- Email: [email protected]
Colour and colour stability of hot processed frozen minced beef. Results from chemical model experiments tested under storage conditions
Meat Science 28 (1990) 87-97
Colour and Colour Stability of Hot Processed Frozen Minced Beef. Results from Chemical Model Experiments Tested under Storage Conditions
H. J. Andersen, ° G. Bertelsen b & L. H. Skibsted °* a Chemistry Department, Royal Veterinary and Agricultural University, Thorvaldsensvej 40, DK- 1871 Frederiksberg C, Denmark b Department of Food Preservation, Royal Veterinary and Agricultural University, Howitzvej 13, DK-2000 Frederiksberg, Denmark (Received 26 May 1989; revised version received 8 September 1989; accepted 10 October 1989)
A BS TRA C T Minced beef based on hot processed meat wasfound to show an improved red colour and, more significantly, to show a better colour stability with respect to brown discoloration, when compared to minced beef resulting from a traditional (cold) process. The colour was followed by tristimulus colorimetry during freezer storage (product temperature - 1 8 ° C ) in a display cabinet illuminated by fluorescent tubes. The development of brown discoloration seems to indicate other oxidative changes and to relate to rancidity as judged from a determination of thiobarbituric acid reactive substances in the mince surface. Fluorescent light and added salt. both known from chemical models to increase the autoxidation rate of red oxymyoglobin to brown metmyoglobin, greatly reduced the colour stability, especiaUy for cold deboned meat. pH of the hot deboned meat was significantly higher ( p H > 6"0) than that of the cold deboned meat (pH ranged from 5"5 to 5"8). The observations are discussed in relation to kinetic salt effect on the acid catalyzed autoxidation of oxymyoglobin and its role in the initiation of oxidative rancidity. * To whom all correspondence should be addressed. 87 Meat Science0309-1740/90/$03"50 © 1990 Elsevier Science Publishers Ltd, England. Printed in Great Britain
88
H.J. Andersen, G. Bertelsen, L. H. Skibsted
INTRODUCTION The gradual change in surface colour from red to brown, often encountered during storage and display of fresh and frozen meat, is largely a result of competitive thermal and photochemical autoxidation of the red oxymyoglobin (MbO2) to the brown metmyoglobin (MMb) (Andersen et aL, 1989). Besides being unacceptable to the consumer (Hood & Riordan, 1973), the brown surface discoloration seems to indicate other oxidative changes in the product, although the interaction between pigment oxidation and the oxidation of lipids and protein is not well understood (Ledward, 1987). However, superoxide is formed in the primary step of oxymyoglobin autoxidation (Gotoh & Shikama, 1976), and this radical-ion is capable, most likely through disproportionation to oxygen and hydrogen peroxide, of oxidizing myoglobin and thereby transforming this pigment into an efficient iron(IV) catalyst for lipid oxidation (Kanner et al., 1987; Harel & Kanner, 1988). Processing techniques and storage conditions should accordingly be optimized to improve colour and colour stability, especially for products such as minced beef with large surface areas. The kinetics and reaction mechanism of the competitive thermal and photochemical autoxidation ofoxymyoglobin have previously been studied under conditions relevant to meat and meat products (Bertelsen & Skibsted, 1987; Andersen et al., 1988b). From these model studies it was concluded that the rate increase noted for solutions with decreasing pH (George & Stratmann, 1954; Shikama & Sugaware, 1978) is due to a specific acid catalysis, which moreover is subject to a significant positive kinetic salt effect and, notably, that this salt effect vanishes gradually with increasing pH. These results imply that a better colour stability is expectable for hot deboned (pre-rigor) meat in which a high pH is maintained, as compared to meat from a traditional (cold) process with a decreasing pH resulting from postmortem glycolysis. Moreover, the colour stability of hot deboned meat should be less sensitive to added salt. This latter prediction, also based on the model experiments (Andersen et ai., 1988b), is of particular interest, since the addition of salt prior to freezing has been shown' to be important in maintaining the desired high pH and the high level of ATP characteristic of pre-rigor meat (Honikel & Hamm, 1978). In order to thoroughly understand the complex autoxidation mechanism ofoxymyoglobin and its relation to meat discoloration and lipid oxidation in meat, further investigations are clearly needed, in particular with respect to the applicability of the results obtained in chemical models to different meat products under conditions of storage and retail display. The exposure of the meat surface to light during display presents further complications, since the photochemical autoxidation is strongly dependent on the
Colour and colour stability of hot processed fro-_en minced bee/
89
wavelength distribution of light, as also shown in model experiments (Bertelsen & Skibsted, 1987). In order to test the applicability of the results obtained with chemical models, the authors have undertaken an investigation of the colour quality of frozen minced beef during storage and have examined the combined effect of light and added salt on pre-rigor, high pH beef in comparison with traditionally deboned beef with low pH. In addition, they have related surface discoloration to lipid oxidation by concomitant determination of surface colour and thiobarbituric acid reactive substances at the surface of the product.
MATERIALS A N D METHODS
Product and packaging Two forequarters from the same bovine were used for the production of minced beef. All muscles from each forequarter were used, and three batches of minced beef(batch size 8 kg) were prepared for each method ofdeboning: (1) pure beef; (2) beef plus 1-0% sodium chloride (Merck, analytical grade), and (3) beef plus 1-0% sodium chloride of commercial grade with 8 ppm K,L[Fe(CN)6-1 added as anticaking agent, as normally used in the Danish food industry. Hot deboned beef', one of the forequarters was deboned immediately after slaughtering. The beef was passed through a 3 cm plate and hand-mixed with the salt. The mince was chilled to a temperature of c. - I°C in a liquid-nitrogen freezer within 2 h of slaughtering (Cryosas Junior, L'air Liquid) and then passed through a 0.3 cm plate for the final grind. Cold deboned beef: the other forequarter of the carcass was conventionally chilled and stored in a chill storage room at 4°C for 6 days. After deboning the beefwas passed through a 3 cm plate, mixed with the salt, and then finally ground as described for the hot deboned beef. The resulting minced beef product, hot or cold deboned (c. 21% fat and c. 17% protein, as analysed by standard methods) was packed in 450g portions in polyethylene tubes with an oxygen transmission rate of c. 1000cm3/m2/24h/atm (25°C, rh 75%). The packs were frozen in a blast freezer (air temperature c. -35°C) to a centre temperature of c. - 10°C. For hot deboned samples, the product reached - 1 0 ° C in less than 6 h after slaughtering.
Storage The packs were placed in the upper layer in a freezer cabinet (gondola, with forced air circulation). Each of the packs was partially covered with black
90
H. J. Andersen, G. Bertelsen, L. H. Skibsted
plastic, allowing a direct comparison during storage of the colour of meat exposed to light and meat protected from light. The product temperature was approximately -18°C; however, during daily defrosting, the product temperature rose to approximately -12°C.
Light source Fluorescent tubes (Philips TLD 18W/36) were used for illumination giving an illuminance of 600 lx on the surface of the product, where a radiant flux density of UV-light (30(0400 nm) of c. 25 mW/m 2 was measured. Colour measurement The surface colour of the product was monitored during storage for approximately 30 days by tristimulus colorimetric measurement of each pack (Hunterlab D-25 equipped with a D25 M sensory head, standardized as described previously (Andersen et al., 1988a)).
pH measurement A Knick Digital model 653 pH meter and a direct insertion probe electrode (Ingold Lot 406-M3) were used for pH measurement. Assessment of lipid oxidation Lipid oxidation was determined by the 2-thiobarbituric acid (TBA) method of Vyncke (1975). The TBA values were expressed as milligrams of malonaldehyde per kilogram of product.
Bacterial sampling Microbial load was determined at the beginning and at the end of the storage period by spiral counting technique, using the selective plating media described in Table 1, after appropriate serial dilutions. TABLE 1 Microbial Determinations: Media and Incubation Conditions
Microflora
Total count Faecal streptococci Coli
Medium
Plate count agar (PCA) Slanetzagar Tryptone Soya agar/ Violet Red Bile agar
Time
Temperature
(h)
(°C)
72 48
25 37
24
44
Colour and colour stability of hot processed fro:en minced beef
91
RESULTS The colour quality of meat and meat products is determined by the initial colour as well as by the colour stability of the product. For the different minced beef products investigated in the present study, the colour was accordingly compared (by tristimulus colorimetry) at the beginning of, and regularly during, the 30-day period of display. Among the tristimulus colorimetric parameters (Hunter L, a and b), the a-value, in agreement with results previously obtained (Andersen et al., 1989), was found to provide the best quantitative description of the progressing colour change from bright red to brown which took place during display of the products. The product containing hot deboned beef appeared more red than the product made from cold deboned beef. The subjective appearance, which, among other factors, is dependent on the reflective properties of the product, was confirmed by the initial a-value, and the changing a-value during display is shown in Fig. 1 for products containing pure meat (Fig. I(A)) and for products with salt added (1.0%, Fig. I(B)). Both the initial a-value and the avalue at any stage of storage were higher for minced beef based on hot deboned meat than for minced beef based on meat deboned in a cold process. The colour quality, as defined above, is thus significantly enhanced by using hot processed meat, and, notably, this observation applies both to
6~"
o.....¢ • hot _,~bo,~ed, ___ light
5
I-
.
.
4.
.
.
8.
.
.
12
16 l i o I
~I! 1 2I 4
I
I
I
4
I
8 112116
|
2 .0. . . 2 4
Fig. I. Surface redness (measured as Hunter a value) of hot and cold dehoned minced beef, during display in a freezer cabinet (product temperature - 18°C) illuminated by fluorescent light. Product surface exposed to light (light) or rigorously protected against light (dark): A, pure beef; B, beef with 1-0% of sodium chloride, analytical grade, added. Standard deviation on measured a-values was estimated to be 0.2.
H../. Andersen, G. Bertelsen, L. H. Skibsted
92
products containing pure meat and to products with salt added. Minced beef based on hot deboned meat thus shows a better red colour, even after 16 days of dark storage, as compared to the initial colour of minced beef based on cold deboned meat. As for the discussion of colour stability, the redness Hunter-a parameter was normalized relative to its initial value, and Aa, the change in redness (Aa(t) = a ( t ) - a ( t = 0)), excludes differences in initial colour quality and will be used for the discussion of colour change during storage. The results presented in Fig. 1 are accordingly transformed to Aa in Fig. 2, which also includes a comparison of the change in redness during display for minced beef with added salt, the latter with and without anticaking agent. Figure 2 allows a direct comparison of the influence of three different factors on the colour stability: (i) the method ofdeboning (cold or hot) for the meat used in the production; (ii) the addition to the product of salt, with and without anticaking agent; (iii) the exposure of the product to fluorescent light during display v e r s u s storage in the dark. In agreement with the results obtained in other investigations (Taylor e t al., 1981), the method of deboning did not affect the bacteriological quality of the product significantly, as determined 1.0
1.0
0 -
0
1.0
-1.0 '\i
,
t
"..
"'" "" .,..
-2.0
-2.0
-3.0
"2.0
-4.0
-4.0
-5.0
-5.0 -6.0
-6.0
, -7.0
................ 2:1
"~
-8.0
o
"'":1:::'::::...
-7.0
.
No NOCI
.
.
1%
o
•
115 NOCI, C.8-
NaCl,
a.8.
-8.0
-9.0
-9.0
"I0.0
-10.0
'U
I
I
4
8
I
lf2
16 20
;4
!
28clays
I
I
4
8
;2
I
|
I
16
20
24
I
t
28da
Fig. 2. Colour stability (measured as changes in Hunter a value) of cold (A) or hot (B) deboned minced beef, non-salted or salted with 1.0% sodium chloride, analytical grade (a.g.) or commercial grade (e.g.), during display in a freezer cabinet with fluorescent light. Product surface exposed to light (light) or rigorously protected against light (dark).
Colour and colour stability of hot processed frozen minced beef
93
by microbial counts at the beginning and at the end of storage, and it was concluded that the noted differences in colour stability were not a result of differences in the bacteriological quality. The total count (log PCA) varied from 4.7 to 6.4 for hot-deboned samples and from 4.9 to 5.7 for productions containing cold deboned meat. Non-salted samples showed the highest microbial counts. Most notable is the dramatic effect of added salt on the colour stability of minced beef produced from cold deboned meat, an effect which is further enhanced by exposure of the product to light (Fig. 2(A)). From a comparison with the effect of light on minced beef based on non-salted, cold deboned beef, it is also evident that, under the actual storage conditions, the addition of salt causes more severe discoloration than does the exposure of the product to light. The storage conditions used in the experiment were similar to those currently in use for display in the retail trade, making this result of immediate interest to the marketing of minced beef products. However, of more general interest is the fact that by each of the deboning methods, discoloration caused by exposure to light and discoloration caused by the addition of salt are additive to a fair approximation. The effect ofdeboning methods (and product pH, Table 2) is less evident, as discussed below in relation to the kinetic models developed. As may also be seen from Fig. 2, discoloration proceeded at almost the same rate for products with salt of analytical grade and for products with salt of commercial grade under all storage conditions used, excluding any effect on the colour stability from processing aids such as K4[Fe(CN)6] used as anticaking agents. Light and salt promote not only discoloration but also lipid oxidation. The a m o u n t of thiobarbituric acid reactive substances in the surface of the product increased during storage (Table 3), and as for the colour, TABLE 2 pH of Frozen, Minced Beef Produced from Hot and Cold
Deboned Meat Prior to and After Storage Treatment
Non-salted Salted. I%: Anal. grade Comm. grade
Hot deboned
Cold deboned
Initial*
Final b
Initial*
Final b
6-6
5.9
5-8
5.6
6.4 6-5
6-2 6-1
5.7 5.7
5-7 5-5
" pH measured at I-2°C in the mince before packaging. b pH measured on thawed samples, stored for 30 days in a display cabinet.
94
H..L Andersen, G. Bertelsen, L. H. Skibsted
TABLE 3 Lipid Oxidation in Surface* of Frozen, Minced Beef Produced from Hot and Cold Deboned Meat, Prior to and After Storage, Measured as TBA-Value b Treatment
Hot deboned Initial"
Non-salted Salted, I%: Anal. grade NaCI Comm. grade NaCI
Cold deboned
Final d L~ht
Da~
0-27
0.58
0-30
0.36 0"33
0.78 1.06
0.61 0-98
Initial"
Final ~ L~ht
Da~
--
0-77
0.43
0-47 0-47
1.18 1'65
0.96 1.01
Analysis samples {10 g) were taken from a c. 10 mm deep layer in the upper surface of the product. Analyzed by the method of Vyncke (1975), and given as milligrams of malonaldehyde per kilogram of product. c Determined prior to storage. Determined after freezer storage for 30 days in a display cabinet (product temperature- 18~C) illuminated by fluorescent tubes. Product surface either exposed to light (600 Ix) or rigorously protected against light.
susceptibility to oxidative changes was found to be larger for the product based on cold deboned beef than for the product based on hot deboned beef. Salt and light each increased the TBA value, and again, for each method of deboning, the effect of added salt and the effect of exposure to light can be added to the combined effect. However, a notable difference was found between the effect of salt of analytical grade and of salt of commercial grade on lipid oxidation. In contrast to what was observed for the oxidation of oxymyoglobin, salt of commercial grade increased lipid oxidation relative to salt of analytical grade, as measured by the TBA method. Admittedly, the results shown in Table 3 include only determination prior to storage and after 30 days of storage and only one determination for each set of conditions. Accordingly, caution should be exercised in a too detailed interpretation. However, the trend shown by these results deserves further attention, and we are currently investigating the role of KJFe(CN)6] and salt as possible prooxidants in relation to the kinetic model developed.
DISCUSSION Hot deboned (pre-rigor) meat offers a number of advantages including reduced cooling space, chilling time and, consequently, reduced energy
Colour and colour stability of hot processed fro:en minced beef
95
requirements (Cuthbertson, 1980). Furthermore, pre-rigor meat is particularly suitable for the production of certain products such as sausages, because of improved water-holding properties, higher emulsifying capacity and lower cooking losses (Hamm, 1972, 1973). These superior processing properties are all related to a high level of ATP in pre-rigor meat, which is maintained coincident with a high pH (Hamm, 1980; Hamm et al., 1980). The result obtained in the present study shows that the colour quality of minced beef during freezer storage is also greatly improved when the product is based on hot deboned meat. Moreover, oxidative changes in the surface of the product based on hot deboned meat are less sensitive to the addition of salt, as judged from the rate of the brown discoloration (caused by oxymyoglobin autoxidation) as well as from the increase in TBA-value during storage. Minced beef with added salt is a common frozen meat product. According to the result obtained in the present study, a better general oxidative stability may be expected during freezer storage when the production is based on hot deboned meat, as compared to cold deboned meat. Salt has previously been found to promote oxidation of both pigments and lipids in ground beef (Tortes et al., 1988). However, most likely because of the relatively short storage period (96 h) covered in the previous study, no difference was found in the degree of lipid oxidation between pre-rigor and post-rigor beef, despite the fact that the pigment oxidation proceeded significantly faster in the latter (Tortes et al., 1988). On the basis of the present result, the authors suggest that acid-catalysed oxymyoglobin oxidation, enhanced by added salt as a result of a primary kinetic salt effect (Andersen et al., 1988b) or by exposure to light (Bertelsen & Skibsted, 1987; Andersen et al., 1989), initiates lipid oxidation with a certain time delay. Admittedly, the experimental data are still very limited and more detailed experiments are in progress. However, when compared with results obtained in model systems, a consistent picture seems to emerge. The autoxidation of oxymyoglobin in the product surface which is responsible for the brown discoloration is the result of a competitive photochemical and thermal process for which the total rate, Vto~,, may be expressed as: V,o~a= Vpho~o+ Vtherm(pH, cs~,t) The photochemical rate, Vpho,o, is, at least under moderate conditions, independent ofpH and salt concentration (Andersen, H. J. & Skibsted, L. H., unpublished), in contrast to the thermal oxidation, for which the rate, Vthe~m(pH, c~z,), increases with increasing salt concentration and decreasing pH (Andersen et al., 1988b). As for the actual storage conditions, it is seen directly from Fig. 2(A) that the photochemical and thermal autoxidation is
96
H. J. Andersen, G. Bertelsen, L. H. Skibsted
almost equally important to the discoloration of non-salted product based on cold deboned beef. For the acidity conditions prevailing in cold deboned meat (pH = 5.7, and only slightly decreasing during storage, Table 2), the effect of added salt may be calculated from the aqueous solution models to amount to a factor of 1.4 for the thermal part (Vtlm.m(l'0% salt)= 1.4 x vth,rm(pure beef), combining transition state theory and an extended Debye-Hiickel expression (Andersen et al., 1988b)). Subtracting the photochemical contribution from the total rate, it is seen--also from Fig. 2(A)---that the effect of added salt on the thermal rate in the actual product is somewhat larger than expected (Vthcrm(l'0% salt)~, 3 x Vtherm(pure beef)). However, the qualitative agreement is encouraging, particularly since the salt effect on the thermal part decreases at the high pH of hot deboned beef (t,th~r,~(1.0% salt) ~ 2 x rLh~rm(purebeef), Fig. 2(B)), which was to be expected according to the kinetic model. As regards the non-salted, hot deboned mince, the product pH decreased by 0.7 units to 5.9, the latter value being hardly different from the pH of the cold deboned minces (Table 2). With a daily defrosting, during which the product temperature rose to approximately - 12°C, post-mortem glycolysis is not completely halted in a non-salted product (Fischer et al., 1980), and the product pH is expected to slowly approach the pH of a cold deboned product. For the non-salted, hot deboned product, the autoxidation rate thus adjusts itself to the value of the decreasing pH according to the proposed kinetic model, in effect accelerating discoloration during storage. In the actual storage experiment, discoloration of the non-salted, hot deboned product followed this pattern, and discoloration proceeded to the same extent as for the cold deboned mince after approximately 20 days, in clear contrast to the better colour stability of the initial stages of storage. The TBA-value (Table 3) indicates, however, that lipid oxidation after 30 days of storage is less significant in the hot deboned product as compared to the cold deboned product. Notably, at this stage of storage, and for both deboning methods, oxymyoglobin oxidation has resulted in the same degree of discoloration. However, the initial stage with a high pH and a low pigment oxidation rate for the hot deboned product apparently has partly postponed lipid oxidation. This observation adds support to a model entailing pigment oxidation as an initiator of lipid oxidation. In conclusion, the present study has shown that ground beef products based on hot deboned meat are of a better colour quality, both with respect to initial colour and to colour stability during storage, and that the colour is less sensitive to added salt. The generally better oxidative stability found for products based on hot deboned beef, with or without added salt, can be understood on the basis of the kinetic model developed previously.
Colour and colour stability of hot processedJ'rozen minced beef
97
ACKNOWLEDGEMENTS The present research was supported by a joint grant from the Danish Technical Research Council and the Danish Agricultural and Veterinary Research Council. The authors wish to thank the Danish Meat Trade School, in particular Berit Larsen, for the production of the minced beef.
REFERENCES Andersen, H. J., Bertelsen, G. & Skibsted, L. H. (1989). Meat Sci., 25, 155. Andersen, H. J., Bertelsen, G., Boegh-Soerensen, L., Shek, C. K. & Skibsted, L. H. (1988a). Meat Sci., 22, 283. Andersen, H. J., Bertelsen, G. & Skibsted, L. H. (1988h). Acta Chem. Scand., A42, 226. Bertelsen, G. & Skibsted, L. H. (1987). Meat Sci., 19, 243. Cuthbertson, A. (1980). In Developments in Meat Science 1, ed. R. Lawrie, Applied Science Publishers Ltd, London, p. 61. Fischer, Ch., Honikel, K. O. & Hamm, R. (1980). Z. Lebensm. Unters. Forsch., 171, 200. George, P. & Stratmann, C. J. (1954). Biochem. J., 57, 568. Gotoh, T. & Shikama, K. (1976). J. Biochem., 80, 397. Harem, R. (1972). Kolloidchemie des Fleisches, Paul Parey Verlag, Berlin, Hamburg. Harem, R. (1973). Fleischwirtsch., 53, 73. Harem, R. (1980). In Developments in Meat Science 1, ed. R. Lawrie. Applied Science Publishers Ltd, London, p. 93. Harem, R., Honikel, K. O., Fischer, Ch. & Hamid, A. (1980). Fh, ischwirtsch., 60, 1567. Harel, S. & Kanner, J. (1988). Free Rad. Res. Comms., 5, 21. Honikel, K. O. & Harem, R. (1978). Meat Sci., 2, 181. Hood, D. E. & Riordan, E. B. (1973). J. Food Technol., 8, 333. Kanner, J., German, J. B. & Kinsella, J. F. (1987). CRC Crit. Rev. FoodSei. Nutr., 25, 317. Ledward, D. (1987). Food Sci. Technol. Today, 1, 153. Shikama, K. & Sugaware, Y. (1978). Eur. J. Biochem., 91,407. Taylor, A. A., Shaw, B. G. & MacDougall, D. M. (1981). Meat Sci., 5, 109. Torres, F., Pearson, A. M., Gray, J. I., Booren, A. M. & Shimokomaki, M. (1988). Meat Sci., 23, ! 5 i. Vyncke, W. (1975). Fette, Seifen, Anstriehmittel, 77, 239.
Colour and Colour Stability of Hot Processed Frozen Minced Beef. Results from Chemical Model Experiments Tested under Storage Conditions
H. J. Andersen, ° G. Bertelsen b & L. H. Skibsted °* a Chemistry Department, Royal Veterinary and Agricultural University, Thorvaldsensvej 40, DK- 1871 Frederiksberg C, Denmark b Department of Food Preservation, Royal Veterinary and Agricultural University, Howitzvej 13, DK-2000 Frederiksberg, Denmark (Received 26 May 1989; revised version received 8 September 1989; accepted 10 October 1989)
A BS TRA C T Minced beef based on hot processed meat wasfound to show an improved red colour and, more significantly, to show a better colour stability with respect to brown discoloration, when compared to minced beef resulting from a traditional (cold) process. The colour was followed by tristimulus colorimetry during freezer storage (product temperature - 1 8 ° C ) in a display cabinet illuminated by fluorescent tubes. The development of brown discoloration seems to indicate other oxidative changes and to relate to rancidity as judged from a determination of thiobarbituric acid reactive substances in the mince surface. Fluorescent light and added salt. both known from chemical models to increase the autoxidation rate of red oxymyoglobin to brown metmyoglobin, greatly reduced the colour stability, especiaUy for cold deboned meat. pH of the hot deboned meat was significantly higher ( p H > 6"0) than that of the cold deboned meat (pH ranged from 5"5 to 5"8). The observations are discussed in relation to kinetic salt effect on the acid catalyzed autoxidation of oxymyoglobin and its role in the initiation of oxidative rancidity. * To whom all correspondence should be addressed. 87 Meat Science0309-1740/90/$03"50 © 1990 Elsevier Science Publishers Ltd, England. Printed in Great Britain
88
H.J. Andersen, G. Bertelsen, L. H. Skibsted
INTRODUCTION The gradual change in surface colour from red to brown, often encountered during storage and display of fresh and frozen meat, is largely a result of competitive thermal and photochemical autoxidation of the red oxymyoglobin (MbO2) to the brown metmyoglobin (MMb) (Andersen et aL, 1989). Besides being unacceptable to the consumer (Hood & Riordan, 1973), the brown surface discoloration seems to indicate other oxidative changes in the product, although the interaction between pigment oxidation and the oxidation of lipids and protein is not well understood (Ledward, 1987). However, superoxide is formed in the primary step of oxymyoglobin autoxidation (Gotoh & Shikama, 1976), and this radical-ion is capable, most likely through disproportionation to oxygen and hydrogen peroxide, of oxidizing myoglobin and thereby transforming this pigment into an efficient iron(IV) catalyst for lipid oxidation (Kanner et al., 1987; Harel & Kanner, 1988). Processing techniques and storage conditions should accordingly be optimized to improve colour and colour stability, especially for products such as minced beef with large surface areas. The kinetics and reaction mechanism of the competitive thermal and photochemical autoxidation ofoxymyoglobin have previously been studied under conditions relevant to meat and meat products (Bertelsen & Skibsted, 1987; Andersen et al., 1988b). From these model studies it was concluded that the rate increase noted for solutions with decreasing pH (George & Stratmann, 1954; Shikama & Sugaware, 1978) is due to a specific acid catalysis, which moreover is subject to a significant positive kinetic salt effect and, notably, that this salt effect vanishes gradually with increasing pH. These results imply that a better colour stability is expectable for hot deboned (pre-rigor) meat in which a high pH is maintained, as compared to meat from a traditional (cold) process with a decreasing pH resulting from postmortem glycolysis. Moreover, the colour stability of hot deboned meat should be less sensitive to added salt. This latter prediction, also based on the model experiments (Andersen et ai., 1988b), is of particular interest, since the addition of salt prior to freezing has been shown' to be important in maintaining the desired high pH and the high level of ATP characteristic of pre-rigor meat (Honikel & Hamm, 1978). In order to thoroughly understand the complex autoxidation mechanism ofoxymyoglobin and its relation to meat discoloration and lipid oxidation in meat, further investigations are clearly needed, in particular with respect to the applicability of the results obtained in chemical models to different meat products under conditions of storage and retail display. The exposure of the meat surface to light during display presents further complications, since the photochemical autoxidation is strongly dependent on the
Colour and colour stability of hot processed fro-_en minced bee/
89
wavelength distribution of light, as also shown in model experiments (Bertelsen & Skibsted, 1987). In order to test the applicability of the results obtained with chemical models, the authors have undertaken an investigation of the colour quality of frozen minced beef during storage and have examined the combined effect of light and added salt on pre-rigor, high pH beef in comparison with traditionally deboned beef with low pH. In addition, they have related surface discoloration to lipid oxidation by concomitant determination of surface colour and thiobarbituric acid reactive substances at the surface of the product.
MATERIALS A N D METHODS
Product and packaging Two forequarters from the same bovine were used for the production of minced beef. All muscles from each forequarter were used, and three batches of minced beef(batch size 8 kg) were prepared for each method ofdeboning: (1) pure beef; (2) beef plus 1-0% sodium chloride (Merck, analytical grade), and (3) beef plus 1-0% sodium chloride of commercial grade with 8 ppm K,L[Fe(CN)6-1 added as anticaking agent, as normally used in the Danish food industry. Hot deboned beef', one of the forequarters was deboned immediately after slaughtering. The beef was passed through a 3 cm plate and hand-mixed with the salt. The mince was chilled to a temperature of c. - I°C in a liquid-nitrogen freezer within 2 h of slaughtering (Cryosas Junior, L'air Liquid) and then passed through a 0.3 cm plate for the final grind. Cold deboned beef: the other forequarter of the carcass was conventionally chilled and stored in a chill storage room at 4°C for 6 days. After deboning the beefwas passed through a 3 cm plate, mixed with the salt, and then finally ground as described for the hot deboned beef. The resulting minced beef product, hot or cold deboned (c. 21% fat and c. 17% protein, as analysed by standard methods) was packed in 450g portions in polyethylene tubes with an oxygen transmission rate of c. 1000cm3/m2/24h/atm (25°C, rh 75%). The packs were frozen in a blast freezer (air temperature c. -35°C) to a centre temperature of c. - 10°C. For hot deboned samples, the product reached - 1 0 ° C in less than 6 h after slaughtering.
Storage The packs were placed in the upper layer in a freezer cabinet (gondola, with forced air circulation). Each of the packs was partially covered with black
90
H. J. Andersen, G. Bertelsen, L. H. Skibsted
plastic, allowing a direct comparison during storage of the colour of meat exposed to light and meat protected from light. The product temperature was approximately -18°C; however, during daily defrosting, the product temperature rose to approximately -12°C.
Light source Fluorescent tubes (Philips TLD 18W/36) were used for illumination giving an illuminance of 600 lx on the surface of the product, where a radiant flux density of UV-light (30(0400 nm) of c. 25 mW/m 2 was measured. Colour measurement The surface colour of the product was monitored during storage for approximately 30 days by tristimulus colorimetric measurement of each pack (Hunterlab D-25 equipped with a D25 M sensory head, standardized as described previously (Andersen et al., 1988a)).
pH measurement A Knick Digital model 653 pH meter and a direct insertion probe electrode (Ingold Lot 406-M3) were used for pH measurement. Assessment of lipid oxidation Lipid oxidation was determined by the 2-thiobarbituric acid (TBA) method of Vyncke (1975). The TBA values were expressed as milligrams of malonaldehyde per kilogram of product.
Bacterial sampling Microbial load was determined at the beginning and at the end of the storage period by spiral counting technique, using the selective plating media described in Table 1, after appropriate serial dilutions. TABLE 1 Microbial Determinations: Media and Incubation Conditions
Microflora
Total count Faecal streptococci Coli
Medium
Plate count agar (PCA) Slanetzagar Tryptone Soya agar/ Violet Red Bile agar
Time
Temperature
(h)
(°C)
72 48
25 37
24
44
Colour and colour stability of hot processed fro:en minced beef
91
RESULTS The colour quality of meat and meat products is determined by the initial colour as well as by the colour stability of the product. For the different minced beef products investigated in the present study, the colour was accordingly compared (by tristimulus colorimetry) at the beginning of, and regularly during, the 30-day period of display. Among the tristimulus colorimetric parameters (Hunter L, a and b), the a-value, in agreement with results previously obtained (Andersen et al., 1989), was found to provide the best quantitative description of the progressing colour change from bright red to brown which took place during display of the products. The product containing hot deboned beef appeared more red than the product made from cold deboned beef. The subjective appearance, which, among other factors, is dependent on the reflective properties of the product, was confirmed by the initial a-value, and the changing a-value during display is shown in Fig. 1 for products containing pure meat (Fig. I(A)) and for products with salt added (1.0%, Fig. I(B)). Both the initial a-value and the avalue at any stage of storage were higher for minced beef based on hot deboned meat than for minced beef based on meat deboned in a cold process. The colour quality, as defined above, is thus significantly enhanced by using hot processed meat, and, notably, this observation applies both to
6~"
o.....¢ • hot _,~bo,~ed, ___ light
5
I-
.
.
4.
.
.
8.
.
.
12
16 l i o I
~I! 1 2I 4
I
I
I
4
I
8 112116
|
2 .0. . . 2 4
Fig. I. Surface redness (measured as Hunter a value) of hot and cold dehoned minced beef, during display in a freezer cabinet (product temperature - 18°C) illuminated by fluorescent light. Product surface exposed to light (light) or rigorously protected against light (dark): A, pure beef; B, beef with 1-0% of sodium chloride, analytical grade, added. Standard deviation on measured a-values was estimated to be 0.2.
H../. Andersen, G. Bertelsen, L. H. Skibsted
92
products containing pure meat and to products with salt added. Minced beef based on hot deboned meat thus shows a better red colour, even after 16 days of dark storage, as compared to the initial colour of minced beef based on cold deboned meat. As for the discussion of colour stability, the redness Hunter-a parameter was normalized relative to its initial value, and Aa, the change in redness (Aa(t) = a ( t ) - a ( t = 0)), excludes differences in initial colour quality and will be used for the discussion of colour change during storage. The results presented in Fig. 1 are accordingly transformed to Aa in Fig. 2, which also includes a comparison of the change in redness during display for minced beef with added salt, the latter with and without anticaking agent. Figure 2 allows a direct comparison of the influence of three different factors on the colour stability: (i) the method ofdeboning (cold or hot) for the meat used in the production; (ii) the addition to the product of salt, with and without anticaking agent; (iii) the exposure of the product to fluorescent light during display v e r s u s storage in the dark. In agreement with the results obtained in other investigations (Taylor e t al., 1981), the method of deboning did not affect the bacteriological quality of the product significantly, as determined 1.0
1.0
0 -
0
1.0
-1.0 '\i
,
t
"..
"'" "" .,..
-2.0
-2.0
-3.0
"2.0
-4.0
-4.0
-5.0
-5.0 -6.0
-6.0
, -7.0
................ 2:1
"~
-8.0
o
"'":1:::'::::...
-7.0
.
No NOCI
.
.
1%
o
•
115 NOCI, C.8-
NaCl,
a.8.
-8.0
-9.0
-9.0
"I0.0
-10.0
'U
I
I
4
8
I
lf2
16 20
;4
!
28clays
I
I
4
8
;2
I
|
I
16
20
24
I
t
28da
Fig. 2. Colour stability (measured as changes in Hunter a value) of cold (A) or hot (B) deboned minced beef, non-salted or salted with 1.0% sodium chloride, analytical grade (a.g.) or commercial grade (e.g.), during display in a freezer cabinet with fluorescent light. Product surface exposed to light (light) or rigorously protected against light (dark).
Colour and colour stability of hot processed frozen minced beef
93
by microbial counts at the beginning and at the end of storage, and it was concluded that the noted differences in colour stability were not a result of differences in the bacteriological quality. The total count (log PCA) varied from 4.7 to 6.4 for hot-deboned samples and from 4.9 to 5.7 for productions containing cold deboned meat. Non-salted samples showed the highest microbial counts. Most notable is the dramatic effect of added salt on the colour stability of minced beef produced from cold deboned meat, an effect which is further enhanced by exposure of the product to light (Fig. 2(A)). From a comparison with the effect of light on minced beef based on non-salted, cold deboned beef, it is also evident that, under the actual storage conditions, the addition of salt causes more severe discoloration than does the exposure of the product to light. The storage conditions used in the experiment were similar to those currently in use for display in the retail trade, making this result of immediate interest to the marketing of minced beef products. However, of more general interest is the fact that by each of the deboning methods, discoloration caused by exposure to light and discoloration caused by the addition of salt are additive to a fair approximation. The effect ofdeboning methods (and product pH, Table 2) is less evident, as discussed below in relation to the kinetic models developed. As may also be seen from Fig. 2, discoloration proceeded at almost the same rate for products with salt of analytical grade and for products with salt of commercial grade under all storage conditions used, excluding any effect on the colour stability from processing aids such as K4[Fe(CN)6] used as anticaking agents. Light and salt promote not only discoloration but also lipid oxidation. The a m o u n t of thiobarbituric acid reactive substances in the surface of the product increased during storage (Table 3), and as for the colour, TABLE 2 pH of Frozen, Minced Beef Produced from Hot and Cold
Deboned Meat Prior to and After Storage Treatment
Non-salted Salted. I%: Anal. grade Comm. grade
Hot deboned
Cold deboned
Initial*
Final b
Initial*
Final b
6-6
5.9
5-8
5.6
6.4 6-5
6-2 6-1
5.7 5.7
5-7 5-5
" pH measured at I-2°C in the mince before packaging. b pH measured on thawed samples, stored for 30 days in a display cabinet.
94
H..L Andersen, G. Bertelsen, L. H. Skibsted
TABLE 3 Lipid Oxidation in Surface* of Frozen, Minced Beef Produced from Hot and Cold Deboned Meat, Prior to and After Storage, Measured as TBA-Value b Treatment
Hot deboned Initial"
Non-salted Salted, I%: Anal. grade NaCI Comm. grade NaCI
Cold deboned
Final d L~ht
Da~
0-27
0.58
0-30
0.36 0"33
0.78 1.06
0.61 0-98
Initial"
Final ~ L~ht
Da~
--
0-77
0.43
0-47 0-47
1.18 1'65
0.96 1.01
Analysis samples {10 g) were taken from a c. 10 mm deep layer in the upper surface of the product. Analyzed by the method of Vyncke (1975), and given as milligrams of malonaldehyde per kilogram of product. c Determined prior to storage. Determined after freezer storage for 30 days in a display cabinet (product temperature- 18~C) illuminated by fluorescent tubes. Product surface either exposed to light (600 Ix) or rigorously protected against light.
susceptibility to oxidative changes was found to be larger for the product based on cold deboned beef than for the product based on hot deboned beef. Salt and light each increased the TBA value, and again, for each method of deboning, the effect of added salt and the effect of exposure to light can be added to the combined effect. However, a notable difference was found between the effect of salt of analytical grade and of salt of commercial grade on lipid oxidation. In contrast to what was observed for the oxidation of oxymyoglobin, salt of commercial grade increased lipid oxidation relative to salt of analytical grade, as measured by the TBA method. Admittedly, the results shown in Table 3 include only determination prior to storage and after 30 days of storage and only one determination for each set of conditions. Accordingly, caution should be exercised in a too detailed interpretation. However, the trend shown by these results deserves further attention, and we are currently investigating the role of KJFe(CN)6] and salt as possible prooxidants in relation to the kinetic model developed.
DISCUSSION Hot deboned (pre-rigor) meat offers a number of advantages including reduced cooling space, chilling time and, consequently, reduced energy
Colour and colour stability of hot processed fro:en minced beef
95
requirements (Cuthbertson, 1980). Furthermore, pre-rigor meat is particularly suitable for the production of certain products such as sausages, because of improved water-holding properties, higher emulsifying capacity and lower cooking losses (Hamm, 1972, 1973). These superior processing properties are all related to a high level of ATP in pre-rigor meat, which is maintained coincident with a high pH (Hamm, 1980; Hamm et al., 1980). The result obtained in the present study shows that the colour quality of minced beef during freezer storage is also greatly improved when the product is based on hot deboned meat. Moreover, oxidative changes in the surface of the product based on hot deboned meat are less sensitive to the addition of salt, as judged from the rate of the brown discoloration (caused by oxymyoglobin autoxidation) as well as from the increase in TBA-value during storage. Minced beef with added salt is a common frozen meat product. According to the result obtained in the present study, a better general oxidative stability may be expected during freezer storage when the production is based on hot deboned meat, as compared to cold deboned meat. Salt has previously been found to promote oxidation of both pigments and lipids in ground beef (Tortes et al., 1988). However, most likely because of the relatively short storage period (96 h) covered in the previous study, no difference was found in the degree of lipid oxidation between pre-rigor and post-rigor beef, despite the fact that the pigment oxidation proceeded significantly faster in the latter (Tortes et al., 1988). On the basis of the present result, the authors suggest that acid-catalysed oxymyoglobin oxidation, enhanced by added salt as a result of a primary kinetic salt effect (Andersen et al., 1988b) or by exposure to light (Bertelsen & Skibsted, 1987; Andersen et al., 1989), initiates lipid oxidation with a certain time delay. Admittedly, the experimental data are still very limited and more detailed experiments are in progress. However, when compared with results obtained in model systems, a consistent picture seems to emerge. The autoxidation of oxymyoglobin in the product surface which is responsible for the brown discoloration is the result of a competitive photochemical and thermal process for which the total rate, Vto~,, may be expressed as: V,o~a= Vpho~o+ Vtherm(pH, cs~,t) The photochemical rate, Vpho,o, is, at least under moderate conditions, independent ofpH and salt concentration (Andersen, H. J. & Skibsted, L. H., unpublished), in contrast to the thermal oxidation, for which the rate, Vthe~m(pH, c~z,), increases with increasing salt concentration and decreasing pH (Andersen et al., 1988b). As for the actual storage conditions, it is seen directly from Fig. 2(A) that the photochemical and thermal autoxidation is
96
H. J. Andersen, G. Bertelsen, L. H. Skibsted
almost equally important to the discoloration of non-salted product based on cold deboned beef. For the acidity conditions prevailing in cold deboned meat (pH = 5.7, and only slightly decreasing during storage, Table 2), the effect of added salt may be calculated from the aqueous solution models to amount to a factor of 1.4 for the thermal part (Vtlm.m(l'0% salt)= 1.4 x vth,rm(pure beef), combining transition state theory and an extended Debye-Hiickel expression (Andersen et al., 1988b)). Subtracting the photochemical contribution from the total rate, it is seen--also from Fig. 2(A)---that the effect of added salt on the thermal rate in the actual product is somewhat larger than expected (Vthcrm(l'0% salt)~, 3 x Vtherm(pure beef)). However, the qualitative agreement is encouraging, particularly since the salt effect on the thermal part decreases at the high pH of hot deboned beef (t,th~r,~(1.0% salt) ~ 2 x rLh~rm(purebeef), Fig. 2(B)), which was to be expected according to the kinetic model. As regards the non-salted, hot deboned mince, the product pH decreased by 0.7 units to 5.9, the latter value being hardly different from the pH of the cold deboned minces (Table 2). With a daily defrosting, during which the product temperature rose to approximately - 12°C, post-mortem glycolysis is not completely halted in a non-salted product (Fischer et al., 1980), and the product pH is expected to slowly approach the pH of a cold deboned product. For the non-salted, hot deboned product, the autoxidation rate thus adjusts itself to the value of the decreasing pH according to the proposed kinetic model, in effect accelerating discoloration during storage. In the actual storage experiment, discoloration of the non-salted, hot deboned product followed this pattern, and discoloration proceeded to the same extent as for the cold deboned mince after approximately 20 days, in clear contrast to the better colour stability of the initial stages of storage. The TBA-value (Table 3) indicates, however, that lipid oxidation after 30 days of storage is less significant in the hot deboned product as compared to the cold deboned product. Notably, at this stage of storage, and for both deboning methods, oxymyoglobin oxidation has resulted in the same degree of discoloration. However, the initial stage with a high pH and a low pigment oxidation rate for the hot deboned product apparently has partly postponed lipid oxidation. This observation adds support to a model entailing pigment oxidation as an initiator of lipid oxidation. In conclusion, the present study has shown that ground beef products based on hot deboned meat are of a better colour quality, both with respect to initial colour and to colour stability during storage, and that the colour is less sensitive to added salt. The generally better oxidative stability found for products based on hot deboned beef, with or without added salt, can be understood on the basis of the kinetic model developed previously.
Colour and colour stability of hot processedJ'rozen minced beef
97
ACKNOWLEDGEMENTS The present research was supported by a joint grant from the Danish Technical Research Council and the Danish Agricultural and Veterinary Research Council. The authors wish to thank the Danish Meat Trade School, in particular Berit Larsen, for the production of the minced beef.
REFERENCES Andersen, H. J., Bertelsen, G. & Skibsted, L. H. (1989). Meat Sci., 25, 155. Andersen, H. J., Bertelsen, G., Boegh-Soerensen, L., Shek, C. K. & Skibsted, L. H. (1988a). Meat Sci., 22, 283. Andersen, H. J., Bertelsen, G. & Skibsted, L. H. (1988h). Acta Chem. Scand., A42, 226. Bertelsen, G. & Skibsted, L. H. (1987). Meat Sci., 19, 243. Cuthbertson, A. (1980). In Developments in Meat Science 1, ed. R. Lawrie, Applied Science Publishers Ltd, London, p. 61. Fischer, Ch., Honikel, K. O. & Hamm, R. (1980). Z. Lebensm. Unters. Forsch., 171, 200. George, P. & Stratmann, C. J. (1954). Biochem. J., 57, 568. Gotoh, T. & Shikama, K. (1976). J. Biochem., 80, 397. Harem, R. (1972). Kolloidchemie des Fleisches, Paul Parey Verlag, Berlin, Hamburg. Harem, R. (1973). Fleischwirtsch., 53, 73. Harem, R. (1980). In Developments in Meat Science 1, ed. R. Lawrie. Applied Science Publishers Ltd, London, p. 93. Harem, R., Honikel, K. O., Fischer, Ch. & Hamid, A. (1980). Fh, ischwirtsch., 60, 1567. Harel, S. & Kanner, J. (1988). Free Rad. Res. Comms., 5, 21. Honikel, K. O. & Harem, R. (1978). Meat Sci., 2, 181. Hood, D. E. & Riordan, E. B. (1973). J. Food Technol., 8, 333. Kanner, J., German, J. B. & Kinsella, J. F. (1987). CRC Crit. Rev. FoodSei. Nutr., 25, 317. Ledward, D. (1987). Food Sci. Technol. Today, 1, 153. Shikama, K. & Sugaware, Y. (1978). Eur. J. Biochem., 91,407. Taylor, A. A., Shaw, B. G. & MacDougall, D. M. (1981). Meat Sci., 5, 109. Torres, F., Pearson, A. M., Gray, J. I., Booren, A. M. & Shimokomaki, M. (1988). Meat Sci., 23, ! 5 i. Vyncke, W. (1975). Fette, Seifen, Anstriehmittel, 77, 239.