. If significance was determined (P < 0.05) in the model, dependent variable means were separated using Least Square Means of SAS.  All experiments were replicated five times. Data was statistically analyzed using PROC MIXED Model of SAS (2002). The experiments involving unfrozen meat (stored 7 d, meat age experiment) and frozen meat were independently analyzed against unfrozen (0 d stored) ground beef. The unfrozen patty data were analyzed as a split-plot design with day (0 vs 7) as the main effect and time of exposure to air after cutting (time, 0 vs 10 min.) as the split plot effect. The frozen patty data were compared to the unfrozen patty (0 d) results also in a split-plot design. The main effect was treatment (unfrozen-no storage vs frozen and stored, 60 d). Time of exposure to air (0 vs 10 min) represented the split plot effect. If significance was determined (P < 0.05) in the model, dependent variable means were separated using Least Square Means of SAS. 

Findings 

Chromameter measurements   

In unfrozen cooked patties (meat age effects), day had an effect on a*, b*, and chroma (Table 1). The results indicated that as meat age increased, a*, b*, and chroma values decreased. Day had no effect (P > 0.05) on L* or hue angle. Time had an effect on all dependent variables recorded. L* and hue angle values decreased between 0 and 10 minutes post cutting. The dependent variables a*, b*, and chroma increased between 0 and 10 minutes post cutting.  

The results indicated that patties produced from frozen meat decreased a* and chroma values while hue angle increased compared to patties produced from beef that had not been frozen (Table 2). Storage condition had no effect (P > 0.05) on L* and b* values in the cooked patties. Time had an effect (P < 0.05) on all dependent variables except hue angle. L* values decreased with time following cutting while a* , b*, and chroma increased with time in patties produced from unfrozen and frozen beef.  

Reflectance measurements  

In the cooked patties made from beef stored unfrozen for 0 and 7 days, the only dependent variable affected (P < 0.05) by the two independent variables (time and day) was the time effect on the estimation of oxymyoglobin (Table 3). The results indicate a reduction in the estimate of oxymyoglobin following patty cutting. On the contrary, in the unfrozen and frozen storage comparison, the only dependent variable affected (P < 0.05) by the independent variables was the storage treatment effect on the estimate of metmyoglobin (Table 4). The results indicate an increase in metmyoglobin as are result of frozen storage. 

Discussion 

Establishing a specific endpoint temperature that produces patties capable of color reversion is contingent on a number of factors. Many of the factors known to affect premature browning would be expected to also pay a role in color reversion. Cooking conditions along with specifics on the beef were established to create patties that demonstrated color reversion. An end-point temperature was established for producing patties capable of undergoing color reversion. Relative to the starting patty cooked color, patties produced from unfrozen beef had mean a* values that were similar to or lower than that which Ryan et al. (2006) reported as well-done in appearance (a*=12.4). Patties produced from thawed frozen beef had even lower initial a* values than the fresh beef.  

Although an endpoint temperature was found that achieved color reversion in our study, this endpoint temperature is potentially contingent upon numerous other factors including the meat source, meat age, pH, chemical state of myoglobin, reducing ability of the muscle, heating rate, and post-cook handling conditions. Increased meat age and frozen storage of beef would tend to decrease the meat’s reducing ability and therefore lower the expected internal temperature and total thermal input necessary to partially denature myoglobin to a level that is still capable of color reversion upon exposure to air. The cow meat used in this experiment was higher in pH than that associated with young fed beef. The higher the pH would tend to raise the thermal input necessary to denature myoglobin to the level necessary to achieve a thoroughly cooked appearance upon initial cutting of the patty while maintaining enough integrity to undergo color reversion. Cooking produces hemichromes which are tan to brown in color. To produce a hemichrome, the globin (protein) portion loses its native functional state through heat denaturation and the heme-iron is oxidized. With sufficient oxidizing conditions, myoglobin can also form metmyoglobin which is also brown. It is unclear as to whether a hemichrome is capable of color reversion.  

Heat would be expected to cause oxymyoglobin to release its oxygen thereby losing the bright cherry red color of fresh beef. Background deoxymyoglobin (dark red) would not be expected to be as visible in a cooked patty as oxymyoglobin. Heating may release the oxygen from myoglobin that has a partially altered native state of the globin such that even deoxymyoglobin does reflect the typical red wavelengths of light until the patty has sufficiently cooled and the globin has regained its native configuration. With insufficient cooking, metmyglobin could be formed. Upon cutting the patty and exposure to air, in the presence of active meat myoglobin reducing enzymes, this pigment could lead to deoxymyoglobin or oxymyoglobin. This line of reasoning agrees with visual observations and the chroma measurements in that the brown-tan color gradually reverted to red.  

Our results agree with Ghorpade and Cornforth (1993). They observed color reversion in pork roasts cooked to 65 C after holding for 15 minutes at room temperature.  

As storage time increased in our study, a* values generally declined. Van Laack et al. (1996) made a similar observation in some patties that were frozen for one year at –27 C. However, they also observed an increase in a* after frozen storage. It was unclear what caused such variation.  

Although a* increased with time, indicating an increase in redness, this was not supported by the reflectance estimate of oxymyoglobin in the meat age study. The reflectance estimator for OMb is recommended for use on fresh meat (AMSA, 1991). In our fresh meat age study, the reflectance estimator results appeared to be in contrast to the a* results relative to air exposure time. In cooked meat, if present at all, oxymyoglobin levels should only represent a minor constituent. Therefore, the reflectance estimator for OMb may not be a reliable estimator for OMb in cooked meat. Van Laack et al. (1996) also found that a ratio of absorbance at 630 and 580 nm could not explain differences in cooked color in ground beef patties that were made from fresh and frozen beef.  

Cow beef as expected, had a higher pH than typically found in fed beef and among the beef sources used, variation in pH was found. High pH has been shown to limit myoglobin denaturation during heating (Trout, 1989). Van Laack et al. (1996) found that higher pH values were correlated with the incidence of red color from incomplete Mb denaturation in ground beef patties cooked to 71 C. This limited myoglobin denaturation would permit the formation of oxymyoglobin upon exposure to oxygen following cutting of the patties.  

Although not quantified, it was noted that the patties were not uniformly red upon color reversion. Oftentimes bright red regions were formed. This may be related to pH effects or higher myoglobin content within the ground particle. 

Implications 

Beef patties evaluated immediately after being removed from the grill can appear fully cooked on the inside but with time of exposure to air will undergo color reversion. Fresh as well as frozen beef can undergo color reversion. Therefore, consumers need to continue to be reminded that the most reliable food safety practice is to use a thermometer.