- Email: [email protected]
Effect of pre-slaughter physiological conditions on the oxidative stability of colour and lipid during chill storage of pork
Meat Science 58 (2001) 347±357
www.elsevier.com/locate/meatsci
Eect of pre-slaughter physiological conditions on the oxidative stability of colour and lipid during chill storage of pork Dorte Juncher a,*, Birgitte Rùnn b, Else T. Mortensen a, Poul Henckel c, Anders Karlsson c, Leif H. Skibsted a, Grete Bertelsen a a
Department of Dairy and Food Science, Royal Veterinary and Agricultural University, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark b Department of Mathematics and Physics, Royal Veterinary and Agricultural University, Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark c Department of Product Quality, Danish Institute of Agricultural Sciences, Research Centre Foulum, Box 50, DK-8830 Tjele, Denmark Received 12 June 2000; received in revised form 5 December 2000; accepted 12 December 2000
Abstract The physiological condition of the live animal was found to signi®cantly aect colour, lipid oxidation and water holding capacity of chill stored pork chops (M. Longissimus dorsi) in a study, where various pre-slaughter conditions were achieved by the following four treatments: (A) control; (B) subjected to treadmill exercise immediately prior to stunning; (C) given epinephrine injection 15 h prior to slaughter; and (D) given epinephrine injection 15 h before slaughter and further subjected to treadmill exercise immediately before stunning. The treatments resulted in variations in energy metabolites (glycogen, lactate, creatine phosphate, ATP) and ultimate pH (pHu), with the lowest pHu in chops from treatments A and B, and in signi®cantly dierent tristimulus colour L*-, a*- and b*-parameters, although the eect of treatment on colour was not consistent during the chill storage period of 6 days. Overall, chops from treatments A and B had signi®cantly higher L*- and b*-values (were paler and less blue) than chops from C and D during storage under conditions typical for retail trade. The initial a*-values were higher (redder) in chops from treatments A and B, but the colour, as judged by the a*-values, was less stable in meat from these treatments compared with treatments C and D. Lipid oxidation, evaluated by thiobarbituric acid reactive substances (TBARS) in the fresh meat, and drip loss, measured after 6 days of storage, were both signi®cantly higher in chops from treatments A and B compared to chops obtained from treatments C and D. Statistical analysis relating the pH and the level of various energy metabolites post-mortem in the individual animals to the measured quality parameters, revealed that pHu was the most important factor aecting product quality. In conclusion, over all product quality depends on obtaining a pHu in the narrow range where both meat quality parameters such as colour, lipid oxidation and drip loss as well as microbiological aspects have to be considered. # 2001 Elsevier Science Ltd. All rights reserved. Keywords: Colour stability; Drip loss; Glycogen; Lipid oxidation; pH; Pre-slaughter treatment
1. Introduction For the consumer the surface colour of meat is the most important quality attribute at the time of purchase. During the last decades concern has risen in the meat trade about the colour of pork, as pork becomes paler. The surface colour of meat depends on the quantity of myoglobin present, on its chemical state and also on the chemical and physical conditions of other * Corresponding author. Tel.: +45-35-28-32-35; fax: +45-35-2833-44. E-mail address: [email protected] (D. Juncher).
components in the meat (Renerre, 1990). Factors aecting the surface colour are related to dierences between breeds and even individual animals, the age of the animal at time of slaughter, pre-slaughter handling, the chilling process, methods of packaging, retail light exposure and time/temperature regime during storage (Renerre, 1990). The rate of post mortem pH decline and the ultimate pH (pHu) have been shown to be important in determining the paleness as well as for the water holding capacity of pork meat. Thus, a high rate of pH decline and a low pHu result in pale meat of low water-holding capacity (Briskey, 1964). In addition, a low pHu may
0309-1740/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved. PII: S0309-1740(00)00156-X
348
D. Juncher et al. / Meat Science 58 (2001) 347±357
result in more pronounced discolouration during chill storage as the rate of oxidation of oxymyoglobin to metmyoglobin is accelerated at low pH (Andersen, Bertelsen, & Skibsted, 1988). Discolouration and lipid oxidation are often coupled, and a high degree of discolouration may indicate decreased oxidative stability with subsequent development of o-¯avours during storage, especially after further processing involving addition of salt or cooking (Skibsted, Mikkelsen, & Bertelsen, 1998). The characteristics of the post mortem pH decline are determined by the physiological condition of the animal at stunning (Bendall, 1973; Warriss, Bevis, & Ekins, 1989). Although the physiological conditions before slaughter may be manipulated experimentally by epinephrine injection, previous experiments using this method (Guignot, Touraille, Ouali, Renerre, & Monin, 1994; Hatton, Lawrie, Ratcli, & Wayne, 1972; Owen & Lawrie, 1975; Yasosky, Aberle, Peng, Mills, & Judge, 1984) did not show any relationship between the physiological conditions of the muscles at the time of stunning and the meat quality during chill storage. The goal of the present study was to investigate the eect of pre-slaughter physiological conditions on colour, water holding capacity and lipid oxidation of chops from M. Longissimus dorsi during retail chill storage. The physiological conditions were manipulated by treadmill exercise and/or epinephrine injection in pigs without the gene for halothane sensitivity. Based on the results obtained in a series of experiments described by Henckel, Karlsson, and Petersen (1997), four sets of conditions were chosen to cover the wide range of variation in level of energy metabolites and pH post-mortem that may be observed in pigs under practical conditions. 2. Materials and methods 2.1. Muscle samples Twenty slaughter pigs (all without the gene for halothane sensitivity), as a part of the herd of 80 pigs raised at the Research Centre Foulum and used in the experiment described by Henckel, Karlsson, Oksbjerg, and Petersen (2000), were used in the present study. The pigs, from ®ve dierent litters, were fed with ®ve consecutive batches of pig feed. The composition of the pig feed batches and the content of vitamin E, dry matter, fat and peroxide values of the feed lipid are given by Henckel et al. (2000). From each litter four females were selected. Each pig was exposed to one of the four treatments: (A) control, no treatment; (B) exercise on a treadmill at a rate of 3.8 km/h for 10 min immediately prior to stunning; (C) 0.2 mg epinephrine/kg live weight 15 h pre slaughter; and
(D) 0.3 mg epinephrine/kg live weight 15 h pre slaughter and exercise on a treadmill at a rate of 3.8 km/h for 5 min immediately prior to stunning. Treatments A and B were given 1 ml of a sterile salt solution as a placebo at the same time as treatments C and D were administered epinephrine (Henckel et al., 2000). The four pigs from each litter (one pig per treatment) were slaughtered on the same day as described by Henckel et al. (2000). The levels of glycogen, creatine phosphate, lactate, ATP and IMP were measured in biopsies from M. Longissimus dorsi immediately before stunning and 1, 15, 30, 45 min, and 1 , 3, 6 and 24 h after exsanguination as described by Henckel, Karlsson, Jensen, Oksbjerg, and Petersen (2001). pH was measured at 1, 15, 30, 45 min, and 1, 3, 6 and 24 h after exsanguination as described by Henckel et al. (2000). Twenty-four hours post mortem, the carcasses were dissected and the right side M. Longissimus dorsi from each of the pigs from treatments A, B, C and D was placed in individual polyethylene bags and transported in a thermobox (1±3 C) from the Research Center Foulum to the Royal Veterinary and Agricultural University. 2.2. Handling and storage conditions Two days post mortem, M. Longissimus dorsi muscles were sliced into 2-cm chops, which were placed in individual plastic trays, wrapped in polyethylene and placed in an illuminated (¯uorescent tubes, Philips 18W/82, average illuminance of 900 lx on the surface of the products) chill cabinet for up to 6 days. The temperature of the chill cabinet was 4 C with an increase in temperature to approximately 8 C during daily defrosting, as monitored by a Grant datalogger. The chops were randomized for analyses to be performed after 0, 1, 4 and 6 days of chill storage; one chop on each day for the colour and chemical analyses and one for drip loss determination at day 0. At each analysis, the colour of the surface of each day of chop was measured, followed by cutting the chop into small pieces which were randomized for the analyses described later. When not otherwise noted, the analyses were made in duplicate. Samples for analyses of total iron and a-tocopherol were frozen in liquid nitrogen and stored at 80 C until analysis. 2.3. Analyses All chemicals used were of analytical grade, and water was puri®ed through a Millipore puri®cation train (Millipore Corp., Bedford, MA). 2.3.1. Colour 2.3.1.1. Colour measurement. The colour measurements were performed using a tristimulus colorimeter (Minolta Chroma Meter CR300, Minolta Corporation, NJ). The
D. Juncher et al. / Meat Science 58 (2001) 347±357
L*-, a*- and b*-values were measured 10 times on the surface of the chops at day 0 (after a blooming period of 2 h in dark at 4 C), and at day 1, 4 and 6 after slicing. 2.3.2. Lipid oxidation 2.3.2.1. TBARS. Lipid oxidation was measured as 2thiobarbituric acid reactive substances (TBARS) by the extraction method described by Vyncke (1975) with a few modi®cations: meat (10 g) from the surface (1 mm layer) of the chop was homogenized (Ultra Turrax T-25, Janke & Kunkel IKA-Labortechnik, Staufen, Germany) with 30 ml of a 7.5% trichloroacetic acid (TCA) solution including 0.1% propylgallate (PG) and 0.1% ethylenediaminetetraacetic acid, disodium salt (EDTA) for 45 s at 13,500 rpm and the homogenate was ®ltered through a Whatmann ®lter no. 42. The extract (5 ml) was mixed with 0.020 M thiobarbituric acid (5 ml), heated and cooled as described by Vyncke (1975). The absorbance was measured at 532 and 600 nm using a HP 8452A diode assay spectrophotometer (Hewlett Packard Co., Palo Alto, CA), and the absorbance difference, A532nm A600nm, was calculated with A600nm correcting for sample turbidity. TBARS, expressed as milligrams of malonaldehyd per kilogram of meat, was calculated using malondialdehyd-bis-(diethylacetate) as standard. 2.3.2.2. Dry matter. Dry matter determinations, for use for calculation of TBARS, were conducted by drying 2 g meat from the surface of the chop in an oven at 104 C for 4 h and weighing after cooling for 15 min. 2.3.2.3. Extraction of lipids. Meat (10 g) was homogenized by an Ultra Turrax (Ultra Turrax T-25, Janke & Kunkel IKA-Labortechnik, Staufen, Germany) with 100 ml chloroform/methanol (2:1 v/v) for 1 min at 13,500 rpm. After homogenization, 25 ml 1.0 mM CaCl2 solution was added and the sample was further homogenized using an Ultra Turrax and centrifuged (MSE Mistral 2000, Crawley, England) for 20 min at 1000 rpm. The chloroform phase was removed and the extraction procedure repeated. The chloroform phase containing the extracted lipids was dried by vacuum evaporation (BuÈchi RE 11, BuÈchi Laboratoriums-Technik AG, Flawill, Schweiz). Finally 22 ml chloroform/ methanol and 2.0 ml CaCl2 were added to the dried sample and the sample was mixed (Vortex-mixer VF2, Janke & Kunkel IKA-Labortechnik) and centrifuged for 20 min at 2500 rpm. The lipid phase was removed, dried by vacuum evaporation and weighed. The percent of intramuscular fat was calculated from the weight of total lipid obtained after solvent extraction and the weight of the meat. 2.3.2.4. Fatty acid composition of the total fat. Extracted lipid (10 mg) was transformed into its constituent
349
methyl esters as described by Jart (1997). Methyl esters (1 ml) was injected into a HP 5890 series II gas chromatograph (Hewlett-Packard) with a ¯ame-ionization detector and with a 25 m0.20 mm0.33 mm HP-FFAP column No. 19091F-102 (Hewlett-Packard) under the following oven temperature program and conditions: 50 C for 1 min; 50±180 C at 15 C/min; 180±220 C at 5 C/min; at 220 C for 10 min; injector temperature was 250 C, detector temperature 300 C. Helium was used as the carrier gas with a split ratio of 1:10 and a ¯ow was 1 ml/min. The relative contents of the fatty acids C4:0, C14:0, C16:0, C18:0, C18:1, C18:2, C18:3, C20:0, C20:4, C22:1 and C24:0 were calculated from the chromatograms and from an external standard containing methyl esters of the fatty acids. 2.3.3. -Tocopherol a-Tocopherol was extracted using the procedure of Buttris and Diplock (1984) with some modi®cations: 2 ml homogenate (4 g meat homogenized with 20 ml 1.15% KCl-solution using an Ultra Turrax blender for 30 s at 13,500 rpm) was added to 0.20 ml saturated potassium hydroxide and 2 ml 0.50% ethanolic pyrrogalol. The sample was Vortex-mixed for 10 s and saponi®ed at 70 C for 30 min. After cooling in ice water, and addition of 1.0 ml water and 4.0 ml hexane with 0.001% BHT, the sample was vigorously shaken, mixed on a Vortex mixer followed by centrifugation (Universal 16A, Hettich Zentrifugen, D-78532 Tuttlingen, Germany) for 5 min at 800g. The upper hexane layer was collected and the residue re-extracted with a further 3 ml of hexane containing 0.001% BHT. The combined fractions were evaporated under nitrogen and a-tocopherol was redissolved in 0.50 ml ethanol with 0.001% BHT. Quanti®cation of tocopherols was performed according to Jensen et al. (1997). 2.3.4. Total iron Total iron was measured according to the methods of the Danish Standardization Board (1982a, b) and of Gilman (1989) with some modi®cations: Meat (2 g) and concentrated nitric acid (7 ml) were kept overnight in a te¯on container. The sample was destroyed in a closed te¯on container in a microwave oven (CEM MDS 81D, CEM Corporation, Matthews NC) with the following program: 100% power for 5 min to a pressure of about 120 Psi followed by cooling and ventilating of the container, 75% power for 10 min to a pressure of about 120 Psi followed by cooling and ventilating of the container, 100% power for 15 min to a pressure of < 120 Psi followed by cooling and ventilating. Five droplets of hydrogen peroxide (30%) were added and the sample was kept overnight in the container. The ashed sample was transfered to a 10 ml ¯ask with water and the iron content in the resulting solution measured by atomic absorptions spectroscopy (Perkin Elmer 3300, The
350
D. Juncher et al. / Meat Science 58 (2001) 347±357
Perkin-Elmer Corporation, Norwalk, CT). The concentration of total iron in the sample was calculated from a standard curve. 2.3.5. Drip loss Drip loss was measured on one chop from each muscle by the method of Barton-Gade et al. (1993) with some modi®cations: The chop was weighed immediately after cutting, hung in a laminate bag, closed tightly with string. The chop was weighed again after 6 days at 4 C, and the drip loss calculated. The drip loss was measured on the last three litters of the storage experiment. 2.4. Statistical analyses The pre-slaughter physiological conditions were expressed by the following physiological variables measured immediately before stunning and after exsanguination, namely pH, glycogen, lactate, creatine phosphate, ATP and IMP taken 4 min before stunning and 1, 15, 30, 45 min, and 1, 3, 6 and 24 h after exsanguination (pH only after exsanguination). For each variable, the nine measurements over time were concentrated into three parameters ( , and ) for each pig by the non-linear normal models given in Table 1. As shown in Table 1 the measurements of pH and glycogen were modelled as an exponential function decreasing in time, the variables lactate and IMP were modelled by a S-shaped function of time, and the variables creatine phosphate and ATP were modelled as very steep exponential functions decreasing in time. The non-linear models were ®tted by PROC NLIN in SAS (1996). To investigate whether the treatments resulted in different physiological variables for the pigs, all the estimated parameters were subjected to analysis of variance with treatment and litter included as ®xed eects. The development in colour, measured as L*-, a*- and b*-values, and in lipid oxidation, measured as TBARS, over time were analyzed by analysis of variance, where treatment, storage time and the interaction between
treatment and storage time were included as ®xed eects. Individual pig, litter and the interaction between litter and storage time were included as random eects. Within this model contrasts between treatment at day 0 were estimated and tested by pairwise comparison via ttests. The treatment eect on colour stability, evaluated as the decrease in the a*-value from day 1 to day 6, was tested similarly. The response variable TBARS was transformed with the natural logarithm prior to analysis in order to obtain homogeneous variance. The eects of treatment on colour and lipid oxidation found in the above analysis may be caused by dierences in pre-slaughter physiological conditions induced by the four treatments. Hence the physiological variables, expressed by the 18 estimated parameters, together with total iron content, a-tocopherol, polyunsaturated fatty acids (expressed as fatty acid with two or more double bonds and as fatty acids with three or more double bonds) were included, one at a time, in a covariance analysis, where treatment, storage time, interaction between treatment and storage time and interaction between the physiological parameter and storage time were included as ®xed eects. Individual pig, litter and the interaction between litter and storage time were included as random eects. Finally the variables showing signi®cant eects on the colour or TBARS values were included simultaneously in a covariance analysis with treatment, storage time and the interaction between treatment and storage time included as ®xed eects, and individual pig, litter and the interaction between litter and storage time included as random eects. Backwards selection was made to obtain the ®nal model for changes in the colour values and TBARS. Drip loss was measured on the chops from pigs from three of the litters and the observations were analysed by an analysis of variance with treatment included as a ®xed eect and litter included as a random eect. The treatment eect on drip loss was analyzed as described for colour and lipid oxidation, but within the simpler model with treatment included as ®xed eect and
Table 1 Models used to describe development in pH, glycogen, lactate, creatine phosphate, ATP and IMP measured in M. Longissimus dorsi immediately before stunning (not for pH) and after exsanguination for up to 24 h Parameter
Model for parameter ®ttinga
Parameters signi®cantly aected by the treatmentsb
pH Glycogen Lactate Creatine phosphate (cp) ATP IMP
pH=a(1+exp( b(time- g))) glycogen=a(1+exp( b(time g))) lactate=a exp(g btime exp( time)) cp=a+g(exp(btime) 1) ATP=a+g(exp(btime) 1) IMP=a/(1+1/g exp( 1/btime))
a*** a*** a***, g** a**, b** a**, g**
a a is a common parameter for the ultimate level of pH, glycogen, lactate and IMP and for the level at t=0 for cp and ATP, b is a common parameter for the rate of decrease in pH, glycogen, cp and ATP, and for the increase in lactate or IMP, and g is a common parameter that re¯ects the time of the decrease in pH and glycogen and the total change in level of lactate, IMP, cp and ATP. b *) **) ***) Parameters are signi®cantly aected by the treatment A: control, B: treadmill exercise, C: epinephrine injection, D: epinephrine injection and treadmill exercise (*:P<0.05, **:P<0.01 and ***:P<0.001).
D. Juncher et al. / Meat Science 58 (2001) 347±357
individual pig as random eect. Backwards selection was made to obtain the ®nal model for drip loss. All analyses of variance and covariance were ®tted by PROC MIXED in SAS (1996). Signi®cant treatment eects for intramuscular fat (IMF), polyunsaturated fatty acids, a-tocopherol and total iron were classi®ed by LSD (SAS, 1996). 3. Results 3.1. Pre-slaughter physiological conditions For each of the 20 pigs the changes in pH, glycogen, lactate, creatine phosphate (cp), ATP and IMP were measured before and up to 24 h after stunning (pH only after stunning) as reported by Henckel et al. (2000, 2001). The values for these six physiological parameters were ®tted to a non-linear model as described in Section 2, and in Fig. 1 the development in pH with time is shown as an example. It appears that the models ®t well to the observed values in each of the individual 20 pigs. Even though some variability in pH development is seen between the litters, a clear treatment eect is seen (Fig. 1). For the other physiological parameters plots of the function described in Section 2 gave plots of comparable ®tness. For each of the physiological variables, the information on the development over time was concentrated into three parameters , and , as described in Section 2 (Table 1) together with the results of the analyses of variance made to investigate whether the treatments result in dierences in the development of energy metabolites (glycogen, lactate, cp, ATP and IMP) and in the development of pH. The most pronounced eect of preslaughter treatment (A, B, C and D) was found in the ultimate level ( ) of pH, glycogen and lactate (Table 1). Thus, pigs from treatments A and B had signi®cantly (P<0.001) lower pHu and signi®cantly (P<0.001) higher ultimate levels of glycogen and lactate compared with pigs from the treatments C and D (Table 2). It also
351
appears from Table 1, that the treatment did not aect the rate of decrease or rate of increase of any of the measured energy metabolites or of pH, except for a weaker eect on the rate of decrease of cp. 3.2. Colour and colour stability The eect of pre-slaughter treatment on initial colour and on colour stability is shown in Fig. 2. The analysis of variance showed a signi®cant eect of the treatment, but the eect of the treatment was not the same during the entire storage period (signi®cant interaction between treatment and storage time (P<0.001 for L*- and a*values and P<0.01 for b*-values)). At the beginning of the storage period (day 0) and during storage, chops from treatments A and B had signi®cantly (P<0.001 and P<0.05 for day 0 and during storage, respectively) higher L*-values (more white) and b*-values (less blue) than chops from treatments C and D, and L*-values of chops from treatment C were signi®cantly (P<0.05) higher than L*-values of chops from treatment D (Fig. 2). Chops from treatments A and B had signi®cantly (P<0.001) higher initial a*-values than chops from treatments C and D, indicating a redder colour in the ®rst two. However, the colour stability during chill storage, evaluated from the decrease in a*-values from day 1 to day 6, was signi®cantly (P<0.01) inferior in chops from treatments A and B compared with chops from treatments C and D (Fig. 2). The higher initial a*values related to the higher b*-values indicate that chops from treatments A and B had more myoglobin in the oxymyoglobin form than chops from treatments C and D at the beginning of the storage period. This agrees with the results of Hunt, Sùrheim, and Slinde (1999) in a study of colour traits of uncooked fresh ground beef patties, in which the highest a*- and b*values were found in patties containing predominantly oxymyoglobin compared with patties containing predominantly deoxy- or metmyoglobin. In order to investigate whether the eects of treatment on the L*-, a*- and b*-values may be caused by dierences
Table 2 Estimated ultimate levels of pH, glycogen and lactate in M. Longissimus dorsi and intramuscular fat content (IMF), fatty acid composition, atocopherol content and total iron content measured at 48 h post mortem in chops (M. Longissimus dorsi) Pre-slaughter treatmenta
pHub
Glycogenub (mmole/kg meat)
Lactateub (mmole/kg meat)
IMFc (g/100 g meat)
Fatty acidsc 52 double bonds (g/kg meat)
Fatty acidsc 53 double bonds (g/kg meat)
a-tocopherolc (mg/kg meat)
Total ironc (mg/kg meat)
A B C D
5.7ad 5.7a 5.9b 6.2c
27.1a 32.3a 8.6b 2.9b
108.3a 114.1a 89.9b 74.2c
1.8a 1.7a 2.5b 2.2ab
2.2a 2.1a 2.3a 2.3a
0.3a 0.3a 0.2a 0.3a
3.5ab 3.0b 3.7a 3.1b
5.2a 5.4a 4.8a 5.4a
a b c d
A, control; B, treadmill exercise; C, epinephrine injection; D, epinephrine injection and treadmill exercise. Estimated levels (see Table 1) on the basis of results from Henckel et al. (2000, 2001). Measured average of two determinations. Values in the same column with the same letter are not signi®cantly dierent at a 5% level.
352
D. Juncher et al. / Meat Science 58 (2001) 347±357
Fig. 1. pH measured in samples of M. Longissimus dorsi after stunning as a function of time (min). Curves are ®tted to experimental values according to adapting an exponential decreasing function. Pre-slaughter treatment A: control; B: treadmill exercise; C: epinephrine injection; D: epinephrine injection and treadmill exercise.
D. Juncher et al. / Meat Science 58 (2001) 347±357
in the development in the measured energy metabolites (glycogen, lactate, cp, ATP, IMP) or in pH, the estimated parameters shown in Table 1 together with the measured levels of total iron, a-tocopherol and polyunsaturated fatty acids (Table 2) were included in a covariance analysis, and a summary of the signi®cant results is given in Table 3.
353
For the L*-value the ultimate level of lactate was found to be signi®cant and a higher ultimate lactate level resulted in a higher L*-value during storage (Fig. 3). A signi®cant interaction between pre-slaughter treatment and storage time was also found, indicating that some variation in the L*-value between the four treatment groups can not be explained entirely by the measured physiological parameters. For the redness (a*-value), a signi®cant interaction between pHu and storage time was seen, and it appears from Fig. 4A that a low pHu resulted in signi®cantly higher a*-value, with the dierence being most marked on days 0 and 1. At days 4 and 6 the eect of pHu on the a*-values was less pronounced. From Fig. 4A a pronounced eect of pHu was found at pH values below 5.8. From the ®nal model for the a*-values (see Table 1), the dierence in colour stability (evaluated as the decrease in the a*-value from day 1 to day 6) between chops from M. Longissimus dorsi with pHu=5.6 and chops from M. Longissimus dorsi with pHu=6.0 (same iron content) was estimated to correspond to an average dierence in a* of 1.2 (standard deviation, 0.2). The content of total iron in the meat seems to aect the a*values, too, as a higher content resulted in signi®cantly higher a*-values (Fig. 4B), as would be expected, but the total iron content was not aected by treatment as no signi®cant dierences between treatments were found (Table 2). No signi®cant eect of the pre-slaughter treatment on the a*-value was seen (Table 3), indicating that most of the variation in meat redness can be explained by pHu and total iron content. A signi®cant interaction between pHu and storage time was also found for the b*-values, and it appears from Fig. 5 that a low pHu resulted in higher b*-values, and again the eect was most pronounced at the beginning of storage. As in the case of a*-values, no signi®cant eects of pre-slaughter treatment were seen (Table 3), indicating that pHu was of great importance for the b*-value. 3.3. Lipid oxidation
Fig. 2. Colour L*-, a*- and b*-values measured at the surface of chops from M. Longissimus dorsi (LSMeans of ®ve animals with standard errors of the means from the model mentioned in Section 2 indicated as bars) exposed to ¯uorescent light at 4 C for up to 6 days. Preslaughter treatment: A (-&-): control; B (-&-): treadmill exercise; C (*-): epinephrine injection; D (-*-): epinephrine injection and treadmill exercise.
The eect of the pre-slaughter treatment on lipid oxidation measured as TBARS during the storage period is shown in Fig. 6. In spite of the low TBARS in chops from all four treatments, the TBARS-values were signi®cantly higher (P<0.001) during storage in chops from treatments A and B than in chops from treatments C and D, and TBARS increased signi®cantly (P<0.05) during storage in all groups. A signi®cant dierence between treatments for intramuscular fat content was found, but the dierence in the amount of polyunsaturated fatty acids in the meat was not signi®cant (Table 2). Neither the content of polyunsaturated fatty acids, nor the content of a-tocopherol [in which there was signi®cant dierence between treatments (Table 2)]
354
D. Juncher et al. / Meat Science 58 (2001) 347±357
was found to have signi®cant in¯uence on TBARS. In fact, pHu was the only measured parameter which aected TBARS signi®cantly (Table 3), and as might be seen from Fig. 7, a low pHu resulted in increased lipid oxidation. Fig. 7 shows that there was a threshold value at approximately 5.8 above which the eect of pHu seems to be of minor importance. As well as the signi®cant eect of pHu, a signi®cant eect of pre-slaughter treatment was seen, indicating that the variation in TBARS between the four treatment groups can not be explained entirely by the measured parameters. 3.4. Drip loss Fig. 8 shows the drip loss of the chops from the four treatment groups. Chops from pre-slaughter treatments A and B had signi®cantly higher (P<0.01) drip loss than chops from treatments C and D. The total increase in lactate was the only pre-slaughter parameter signi®cantly aecting drip loss (Table 3), and a higher total increase in lactate resulted in a higher drip loss (Fig. 9). However, a signi®cant eect of the pre-slaughter
treatment was still left, indicating that the variation in drip loss seen between the four groups can not be explained entirely by the measured parameters. 4. Discussion Pre-slaughter conditions for the pigs in the present study have clearly shown their aect on both the initial colour of pork chops from M. Longissimus dorsi muscles and their colour stability, lipid oxidation and drip loss during subsequent chill storage in light. Moreover, the systematic variation in pre-slaughter treatment and a detailed statistical analysis of the results relating to pork quality has allowed certain conclusions to be made relating to the factors of importance for quality. Moreover, pattern of interaction between pigment oxidation (resulting in discolouration) and lipid oxidation became evident, although the reasons for this interaction will have to await further investigations. However, it should be noted that colour changes are occurring more rapidly than lipid oxidation, as seen from a comparison of the
Table 3 Summary of the results of the covariance analyses testing the eect of the energy metabolites and pH on colour parameters, TBARS and drip loss of chops (M. Longissimus dorsi) Quality parameter
Parameter showing signi®cant eect
Colour L*-values a*-values b*-values TBARS Drip loss
alactate***a,b, (treatmentstorage time)*** (apHstorage time)***, total iron*** (apHstorage time)*** apH*, treatment**, storage time* glactate*c, treatment*
a
Fig. 3. Eect of ultimate lactate concentration on the L*-values measured at the surface of chops from M. Longissimus dorsi exposed to ¯uorescent light at 4 C. Days of storage: 0: (&), 1: (~), 4: (), 6: ( ).
a means ultimate level (see Table 1). *, **, *** means signi®cant at P<0.05, P<0.01 and P<0.001, respectively. c g means total change (see Table 1). b
Fig. 4. Eect of ultimate pH (A) and iron content (B) on the a*-values measured at the surface of chops from M. Longissimus dorsi exposed to ¯uorescent light at 4 C. Days of storage: 0: (&), 1: (~), 4: (), 6: ( ).
D. Juncher et al. / Meat Science 58 (2001) 347±357
time development of a* (Fig. 2) compared with TBARS values (Fig. 6). This gives an indication of the coupling between pigment oxidation and lipid oxidation through a mechanism where the catalytic species (i.e. free radicals) involved in the initiation of lipid oxidation are generated through oxymyoglobin oxidation, as recently discussed (Skibsted et al., 1998). Since autoxidation of oxymyoglobin is speci®cally acid-catalyzed, i.e. a decrease in pH of one unit accelerates autoxidation by a factor of 10, the lower pH of meat from treatments A and B readily explains the decreased colour stability and increased lipid oxidation of chops from animals in these treatments. As for colour stability, chops from treatments A and B, with signi®cantly lower pHu and higher ultimate levels of glycogen and lactate (Table 2) were found to be paler than chops from treatments C and D in agreement with the results of Hatton et al. (1972) on pigs and Guignot et al. (1994) on calves. For paleness, the ultimate
Fig. 5. Eect of ultimate pH on the b*-values measured at the surface of chops from M. Longissimus dorsi exposed to ¯uorescent light at 4 C. Days of storage: 0: (&), 1: (~), 4: (), 6: ( ).
Fig. 6. TBARS measured on the surface (approximately 1 mm thick) of chops from M. Longissimus dorsi (LSMeans of ®ve animals with standard errors of the means from the model mentioned in Section 2 indicated as bars) exposed to ¯uorescent light at 4 C for up to 6 days. Pre-slaughter treatment: A (-&-): control; B (-&-): treadmill exercise; C (-*-): epinephrine injection; D (-*-): epinephrine injection and treadmill exercise.
355
level of lactate was shown to be the most important measured parameter (Table 3), and as shown in Fig. 3, an almost linear relationship exists between the ultimate level of lactate and the initial L*-values of the chops. The ultimate level of lactate is closely related to the initial level of glycogen and accordingly inversely related to the ultimate pH of the muscle. The production of acid during post mortem glycolysis will initiate the denaturation of the meat proteins resulting in a more light re¯ection and a paler appearance. Chops from treatments A and B had a higher ratio of oxymyoglobin/myoglobin at the beginning of the storage period than chops from treatments C and D, and these dierences might also be related to dierences in pHu (Figs. 4A, 5). At high pH, the rate of oxygen consumption at the surface of the meat, where oxygen is available, is relatively high and myoglobin is thereby maintained in the reduced state (Ashmore, Parker, & Doerr, 1972). The oxygenation of myoglobin to oxymyoglobin is also reduced by the poorer oxygen penetration in meat of
Fig. 7. Eect of ultimate pH on TBARS measured at the surface (approximately 1 mm thick) of chops from M. Longissimus dorsi exposed to ¯uorescent light at 4 C. Days of storage: 0: (&), 1: (~), 4: (), 6: ( ).
Fig. 8. Drip loss from chops from M. Longissimus dorsi (average of three animals with standard deviations indicated as bars) exposed to ¯uorescent light at 4 C for 6 days. Pre-slaughter treatment: A: control; B: treadmill exercise; C: epinephrine injection; D: epinephrine injection and treadmill exercise.
356
D. Juncher et al. / Meat Science 58 (2001) 347±357
5. Conclusion
Fig. 9. Eect of total increase in lactate concentration on drip loss measured on chops from M. Longissimus dorsi exposed to ¯uorescent light at 4 C for 6 days.
high pH (Gregory, 1998). At low pH, inactivation of mitochondria inhibits oxygen consumption (Ashmore et al., 1972) and oxygen is therefore more available for oxygenation of myoglobin, resulting in a deeper layer of oxymyoglobin and a redder (higher a*-values) and less blue (higher b*-values) colour at the surface. Similar relationships between oxymyoglobin/myoglobin and a*and b*-values were found by Hunt et al. (1999). The colour stability is an important quality parameter (Faustman & Cassens, 1990), and the higher rate of discolouration in chops of low pHu (A and B) can be explained by speci®c acid catalysis of oxidation of the bright-red oxymyoglobin to the brownish-red metmyoglobin (Renerre, Anton, & Gatellier, 1992). A similar eect of low pHu on discolouration was described by Owen and Lawrie (1975). In agreement with other studies (Owen & Lawrie, 1975; Yasosky et al., 1984), the ultimate pH was found to be a very important factor for lipid oxidation, especially at low pHu. The pronounced eect of pHu below 5.8 (Fig. 7) is the same threshold value as for improved red colour (Fig. 4A). Though the TBARS values in general were low in the fresh meat, it might result in severe quality problems, if meat of pHu below 5.8 is further processed into meat product or is used for the production of precooked meat, since lipid oxidation is more pronounced in such products (Nielsen, Sùrensen, Skibsted, & Bertelsen, 1997). In an experiment with ground pork, Yasosky et al. (1984) found a critical pH value of 6.10, or even higher was necessary to obtain maximum inhibition of lipid oxidation. The drip loss was found to be highest in the chops with signi®cantly lower pHu (treatments A and B), which is in agreement with results obtained by Hatton et al. (1972). The total increase of lactate, which was closely correlated to the ultimate level of lactate, was the most important determinant for drip loss, an observation which can be related to protein denaturation, as described above for meat lightness.
Pork chops from meat with lower ultimate pH (treatments A and B) were found to have signi®cantly higher initial L*-values, resulting in an overall paler appearance, and signi®cantly higher a*- and b*-values, indicating a relatively higher proportion of the myoglobin in the oxygenated form on the surface of the chops compared with chops from meat with higher pH (treatments C and D). The colour stability (decrease in a*value) during chill storage was signi®cantly poorer, and lipid oxidation and drip loss were signi®cantly higher for meat of lower pH. While pork chops from meat of lower ultimate pH had an intial high redness, discolouration was faster, and the chops were only redder than chops from meat of higher ultimate pH at the beginning of storage. A detailed statistical analyses, indicated, that the most important parameters to explain the dierences in colour, lipid oxidation and drip loss were pHu and ultimate lactate concentration in the meat. As pH and lactate were highly correlated, these results emphazise the importance of obtaining an ultimate pH in the meat in the narrow range where both microbiological aspects as well as meat quality parameters such as colour, lipid oxidation and drip loss can be controlled. Measurements of pH should accordingly be considered important in any quality assurance system for pork. Acknowledgements This work was part of a collaboration project between LMC-Centre for Advanced Food Studies and The Federation of Danish Pig Producers and Slaughterhouses and sponsored by the FéTEK programme through the Danish Ministry of Food, Agriculture and Fisheries. We wish to thank Jùrgen Steen Jensen and Bente Sùrensen for their excellent technical assistance. References Andersen, H. J., Bertelsen, G., & Skibsted, L. H. (1988). Salt eect on acid-catalyzed autoxidation of oxymyoglobin. Acta Chemica Scandinavia, A42, 226±236. Ashmore, C. R., Parker, W., & Doerr, L. (1972). Respiration of mitochondria isolated from dark-cutting beef: postmortem changes. Journal of Animal Science, 34, 46±48. Barton-Gade, P. A., Demeyer, D., Honikel, K. O., Joseph, R. L., Poulanne, E., Severini, M., Smulders, F. J. M., & Tornberg, E. (1993). Reference methods for water holding capacity in meat and meat products: procedures recommended by an OECD working group. 39th International Congress of Meat Science and Technology, August 1±6, 1993, Calgary Alberta, Canada. Bendall, J. R. (1973). Post mortem changes in muscles. In G. H. Bourne, Structure and function of muscle (pp. 243±309). New York: Academic Press. Briskey, E. J. (1964). Etiological status and associated studies of pale,
D. Juncher et al. / Meat Science 58 (2001) 347±357 soft, exudative porcine musculature. In C. O. Chichester, E. M. Mrak, & G. F. Stewart, Advances in food research (vol. 13) (pp. 89±178). New York: Academic Press. Buttriss, J. L., & Diplock, A. T. (1984). High-performance liquid chromatography methods for vitamin E in tissues. Methods in Enzymology, 5, 131±138. Danish Standardization Board (Dansk StandardiseringsraÊd). (1982a). Water quality Ð determination of metal in water, sludge and sediment by atomic absorption spectrophotometry. Special guidelines for cadmium, cobalt, iron, nickel, lead and zink (in Danish). Dansk Standard, DS 263, (1st ed.). January 1982. Danish Standardization Board (Dansk StandardiseringsraÊd). (1982b). Determination of metals in water, sludge and sediments Ð general guidelines for determination by atomic absorption spectrophotometry in ¯ame (in Danish). Dansk Standard, DS 259, (1st ed.). January 1982. Faustmann, C., & Cassens, R. G. (1990). The biochemical basis for discoloration in fresh meat: a review. Journal of Muscle Foods, 1, 217±243. Gilman, L. B. (1989). Microwave sample preparation. CEM Corporation, Matthews, NC28106. Gregory, N. G. (1998). Animal welfare and meat science. Cambridge, UK: University Press (Chapter 6).. Guignot, F., Touraille, C., Ouali, A., Renerre, M., & Monin, G. (1994). Relationships between post-mortem pH changes and some traits of sensory quality in veal. Meat Science, 37, 315±325. Hatton, M. W. C., Lawrie, R. A., Ratcli, P. W., & Wayne, N. (1972). Eects of preslaughter adrenalin injection on muscle metabolites and meat quality of pigs. Journal of Food Technology, 7, 443±453. Henckel, P., Karlsson, A., Jensen, M. T., Oksbjerg, N., Petersen, J. S. (2001). Energy levels in pig muscles and their development post mortem after standardised treatments prior to slaughter. Submitted to Meat Science. Henckel, P., Karlsson, A., Oksbjerg, N., & Petersen, J. S. (2000). Control of post mortem pH-decrease in pig muscles: experimental design and testing of animal models. Meat Science, 55(1), 131±138. Henckel, P., Karlsson, A., Petersen, J. S. (1997). Exercise, fasting, and epinephrine administration Ð eects on muscle glycogen degradation in vivo in pig M. Longissimus dorsi. Proceeding 43rd International Congress of Meat Science and Technology, (pp. 294±295). New Zealand.
357
Hunt, M. C., Sùrheim, O., & Slinde, E. (1999). Color and heat denaturation of myoglobin forms in ground beef. Journal of Food Science, 64(5), 847±851. Jart, A. (1997). The magnetic ®eld as an additional selectivity parameter in fat hydrogenation. Journal American Oil Chemist Society, 74, 615±617. Jensen, C., Guidera, J., Skovgaard, I. M., Staun, H., Skibsted, L. H., Jensen, S. K., Mùller, A. J., Buckly, J., & Bertelsen, G. (1997). Eects of dietary a-tocopheryl acetate supplementation on a-tocopherol deposition in porcine m. psoas major and m. longissimus dorsi and on drip loss, colour stability and oxidative stability of pork meat. Meat Science, 45, 491±500. Nielsen, J. H., Sùrensen, B., Skibsted, L. H., & Bertelsen, G. (1997). Oxidation in pre-cooked minced pork as in¯uenced by chill storage of raw muscle. Meat Science, 46, 191±197. Owen, J. E., & Lawrie, R. A. (1975). The eect of an arti®cially induced high pH on the susceptibility of minced porcine muscle to undergo oxidative rancidity under frozen storage. Journal of Food Technology, 10, 169±180. Renerre, M. (1990). Review: factors involved in the discoloration of beef meat. International Journal of Food Science and Technology, 25, 613±630. Renerre, M., Anton, M., & Gatellier, P. (1992). Autoxidation of puri®ed myoglobin from two bovine muscles. Meat Science, 32, 331±342. SAS (1996). SAS/STATTM software (ver. 6.12). Cary, NC, USA: SAS Institute Inc. Skibsted, L. H., Mikkelsen, A., & Bertelsen, G. (1998). Lipid-derived o-¯avours in meat. In F. Shahidi, Flavour of meat and meat products (2nd ed.) (pp. 217±256). London: Blackie Academic & Professional. Vyncke, W. (1975). Evaluation of the direct thiobarbituric acid extraction method for determining oxidative rancidity in mackerel (Scomber scombrus L.). Fette-Seifen-Anstrichmittel, 77, 239±240. Warriss, P. D., Bevis, E. A., & Ekins, P. J. (1989). The relationships between glycogen stores and muscle ultimate pH in commercially slaughtered pigs. British Veterinary Journal, 145, 378±383. Yasosky, J. J., Aberle, E. D., Peng, I. C., Mills, E. W., & Judge, M. D. (1984). Eects of pH and time of grinding on lipid oxidation of fresh ground pork. Journal of Food Science, 49, 1510±1512.
www.elsevier.com/locate/meatsci
Eect of pre-slaughter physiological conditions on the oxidative stability of colour and lipid during chill storage of pork Dorte Juncher a,*, Birgitte Rùnn b, Else T. Mortensen a, Poul Henckel c, Anders Karlsson c, Leif H. Skibsted a, Grete Bertelsen a a
Department of Dairy and Food Science, Royal Veterinary and Agricultural University, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark b Department of Mathematics and Physics, Royal Veterinary and Agricultural University, Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark c Department of Product Quality, Danish Institute of Agricultural Sciences, Research Centre Foulum, Box 50, DK-8830 Tjele, Denmark Received 12 June 2000; received in revised form 5 December 2000; accepted 12 December 2000
Abstract The physiological condition of the live animal was found to signi®cantly aect colour, lipid oxidation and water holding capacity of chill stored pork chops (M. Longissimus dorsi) in a study, where various pre-slaughter conditions were achieved by the following four treatments: (A) control; (B) subjected to treadmill exercise immediately prior to stunning; (C) given epinephrine injection 15 h prior to slaughter; and (D) given epinephrine injection 15 h before slaughter and further subjected to treadmill exercise immediately before stunning. The treatments resulted in variations in energy metabolites (glycogen, lactate, creatine phosphate, ATP) and ultimate pH (pHu), with the lowest pHu in chops from treatments A and B, and in signi®cantly dierent tristimulus colour L*-, a*- and b*-parameters, although the eect of treatment on colour was not consistent during the chill storage period of 6 days. Overall, chops from treatments A and B had signi®cantly higher L*- and b*-values (were paler and less blue) than chops from C and D during storage under conditions typical for retail trade. The initial a*-values were higher (redder) in chops from treatments A and B, but the colour, as judged by the a*-values, was less stable in meat from these treatments compared with treatments C and D. Lipid oxidation, evaluated by thiobarbituric acid reactive substances (TBARS) in the fresh meat, and drip loss, measured after 6 days of storage, were both signi®cantly higher in chops from treatments A and B compared to chops obtained from treatments C and D. Statistical analysis relating the pH and the level of various energy metabolites post-mortem in the individual animals to the measured quality parameters, revealed that pHu was the most important factor aecting product quality. In conclusion, over all product quality depends on obtaining a pHu in the narrow range where both meat quality parameters such as colour, lipid oxidation and drip loss as well as microbiological aspects have to be considered. # 2001 Elsevier Science Ltd. All rights reserved. Keywords: Colour stability; Drip loss; Glycogen; Lipid oxidation; pH; Pre-slaughter treatment
1. Introduction For the consumer the surface colour of meat is the most important quality attribute at the time of purchase. During the last decades concern has risen in the meat trade about the colour of pork, as pork becomes paler. The surface colour of meat depends on the quantity of myoglobin present, on its chemical state and also on the chemical and physical conditions of other * Corresponding author. Tel.: +45-35-28-32-35; fax: +45-35-2833-44. E-mail address: [email protected] (D. Juncher).
components in the meat (Renerre, 1990). Factors aecting the surface colour are related to dierences between breeds and even individual animals, the age of the animal at time of slaughter, pre-slaughter handling, the chilling process, methods of packaging, retail light exposure and time/temperature regime during storage (Renerre, 1990). The rate of post mortem pH decline and the ultimate pH (pHu) have been shown to be important in determining the paleness as well as for the water holding capacity of pork meat. Thus, a high rate of pH decline and a low pHu result in pale meat of low water-holding capacity (Briskey, 1964). In addition, a low pHu may
0309-1740/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved. PII: S0309-1740(00)00156-X
348
D. Juncher et al. / Meat Science 58 (2001) 347±357
result in more pronounced discolouration during chill storage as the rate of oxidation of oxymyoglobin to metmyoglobin is accelerated at low pH (Andersen, Bertelsen, & Skibsted, 1988). Discolouration and lipid oxidation are often coupled, and a high degree of discolouration may indicate decreased oxidative stability with subsequent development of o-¯avours during storage, especially after further processing involving addition of salt or cooking (Skibsted, Mikkelsen, & Bertelsen, 1998). The characteristics of the post mortem pH decline are determined by the physiological condition of the animal at stunning (Bendall, 1973; Warriss, Bevis, & Ekins, 1989). Although the physiological conditions before slaughter may be manipulated experimentally by epinephrine injection, previous experiments using this method (Guignot, Touraille, Ouali, Renerre, & Monin, 1994; Hatton, Lawrie, Ratcli, & Wayne, 1972; Owen & Lawrie, 1975; Yasosky, Aberle, Peng, Mills, & Judge, 1984) did not show any relationship between the physiological conditions of the muscles at the time of stunning and the meat quality during chill storage. The goal of the present study was to investigate the eect of pre-slaughter physiological conditions on colour, water holding capacity and lipid oxidation of chops from M. Longissimus dorsi during retail chill storage. The physiological conditions were manipulated by treadmill exercise and/or epinephrine injection in pigs without the gene for halothane sensitivity. Based on the results obtained in a series of experiments described by Henckel, Karlsson, and Petersen (1997), four sets of conditions were chosen to cover the wide range of variation in level of energy metabolites and pH post-mortem that may be observed in pigs under practical conditions. 2. Materials and methods 2.1. Muscle samples Twenty slaughter pigs (all without the gene for halothane sensitivity), as a part of the herd of 80 pigs raised at the Research Centre Foulum and used in the experiment described by Henckel, Karlsson, Oksbjerg, and Petersen (2000), were used in the present study. The pigs, from ®ve dierent litters, were fed with ®ve consecutive batches of pig feed. The composition of the pig feed batches and the content of vitamin E, dry matter, fat and peroxide values of the feed lipid are given by Henckel et al. (2000). From each litter four females were selected. Each pig was exposed to one of the four treatments: (A) control, no treatment; (B) exercise on a treadmill at a rate of 3.8 km/h for 10 min immediately prior to stunning; (C) 0.2 mg epinephrine/kg live weight 15 h pre slaughter; and
(D) 0.3 mg epinephrine/kg live weight 15 h pre slaughter and exercise on a treadmill at a rate of 3.8 km/h for 5 min immediately prior to stunning. Treatments A and B were given 1 ml of a sterile salt solution as a placebo at the same time as treatments C and D were administered epinephrine (Henckel et al., 2000). The four pigs from each litter (one pig per treatment) were slaughtered on the same day as described by Henckel et al. (2000). The levels of glycogen, creatine phosphate, lactate, ATP and IMP were measured in biopsies from M. Longissimus dorsi immediately before stunning and 1, 15, 30, 45 min, and 1 , 3, 6 and 24 h after exsanguination as described by Henckel, Karlsson, Jensen, Oksbjerg, and Petersen (2001). pH was measured at 1, 15, 30, 45 min, and 1, 3, 6 and 24 h after exsanguination as described by Henckel et al. (2000). Twenty-four hours post mortem, the carcasses were dissected and the right side M. Longissimus dorsi from each of the pigs from treatments A, B, C and D was placed in individual polyethylene bags and transported in a thermobox (1±3 C) from the Research Center Foulum to the Royal Veterinary and Agricultural University. 2.2. Handling and storage conditions Two days post mortem, M. Longissimus dorsi muscles were sliced into 2-cm chops, which were placed in individual plastic trays, wrapped in polyethylene and placed in an illuminated (¯uorescent tubes, Philips 18W/82, average illuminance of 900 lx on the surface of the products) chill cabinet for up to 6 days. The temperature of the chill cabinet was 4 C with an increase in temperature to approximately 8 C during daily defrosting, as monitored by a Grant datalogger. The chops were randomized for analyses to be performed after 0, 1, 4 and 6 days of chill storage; one chop on each day for the colour and chemical analyses and one for drip loss determination at day 0. At each analysis, the colour of the surface of each day of chop was measured, followed by cutting the chop into small pieces which were randomized for the analyses described later. When not otherwise noted, the analyses were made in duplicate. Samples for analyses of total iron and a-tocopherol were frozen in liquid nitrogen and stored at 80 C until analysis. 2.3. Analyses All chemicals used were of analytical grade, and water was puri®ed through a Millipore puri®cation train (Millipore Corp., Bedford, MA). 2.3.1. Colour 2.3.1.1. Colour measurement. The colour measurements were performed using a tristimulus colorimeter (Minolta Chroma Meter CR300, Minolta Corporation, NJ). The
D. Juncher et al. / Meat Science 58 (2001) 347±357
L*-, a*- and b*-values were measured 10 times on the surface of the chops at day 0 (after a blooming period of 2 h in dark at 4 C), and at day 1, 4 and 6 after slicing. 2.3.2. Lipid oxidation 2.3.2.1. TBARS. Lipid oxidation was measured as 2thiobarbituric acid reactive substances (TBARS) by the extraction method described by Vyncke (1975) with a few modi®cations: meat (10 g) from the surface (1 mm layer) of the chop was homogenized (Ultra Turrax T-25, Janke & Kunkel IKA-Labortechnik, Staufen, Germany) with 30 ml of a 7.5% trichloroacetic acid (TCA) solution including 0.1% propylgallate (PG) and 0.1% ethylenediaminetetraacetic acid, disodium salt (EDTA) for 45 s at 13,500 rpm and the homogenate was ®ltered through a Whatmann ®lter no. 42. The extract (5 ml) was mixed with 0.020 M thiobarbituric acid (5 ml), heated and cooled as described by Vyncke (1975). The absorbance was measured at 532 and 600 nm using a HP 8452A diode assay spectrophotometer (Hewlett Packard Co., Palo Alto, CA), and the absorbance difference, A532nm A600nm, was calculated with A600nm correcting for sample turbidity. TBARS, expressed as milligrams of malonaldehyd per kilogram of meat, was calculated using malondialdehyd-bis-(diethylacetate) as standard. 2.3.2.2. Dry matter. Dry matter determinations, for use for calculation of TBARS, were conducted by drying 2 g meat from the surface of the chop in an oven at 104 C for 4 h and weighing after cooling for 15 min. 2.3.2.3. Extraction of lipids. Meat (10 g) was homogenized by an Ultra Turrax (Ultra Turrax T-25, Janke & Kunkel IKA-Labortechnik, Staufen, Germany) with 100 ml chloroform/methanol (2:1 v/v) for 1 min at 13,500 rpm. After homogenization, 25 ml 1.0 mM CaCl2 solution was added and the sample was further homogenized using an Ultra Turrax and centrifuged (MSE Mistral 2000, Crawley, England) for 20 min at 1000 rpm. The chloroform phase was removed and the extraction procedure repeated. The chloroform phase containing the extracted lipids was dried by vacuum evaporation (BuÈchi RE 11, BuÈchi Laboratoriums-Technik AG, Flawill, Schweiz). Finally 22 ml chloroform/ methanol and 2.0 ml CaCl2 were added to the dried sample and the sample was mixed (Vortex-mixer VF2, Janke & Kunkel IKA-Labortechnik) and centrifuged for 20 min at 2500 rpm. The lipid phase was removed, dried by vacuum evaporation and weighed. The percent of intramuscular fat was calculated from the weight of total lipid obtained after solvent extraction and the weight of the meat. 2.3.2.4. Fatty acid composition of the total fat. Extracted lipid (10 mg) was transformed into its constituent
349
methyl esters as described by Jart (1997). Methyl esters (1 ml) was injected into a HP 5890 series II gas chromatograph (Hewlett-Packard) with a ¯ame-ionization detector and with a 25 m0.20 mm0.33 mm HP-FFAP column No. 19091F-102 (Hewlett-Packard) under the following oven temperature program and conditions: 50 C for 1 min; 50±180 C at 15 C/min; 180±220 C at 5 C/min; at 220 C for 10 min; injector temperature was 250 C, detector temperature 300 C. Helium was used as the carrier gas with a split ratio of 1:10 and a ¯ow was 1 ml/min. The relative contents of the fatty acids C4:0, C14:0, C16:0, C18:0, C18:1, C18:2, C18:3, C20:0, C20:4, C22:1 and C24:0 were calculated from the chromatograms and from an external standard containing methyl esters of the fatty acids. 2.3.3. -Tocopherol a-Tocopherol was extracted using the procedure of Buttris and Diplock (1984) with some modi®cations: 2 ml homogenate (4 g meat homogenized with 20 ml 1.15% KCl-solution using an Ultra Turrax blender for 30 s at 13,500 rpm) was added to 0.20 ml saturated potassium hydroxide and 2 ml 0.50% ethanolic pyrrogalol. The sample was Vortex-mixed for 10 s and saponi®ed at 70 C for 30 min. After cooling in ice water, and addition of 1.0 ml water and 4.0 ml hexane with 0.001% BHT, the sample was vigorously shaken, mixed on a Vortex mixer followed by centrifugation (Universal 16A, Hettich Zentrifugen, D-78532 Tuttlingen, Germany) for 5 min at 800g. The upper hexane layer was collected and the residue re-extracted with a further 3 ml of hexane containing 0.001% BHT. The combined fractions were evaporated under nitrogen and a-tocopherol was redissolved in 0.50 ml ethanol with 0.001% BHT. Quanti®cation of tocopherols was performed according to Jensen et al. (1997). 2.3.4. Total iron Total iron was measured according to the methods of the Danish Standardization Board (1982a, b) and of Gilman (1989) with some modi®cations: Meat (2 g) and concentrated nitric acid (7 ml) were kept overnight in a te¯on container. The sample was destroyed in a closed te¯on container in a microwave oven (CEM MDS 81D, CEM Corporation, Matthews NC) with the following program: 100% power for 5 min to a pressure of about 120 Psi followed by cooling and ventilating of the container, 75% power for 10 min to a pressure of about 120 Psi followed by cooling and ventilating of the container, 100% power for 15 min to a pressure of < 120 Psi followed by cooling and ventilating. Five droplets of hydrogen peroxide (30%) were added and the sample was kept overnight in the container. The ashed sample was transfered to a 10 ml ¯ask with water and the iron content in the resulting solution measured by atomic absorptions spectroscopy (Perkin Elmer 3300, The
350
D. Juncher et al. / Meat Science 58 (2001) 347±357
Perkin-Elmer Corporation, Norwalk, CT). The concentration of total iron in the sample was calculated from a standard curve. 2.3.5. Drip loss Drip loss was measured on one chop from each muscle by the method of Barton-Gade et al. (1993) with some modi®cations: The chop was weighed immediately after cutting, hung in a laminate bag, closed tightly with string. The chop was weighed again after 6 days at 4 C, and the drip loss calculated. The drip loss was measured on the last three litters of the storage experiment. 2.4. Statistical analyses The pre-slaughter physiological conditions were expressed by the following physiological variables measured immediately before stunning and after exsanguination, namely pH, glycogen, lactate, creatine phosphate, ATP and IMP taken 4 min before stunning and 1, 15, 30, 45 min, and 1, 3, 6 and 24 h after exsanguination (pH only after exsanguination). For each variable, the nine measurements over time were concentrated into three parameters ( , and ) for each pig by the non-linear normal models given in Table 1. As shown in Table 1 the measurements of pH and glycogen were modelled as an exponential function decreasing in time, the variables lactate and IMP were modelled by a S-shaped function of time, and the variables creatine phosphate and ATP were modelled as very steep exponential functions decreasing in time. The non-linear models were ®tted by PROC NLIN in SAS (1996). To investigate whether the treatments resulted in different physiological variables for the pigs, all the estimated parameters were subjected to analysis of variance with treatment and litter included as ®xed eects. The development in colour, measured as L*-, a*- and b*-values, and in lipid oxidation, measured as TBARS, over time were analyzed by analysis of variance, where treatment, storage time and the interaction between
treatment and storage time were included as ®xed eects. Individual pig, litter and the interaction between litter and storage time were included as random eects. Within this model contrasts between treatment at day 0 were estimated and tested by pairwise comparison via ttests. The treatment eect on colour stability, evaluated as the decrease in the a*-value from day 1 to day 6, was tested similarly. The response variable TBARS was transformed with the natural logarithm prior to analysis in order to obtain homogeneous variance. The eects of treatment on colour and lipid oxidation found in the above analysis may be caused by dierences in pre-slaughter physiological conditions induced by the four treatments. Hence the physiological variables, expressed by the 18 estimated parameters, together with total iron content, a-tocopherol, polyunsaturated fatty acids (expressed as fatty acid with two or more double bonds and as fatty acids with three or more double bonds) were included, one at a time, in a covariance analysis, where treatment, storage time, interaction between treatment and storage time and interaction between the physiological parameter and storage time were included as ®xed eects. Individual pig, litter and the interaction between litter and storage time were included as random eects. Finally the variables showing signi®cant eects on the colour or TBARS values were included simultaneously in a covariance analysis with treatment, storage time and the interaction between treatment and storage time included as ®xed eects, and individual pig, litter and the interaction between litter and storage time included as random eects. Backwards selection was made to obtain the ®nal model for changes in the colour values and TBARS. Drip loss was measured on the chops from pigs from three of the litters and the observations were analysed by an analysis of variance with treatment included as a ®xed eect and litter included as a random eect. The treatment eect on drip loss was analyzed as described for colour and lipid oxidation, but within the simpler model with treatment included as ®xed eect and
Table 1 Models used to describe development in pH, glycogen, lactate, creatine phosphate, ATP and IMP measured in M. Longissimus dorsi immediately before stunning (not for pH) and after exsanguination for up to 24 h Parameter
Model for parameter ®ttinga
Parameters signi®cantly aected by the treatmentsb
pH Glycogen Lactate Creatine phosphate (cp) ATP IMP
pH=a(1+exp( b(time- g))) glycogen=a(1+exp( b(time g))) lactate=a exp(g btime exp( time)) cp=a+g(exp(btime) 1) ATP=a+g(exp(btime) 1) IMP=a/(1+1/g exp( 1/btime))
a*** a*** a***, g** a**, b** a**, g**
a a is a common parameter for the ultimate level of pH, glycogen, lactate and IMP and for the level at t=0 for cp and ATP, b is a common parameter for the rate of decrease in pH, glycogen, cp and ATP, and for the increase in lactate or IMP, and g is a common parameter that re¯ects the time of the decrease in pH and glycogen and the total change in level of lactate, IMP, cp and ATP. b *) **) ***) Parameters are signi®cantly aected by the treatment A: control, B: treadmill exercise, C: epinephrine injection, D: epinephrine injection and treadmill exercise (*:P<0.05, **:P<0.01 and ***:P<0.001).
D. Juncher et al. / Meat Science 58 (2001) 347±357
individual pig as random eect. Backwards selection was made to obtain the ®nal model for drip loss. All analyses of variance and covariance were ®tted by PROC MIXED in SAS (1996). Signi®cant treatment eects for intramuscular fat (IMF), polyunsaturated fatty acids, a-tocopherol and total iron were classi®ed by LSD (SAS, 1996). 3. Results 3.1. Pre-slaughter physiological conditions For each of the 20 pigs the changes in pH, glycogen, lactate, creatine phosphate (cp), ATP and IMP were measured before and up to 24 h after stunning (pH only after stunning) as reported by Henckel et al. (2000, 2001). The values for these six physiological parameters were ®tted to a non-linear model as described in Section 2, and in Fig. 1 the development in pH with time is shown as an example. It appears that the models ®t well to the observed values in each of the individual 20 pigs. Even though some variability in pH development is seen between the litters, a clear treatment eect is seen (Fig. 1). For the other physiological parameters plots of the function described in Section 2 gave plots of comparable ®tness. For each of the physiological variables, the information on the development over time was concentrated into three parameters , and , as described in Section 2 (Table 1) together with the results of the analyses of variance made to investigate whether the treatments result in dierences in the development of energy metabolites (glycogen, lactate, cp, ATP and IMP) and in the development of pH. The most pronounced eect of preslaughter treatment (A, B, C and D) was found in the ultimate level ( ) of pH, glycogen and lactate (Table 1). Thus, pigs from treatments A and B had signi®cantly (P<0.001) lower pHu and signi®cantly (P<0.001) higher ultimate levels of glycogen and lactate compared with pigs from the treatments C and D (Table 2). It also
351
appears from Table 1, that the treatment did not aect the rate of decrease or rate of increase of any of the measured energy metabolites or of pH, except for a weaker eect on the rate of decrease of cp. 3.2. Colour and colour stability The eect of pre-slaughter treatment on initial colour and on colour stability is shown in Fig. 2. The analysis of variance showed a signi®cant eect of the treatment, but the eect of the treatment was not the same during the entire storage period (signi®cant interaction between treatment and storage time (P<0.001 for L*- and a*values and P<0.01 for b*-values)). At the beginning of the storage period (day 0) and during storage, chops from treatments A and B had signi®cantly (P<0.001 and P<0.05 for day 0 and during storage, respectively) higher L*-values (more white) and b*-values (less blue) than chops from treatments C and D, and L*-values of chops from treatment C were signi®cantly (P<0.05) higher than L*-values of chops from treatment D (Fig. 2). Chops from treatments A and B had signi®cantly (P<0.001) higher initial a*-values than chops from treatments C and D, indicating a redder colour in the ®rst two. However, the colour stability during chill storage, evaluated from the decrease in a*-values from day 1 to day 6, was signi®cantly (P<0.01) inferior in chops from treatments A and B compared with chops from treatments C and D (Fig. 2). The higher initial a*values related to the higher b*-values indicate that chops from treatments A and B had more myoglobin in the oxymyoglobin form than chops from treatments C and D at the beginning of the storage period. This agrees with the results of Hunt, Sùrheim, and Slinde (1999) in a study of colour traits of uncooked fresh ground beef patties, in which the highest a*- and b*values were found in patties containing predominantly oxymyoglobin compared with patties containing predominantly deoxy- or metmyoglobin. In order to investigate whether the eects of treatment on the L*-, a*- and b*-values may be caused by dierences
Table 2 Estimated ultimate levels of pH, glycogen and lactate in M. Longissimus dorsi and intramuscular fat content (IMF), fatty acid composition, atocopherol content and total iron content measured at 48 h post mortem in chops (M. Longissimus dorsi) Pre-slaughter treatmenta
pHub
Glycogenub (mmole/kg meat)
Lactateub (mmole/kg meat)
IMFc (g/100 g meat)
Fatty acidsc 52 double bonds (g/kg meat)
Fatty acidsc 53 double bonds (g/kg meat)
a-tocopherolc (mg/kg meat)
Total ironc (mg/kg meat)
A B C D
5.7ad 5.7a 5.9b 6.2c
27.1a 32.3a 8.6b 2.9b
108.3a 114.1a 89.9b 74.2c
1.8a 1.7a 2.5b 2.2ab
2.2a 2.1a 2.3a 2.3a
0.3a 0.3a 0.2a 0.3a
3.5ab 3.0b 3.7a 3.1b
5.2a 5.4a 4.8a 5.4a
a b c d
A, control; B, treadmill exercise; C, epinephrine injection; D, epinephrine injection and treadmill exercise. Estimated levels (see Table 1) on the basis of results from Henckel et al. (2000, 2001). Measured average of two determinations. Values in the same column with the same letter are not signi®cantly dierent at a 5% level.
352
D. Juncher et al. / Meat Science 58 (2001) 347±357
Fig. 1. pH measured in samples of M. Longissimus dorsi after stunning as a function of time (min). Curves are ®tted to experimental values according to adapting an exponential decreasing function. Pre-slaughter treatment A: control; B: treadmill exercise; C: epinephrine injection; D: epinephrine injection and treadmill exercise.
D. Juncher et al. / Meat Science 58 (2001) 347±357
in the development in the measured energy metabolites (glycogen, lactate, cp, ATP, IMP) or in pH, the estimated parameters shown in Table 1 together with the measured levels of total iron, a-tocopherol and polyunsaturated fatty acids (Table 2) were included in a covariance analysis, and a summary of the signi®cant results is given in Table 3.
353
For the L*-value the ultimate level of lactate was found to be signi®cant and a higher ultimate lactate level resulted in a higher L*-value during storage (Fig. 3). A signi®cant interaction between pre-slaughter treatment and storage time was also found, indicating that some variation in the L*-value between the four treatment groups can not be explained entirely by the measured physiological parameters. For the redness (a*-value), a signi®cant interaction between pHu and storage time was seen, and it appears from Fig. 4A that a low pHu resulted in signi®cantly higher a*-value, with the dierence being most marked on days 0 and 1. At days 4 and 6 the eect of pHu on the a*-values was less pronounced. From Fig. 4A a pronounced eect of pHu was found at pH values below 5.8. From the ®nal model for the a*-values (see Table 1), the dierence in colour stability (evaluated as the decrease in the a*-value from day 1 to day 6) between chops from M. Longissimus dorsi with pHu=5.6 and chops from M. Longissimus dorsi with pHu=6.0 (same iron content) was estimated to correspond to an average dierence in a* of 1.2 (standard deviation, 0.2). The content of total iron in the meat seems to aect the a*values, too, as a higher content resulted in signi®cantly higher a*-values (Fig. 4B), as would be expected, but the total iron content was not aected by treatment as no signi®cant dierences between treatments were found (Table 2). No signi®cant eect of the pre-slaughter treatment on the a*-value was seen (Table 3), indicating that most of the variation in meat redness can be explained by pHu and total iron content. A signi®cant interaction between pHu and storage time was also found for the b*-values, and it appears from Fig. 5 that a low pHu resulted in higher b*-values, and again the eect was most pronounced at the beginning of storage. As in the case of a*-values, no signi®cant eects of pre-slaughter treatment were seen (Table 3), indicating that pHu was of great importance for the b*-value. 3.3. Lipid oxidation
Fig. 2. Colour L*-, a*- and b*-values measured at the surface of chops from M. Longissimus dorsi (LSMeans of ®ve animals with standard errors of the means from the model mentioned in Section 2 indicated as bars) exposed to ¯uorescent light at 4 C for up to 6 days. Preslaughter treatment: A (-&-): control; B (-&-): treadmill exercise; C (*-): epinephrine injection; D (-*-): epinephrine injection and treadmill exercise.
The eect of the pre-slaughter treatment on lipid oxidation measured as TBARS during the storage period is shown in Fig. 6. In spite of the low TBARS in chops from all four treatments, the TBARS-values were signi®cantly higher (P<0.001) during storage in chops from treatments A and B than in chops from treatments C and D, and TBARS increased signi®cantly (P<0.05) during storage in all groups. A signi®cant dierence between treatments for intramuscular fat content was found, but the dierence in the amount of polyunsaturated fatty acids in the meat was not signi®cant (Table 2). Neither the content of polyunsaturated fatty acids, nor the content of a-tocopherol [in which there was signi®cant dierence between treatments (Table 2)]
354
D. Juncher et al. / Meat Science 58 (2001) 347±357
was found to have signi®cant in¯uence on TBARS. In fact, pHu was the only measured parameter which aected TBARS signi®cantly (Table 3), and as might be seen from Fig. 7, a low pHu resulted in increased lipid oxidation. Fig. 7 shows that there was a threshold value at approximately 5.8 above which the eect of pHu seems to be of minor importance. As well as the signi®cant eect of pHu, a signi®cant eect of pre-slaughter treatment was seen, indicating that the variation in TBARS between the four treatment groups can not be explained entirely by the measured parameters. 3.4. Drip loss Fig. 8 shows the drip loss of the chops from the four treatment groups. Chops from pre-slaughter treatments A and B had signi®cantly higher (P<0.01) drip loss than chops from treatments C and D. The total increase in lactate was the only pre-slaughter parameter signi®cantly aecting drip loss (Table 3), and a higher total increase in lactate resulted in a higher drip loss (Fig. 9). However, a signi®cant eect of the pre-slaughter
treatment was still left, indicating that the variation in drip loss seen between the four groups can not be explained entirely by the measured parameters. 4. Discussion Pre-slaughter conditions for the pigs in the present study have clearly shown their aect on both the initial colour of pork chops from M. Longissimus dorsi muscles and their colour stability, lipid oxidation and drip loss during subsequent chill storage in light. Moreover, the systematic variation in pre-slaughter treatment and a detailed statistical analysis of the results relating to pork quality has allowed certain conclusions to be made relating to the factors of importance for quality. Moreover, pattern of interaction between pigment oxidation (resulting in discolouration) and lipid oxidation became evident, although the reasons for this interaction will have to await further investigations. However, it should be noted that colour changes are occurring more rapidly than lipid oxidation, as seen from a comparison of the
Table 3 Summary of the results of the covariance analyses testing the eect of the energy metabolites and pH on colour parameters, TBARS and drip loss of chops (M. Longissimus dorsi) Quality parameter
Parameter showing signi®cant eect
Colour L*-values a*-values b*-values TBARS Drip loss
alactate***a,b, (treatmentstorage time)*** (apHstorage time)***, total iron*** (apHstorage time)*** apH*, treatment**, storage time* glactate*c, treatment*
a
Fig. 3. Eect of ultimate lactate concentration on the L*-values measured at the surface of chops from M. Longissimus dorsi exposed to ¯uorescent light at 4 C. Days of storage: 0: (&), 1: (~), 4: (), 6: ( ).
a means ultimate level (see Table 1). *, **, *** means signi®cant at P<0.05, P<0.01 and P<0.001, respectively. c g means total change (see Table 1). b
Fig. 4. Eect of ultimate pH (A) and iron content (B) on the a*-values measured at the surface of chops from M. Longissimus dorsi exposed to ¯uorescent light at 4 C. Days of storage: 0: (&), 1: (~), 4: (), 6: ( ).
D. Juncher et al. / Meat Science 58 (2001) 347±357
time development of a* (Fig. 2) compared with TBARS values (Fig. 6). This gives an indication of the coupling between pigment oxidation and lipid oxidation through a mechanism where the catalytic species (i.e. free radicals) involved in the initiation of lipid oxidation are generated through oxymyoglobin oxidation, as recently discussed (Skibsted et al., 1998). Since autoxidation of oxymyoglobin is speci®cally acid-catalyzed, i.e. a decrease in pH of one unit accelerates autoxidation by a factor of 10, the lower pH of meat from treatments A and B readily explains the decreased colour stability and increased lipid oxidation of chops from animals in these treatments. As for colour stability, chops from treatments A and B, with signi®cantly lower pHu and higher ultimate levels of glycogen and lactate (Table 2) were found to be paler than chops from treatments C and D in agreement with the results of Hatton et al. (1972) on pigs and Guignot et al. (1994) on calves. For paleness, the ultimate
Fig. 5. Eect of ultimate pH on the b*-values measured at the surface of chops from M. Longissimus dorsi exposed to ¯uorescent light at 4 C. Days of storage: 0: (&), 1: (~), 4: (), 6: ( ).
Fig. 6. TBARS measured on the surface (approximately 1 mm thick) of chops from M. Longissimus dorsi (LSMeans of ®ve animals with standard errors of the means from the model mentioned in Section 2 indicated as bars) exposed to ¯uorescent light at 4 C for up to 6 days. Pre-slaughter treatment: A (-&-): control; B (-&-): treadmill exercise; C (-*-): epinephrine injection; D (-*-): epinephrine injection and treadmill exercise.
355
level of lactate was shown to be the most important measured parameter (Table 3), and as shown in Fig. 3, an almost linear relationship exists between the ultimate level of lactate and the initial L*-values of the chops. The ultimate level of lactate is closely related to the initial level of glycogen and accordingly inversely related to the ultimate pH of the muscle. The production of acid during post mortem glycolysis will initiate the denaturation of the meat proteins resulting in a more light re¯ection and a paler appearance. Chops from treatments A and B had a higher ratio of oxymyoglobin/myoglobin at the beginning of the storage period than chops from treatments C and D, and these dierences might also be related to dierences in pHu (Figs. 4A, 5). At high pH, the rate of oxygen consumption at the surface of the meat, where oxygen is available, is relatively high and myoglobin is thereby maintained in the reduced state (Ashmore, Parker, & Doerr, 1972). The oxygenation of myoglobin to oxymyoglobin is also reduced by the poorer oxygen penetration in meat of
Fig. 7. Eect of ultimate pH on TBARS measured at the surface (approximately 1 mm thick) of chops from M. Longissimus dorsi exposed to ¯uorescent light at 4 C. Days of storage: 0: (&), 1: (~), 4: (), 6: ( ).
Fig. 8. Drip loss from chops from M. Longissimus dorsi (average of three animals with standard deviations indicated as bars) exposed to ¯uorescent light at 4 C for 6 days. Pre-slaughter treatment: A: control; B: treadmill exercise; C: epinephrine injection; D: epinephrine injection and treadmill exercise.
356
D. Juncher et al. / Meat Science 58 (2001) 347±357
5. Conclusion
Fig. 9. Eect of total increase in lactate concentration on drip loss measured on chops from M. Longissimus dorsi exposed to ¯uorescent light at 4 C for 6 days.
high pH (Gregory, 1998). At low pH, inactivation of mitochondria inhibits oxygen consumption (Ashmore et al., 1972) and oxygen is therefore more available for oxygenation of myoglobin, resulting in a deeper layer of oxymyoglobin and a redder (higher a*-values) and less blue (higher b*-values) colour at the surface. Similar relationships between oxymyoglobin/myoglobin and a*and b*-values were found by Hunt et al. (1999). The colour stability is an important quality parameter (Faustman & Cassens, 1990), and the higher rate of discolouration in chops of low pHu (A and B) can be explained by speci®c acid catalysis of oxidation of the bright-red oxymyoglobin to the brownish-red metmyoglobin (Renerre, Anton, & Gatellier, 1992). A similar eect of low pHu on discolouration was described by Owen and Lawrie (1975). In agreement with other studies (Owen & Lawrie, 1975; Yasosky et al., 1984), the ultimate pH was found to be a very important factor for lipid oxidation, especially at low pHu. The pronounced eect of pHu below 5.8 (Fig. 7) is the same threshold value as for improved red colour (Fig. 4A). Though the TBARS values in general were low in the fresh meat, it might result in severe quality problems, if meat of pHu below 5.8 is further processed into meat product or is used for the production of precooked meat, since lipid oxidation is more pronounced in such products (Nielsen, Sùrensen, Skibsted, & Bertelsen, 1997). In an experiment with ground pork, Yasosky et al. (1984) found a critical pH value of 6.10, or even higher was necessary to obtain maximum inhibition of lipid oxidation. The drip loss was found to be highest in the chops with signi®cantly lower pHu (treatments A and B), which is in agreement with results obtained by Hatton et al. (1972). The total increase of lactate, which was closely correlated to the ultimate level of lactate, was the most important determinant for drip loss, an observation which can be related to protein denaturation, as described above for meat lightness.
Pork chops from meat with lower ultimate pH (treatments A and B) were found to have signi®cantly higher initial L*-values, resulting in an overall paler appearance, and signi®cantly higher a*- and b*-values, indicating a relatively higher proportion of the myoglobin in the oxygenated form on the surface of the chops compared with chops from meat with higher pH (treatments C and D). The colour stability (decrease in a*value) during chill storage was signi®cantly poorer, and lipid oxidation and drip loss were signi®cantly higher for meat of lower pH. While pork chops from meat of lower ultimate pH had an intial high redness, discolouration was faster, and the chops were only redder than chops from meat of higher ultimate pH at the beginning of storage. A detailed statistical analyses, indicated, that the most important parameters to explain the dierences in colour, lipid oxidation and drip loss were pHu and ultimate lactate concentration in the meat. As pH and lactate were highly correlated, these results emphazise the importance of obtaining an ultimate pH in the meat in the narrow range where both microbiological aspects as well as meat quality parameters such as colour, lipid oxidation and drip loss can be controlled. Measurements of pH should accordingly be considered important in any quality assurance system for pork. Acknowledgements This work was part of a collaboration project between LMC-Centre for Advanced Food Studies and The Federation of Danish Pig Producers and Slaughterhouses and sponsored by the FéTEK programme through the Danish Ministry of Food, Agriculture and Fisheries. We wish to thank Jùrgen Steen Jensen and Bente Sùrensen for their excellent technical assistance. References Andersen, H. J., Bertelsen, G., & Skibsted, L. H. (1988). Salt eect on acid-catalyzed autoxidation of oxymyoglobin. Acta Chemica Scandinavia, A42, 226±236. Ashmore, C. R., Parker, W., & Doerr, L. (1972). Respiration of mitochondria isolated from dark-cutting beef: postmortem changes. Journal of Animal Science, 34, 46±48. Barton-Gade, P. A., Demeyer, D., Honikel, K. O., Joseph, R. L., Poulanne, E., Severini, M., Smulders, F. J. M., & Tornberg, E. (1993). Reference methods for water holding capacity in meat and meat products: procedures recommended by an OECD working group. 39th International Congress of Meat Science and Technology, August 1±6, 1993, Calgary Alberta, Canada. Bendall, J. R. (1973). Post mortem changes in muscles. In G. H. Bourne, Structure and function of muscle (pp. 243±309). New York: Academic Press. Briskey, E. J. (1964). Etiological status and associated studies of pale,
D. Juncher et al. / Meat Science 58 (2001) 347±357 soft, exudative porcine musculature. In C. O. Chichester, E. M. Mrak, & G. F. Stewart, Advances in food research (vol. 13) (pp. 89±178). New York: Academic Press. Buttriss, J. L., & Diplock, A. T. (1984). High-performance liquid chromatography methods for vitamin E in tissues. Methods in Enzymology, 5, 131±138. Danish Standardization Board (Dansk StandardiseringsraÊd). (1982a). Water quality Ð determination of metal in water, sludge and sediment by atomic absorption spectrophotometry. Special guidelines for cadmium, cobalt, iron, nickel, lead and zink (in Danish). Dansk Standard, DS 263, (1st ed.). January 1982. Danish Standardization Board (Dansk StandardiseringsraÊd). (1982b). Determination of metals in water, sludge and sediments Ð general guidelines for determination by atomic absorption spectrophotometry in ¯ame (in Danish). Dansk Standard, DS 259, (1st ed.). January 1982. Faustmann, C., & Cassens, R. G. (1990). The biochemical basis for discoloration in fresh meat: a review. Journal of Muscle Foods, 1, 217±243. Gilman, L. B. (1989). Microwave sample preparation. CEM Corporation, Matthews, NC28106. Gregory, N. G. (1998). Animal welfare and meat science. Cambridge, UK: University Press (Chapter 6).. Guignot, F., Touraille, C., Ouali, A., Renerre, M., & Monin, G. (1994). Relationships between post-mortem pH changes and some traits of sensory quality in veal. Meat Science, 37, 315±325. Hatton, M. W. C., Lawrie, R. A., Ratcli, P. W., & Wayne, N. (1972). Eects of preslaughter adrenalin injection on muscle metabolites and meat quality of pigs. Journal of Food Technology, 7, 443±453. Henckel, P., Karlsson, A., Jensen, M. T., Oksbjerg, N., Petersen, J. S. (2001). Energy levels in pig muscles and their development post mortem after standardised treatments prior to slaughter. Submitted to Meat Science. Henckel, P., Karlsson, A., Oksbjerg, N., & Petersen, J. S. (2000). Control of post mortem pH-decrease in pig muscles: experimental design and testing of animal models. Meat Science, 55(1), 131±138. Henckel, P., Karlsson, A., Petersen, J. S. (1997). Exercise, fasting, and epinephrine administration Ð eects on muscle glycogen degradation in vivo in pig M. Longissimus dorsi. Proceeding 43rd International Congress of Meat Science and Technology, (pp. 294±295). New Zealand.
357
Hunt, M. C., Sùrheim, O., & Slinde, E. (1999). Color and heat denaturation of myoglobin forms in ground beef. Journal of Food Science, 64(5), 847±851. Jart, A. (1997). The magnetic ®eld as an additional selectivity parameter in fat hydrogenation. Journal American Oil Chemist Society, 74, 615±617. Jensen, C., Guidera, J., Skovgaard, I. M., Staun, H., Skibsted, L. H., Jensen, S. K., Mùller, A. J., Buckly, J., & Bertelsen, G. (1997). Eects of dietary a-tocopheryl acetate supplementation on a-tocopherol deposition in porcine m. psoas major and m. longissimus dorsi and on drip loss, colour stability and oxidative stability of pork meat. Meat Science, 45, 491±500. Nielsen, J. H., Sùrensen, B., Skibsted, L. H., & Bertelsen, G. (1997). Oxidation in pre-cooked minced pork as in¯uenced by chill storage of raw muscle. Meat Science, 46, 191±197. Owen, J. E., & Lawrie, R. A. (1975). The eect of an arti®cially induced high pH on the susceptibility of minced porcine muscle to undergo oxidative rancidity under frozen storage. Journal of Food Technology, 10, 169±180. Renerre, M. (1990). Review: factors involved in the discoloration of beef meat. International Journal of Food Science and Technology, 25, 613±630. Renerre, M., Anton, M., & Gatellier, P. (1992). Autoxidation of puri®ed myoglobin from two bovine muscles. Meat Science, 32, 331±342. SAS (1996). SAS/STATTM software (ver. 6.12). Cary, NC, USA: SAS Institute Inc. Skibsted, L. H., Mikkelsen, A., & Bertelsen, G. (1998). Lipid-derived o-¯avours in meat. In F. Shahidi, Flavour of meat and meat products (2nd ed.) (pp. 217±256). London: Blackie Academic & Professional. Vyncke, W. (1975). Evaluation of the direct thiobarbituric acid extraction method for determining oxidative rancidity in mackerel (Scomber scombrus L.). Fette-Seifen-Anstrichmittel, 77, 239±240. Warriss, P. D., Bevis, E. A., & Ekins, P. J. (1989). The relationships between glycogen stores and muscle ultimate pH in commercially slaughtered pigs. British Veterinary Journal, 145, 378±383. Yasosky, J. J., Aberle, E. D., Peng, I. C., Mills, E. W., & Judge, M. D. (1984). Eects of pH and time of grinding on lipid oxidation of fresh ground pork. Journal of Food Science, 49, 1510±1512.