- Email: [email protected]
Early Post Mortem pH decrease in porcine M. longissimus dorsi of PSE, Normal and DFD quality
Meat Science 34 (1993) 131-143
Early Post Mortem p H D e c r e a s e in Porcine
M. Longissimus dorsi of PSE, Normal and DFD Quality Anne-Charlotte Enf/ilt, a Kerstin Lundstr6m ~ & Ulla Engstrand b a Department of Food Science, b Department of Statistics Data Processing and Agricultural Extension Swedish University of Agricultural Sciences, S-750 07 Uppsala, Sweden (Received 3 December 1991; revised version received 18 February 1992; accepted 22 February 1992)
A BS TRA CT The purpose of this investigation was to compare the muscle p H at exsanguination and the rate of p H changes in porcine M. longissimus dorsi (LD) of normal DFD (Dark, Firm, Dry) and PSE (Pale, Soft, Exudative) quality. The p H was continuously measured in the LD in 116 carcasses during the first 50 min post mortem. Calculations were made both on measured pH-values and on pH-values transformed to hydrogen ion concentrations. A regression of p H or hydrogen ion concentration on time was made for each animal These individual regressions were then combined, using a multivariate analysis to estimate regression curves for each meat quality class. The two methods for expressing p H gave somewhat different results. The relationship between the hydrogen ion concentration and time was found to be linear for normal and DFD muscles and quadratic for PSE muscles. As a consequence of the mathematical properties of pH, the relationship between p H and time was found to be quadratic for the normal and DFD qualities, and linear for the PSE quality. For both methods of calculations the slopes for the regression curves were significantly different between PSE and the other two quality classes with both calculating methods, while the slopes did not differ between normal and DFD muscle qualities. The intercepts of the regression curves differed significantly between PSE and the other two quality 131
Meat Science 0309-1740/93/$06.00 © 1993 Elsevier Science Publishers Ltd, England. Printed in Great Britain
132
Ann-Charlotte Enfalt, Kerstin Lundstrrm, Ulla Engstrand classes only when the calculations were made on measured pH-values without transformation. A temporary increase in p H was seen in some normal and PSE carcasses during the measured time period. Development of muscles with PSE characteristics thus seems to be initiated by a combination of a lower muscle-pH already at exsanguination and a faster pH decrease. It is also of importance to consider the special mathematical properties of the pH-value.
INTRODUCTION The occurrence of muscles with PSE (Pale, Soft, Exudative) and DFD (Dark, Firm, Dry) characteristics is unfavourable since both these deviations cause meat with less good quality. PSE meat has pale colour, soft consistency and less good water-holding capacity. DFD meat has darker colour and higher ultimate pH than normal meat. Many investigations have been performed to find out the causes of these deviations. Pigs carrying the halothane-gene are generally more stress-susceptible, which may result in a higher incidence of PSE and less good meat quality than pigs without this gene (Jensen & Barton-Gade, 1985; Lundstrrm et al., 1989). The handling of the animals prior to slaughter, and too short a resting period at the slaughterhouse also has an influence on the development of both PSE and DFD as reviewed by Warris (1987). The development of PSE is usually attributed to increased glycolysis rate post mortem (Wismer-Pedersen, 1959; Wismer-Pedersen & Briskey, 1961). In muscles which develop DFD, the muscle glycogen is already depleted before slaughter. This gives less substrate for the post mortem glycolysis and the ultimate pH becomes higher than normal, as reviewed by Bendall & Swatland (1988). When PSE develops in a muscle, pH drops to values lower than 5.8 at 45 min post mortem (Wismer-Pedersen, 1959; Scheper, 1971). This should be compared with normal muscles in which the pH decreases from approximately 7 in living muscles (range 6-9-7.3) (Tarrant et al., 1972; Bendall, 1973; L/Swe et al., 1977) to values between 5-3 and 5-8 at 24 h post mortem (Wismer-Pedersen, 1959; Briskey & Wismer-Pedersen, 1961). The lower pH in PSE muscles, combined with a high carcass temperature within the first hour post mortem, causes the proteins in the muscles to denature (Wismer-Pedersen, 1959; Penny, 1967; Honikel & Kim, 1986; Offer, 1991). This contributes to the pale colour of PSE meat (Wismer-Pedersen & Briskey, 1961; Martin et al., 1975; Honikel & Kim, 1986) and also to the reduced water-holding capacity (Wismer-Pedersen, 1959; Offer et al., 1988). Offer (1991) suggested that denaturation of
Early post mortem pH decrease in porcine M. longissimusdorsi
133
sarcoplasmic proteins in the PSE muscle had a major influence on the increase in paleness, while denaturation of the myofibrillar proteins was responsible for the decrease in water-holding capacity. If pigs are stressed prior to slaughter there may be an increased metabolic activity in the muscles. This could continue after slaughter, producing the low pH at 45 min post mortem which characterizes carcasses with PSE (Bendall, 1966; L6we et aL, 1977). The higher metabolic rate in the muscles before slaughter may result in an accumulation of lactate and a low pH in the muscles before post mortem glycolysis starts. It is thus not clear if the lower pH in PSE carcasses seen at, e.g 45 min post mortem, is mainly the result of an increased rate of pH decrease or is due to an accumulation of lactate in the muscle before slaughter. Since pH has a major influence on the denaturation of the muscle proteins when PSE develops, it is important to study changes in pH which occur in the muscles due to the post mortem process immediately after slaughter. The buffering capacity of the muscle will also influence the slope as well as the rate of the pH decrease (Bendall, 1973). Little is known about the importance of the buffering systems in the muscles post mortem. Earlier studies have mostly dealt with the decrease in pH during a longer period after slaughter, and only a few repeated measurements have been made within the first hour post mortem. As the most critical time for protein denaturation is soon after slaughter, when the temperature is still high (Penny, 1967), it would be of interest to have repeated registrations of pH within the first hour post mortem. The purpose of this study was to compare normal, D F D and PSE longissimus dorsi muscles concerning the muscle pH at exsanguination and the rate of the pH changes during the first 50 min post mortem.
MATERIALS A N D METHODS
pH measurements The animals used in this study were of purebred Swedish Yorkshire (n = 83) from the Experimental station at the Department of Animal Breeding and Genetics, Uppsala, and commercial pigs (n = 33) from the routine slaughter. The pigs were electrically stunned (90 V, 0.8-1.0 A, 15 s) and shackled by one hind leg at stunning. The carcasses remained shackled during the bleeding procedure. The scalding (63°C, 5 min) started approximately 10 min after exsanguination. After exsanguination, a cut was made with a scalpel through the skin and subcutaneous fat and the electrode was inserted into the longissimus
134
Ann-Charlotte Enftilt, Kerstin LundstrOm, Ulla Engstrand
dorsi (LD) muscle at the last rib. pH was measured using a KnickPortamess pH-meter (651-2, Berlin, BRD) with a Xerolyt® electrode (Dr. W. Ingold, Ztirich, Switzerland). Measurements started immediately after exsanguination in the shackled side, and continued after scalding until 50 min post mortem, when the carcasses were placed in a chillingroom (-17°C) for 60 min. The same measurement site was used from exsanguination to scalding, but after scalding a new cut was made approximately 10 mm cranially of the earlier cut. The carcasses were split along the spine at approximately 30 min post mortem, and measurements were thereafter made from the side of the split vertebral column in the same area as before, using the same cut for all measurements. The difference in the number of observations and times of measurements between the carcasses was the result of practical circumstances. The carcasses were kept at +4°C until cutting (approx. 24 h post mortem) and the ultimate pH (PHu) was measured using a PHM62 Standard pH-meter (Radiometer, Copenhagen, Denmark) equipped with a Xerolyt® electrode. The carcasses were divided into three quality classes, normal, PSE and D F D muscles, with PSE defined as pH at 45 min post mortem (pH45) lower than or equal to 5-8 which is often used in the literature (e.g. Scheper, 1971) and D F D defined as pH u greater than or equal to 5.87. This threshold was chosen as the mean for pHu +2 times the standard deviation in the present material (Barton-Gade, 1979).
Statistical analyses Statistical analyses of the data were carried out using the Statistical Analysis System (SAS Institute Inc., 1985). For comparison, analyses were made both on pH-values transformed to hydrogen ion concentrations and on measured pH-values. Observations for each animal were fitted to polynoms of the first and second orders using the REG-procedure. A multivariate analysis of variance (MANOVA in the GLM-procedure) was carried out, with the intercept and the regression coefficient from each individual as dependent variables and quality class as the independent variable. This gave a regression for the decrease in pH on time for each quality class, and made it possible to compare the slopes of the different quality classes. To test the differences in initial pH between quality classes, the intercepts as well as the initial pH-values were compared. The same multivariate approach was also used by Eskridge & Stevens (1987), where a detailed appendix shows the calculation and programming procedure. The multivariate approach for this analysis was favourable compared with general least-squares analysis of variance, since the variances were somewhat influenced by the time of measurement.
Early post mortem pH decrease in porcine M. longissimus dorsi
135
RESULTS Division o f the longissimus dorsi muscles into three quality classes gave 95 with normal, 13 with PSE and 8 with D F D quality. The n u m b e r o f pH measurements per carcass varied from 3 to 16 (mean = 9-1; SD = 3.5% ). Registered pH-values for the three quality classes are shown in Fig. 1. Observations are presented as means for every 5th minute. Values based on only one observation were excluded due to large individual variation. The calculations are made on the hydrogen ion concentrations and then transformed back to pH for illustration. W h e n the analyses were performed on hydrogen ion concentrations, the relationship between the decrease in pH (i.e. the increase in hydrogen ion concentration) and time during the first 50 min post mortem was found to be linear for normal and D F D quality, and nonlinear for the PSE quality. However, the linear regression curve for the PSE quality corresponded better to the measured pH-values. Therefore, a polynom of the first order was fitted for all quality classes in order to be able to make a simultaneous solution for all classes. In the PSE quality, we noticed a temporary increase in pH at 10 to 15 min post mortem in some animals. A similar increase was also seen at 20 min post mortem in the normal quality (Fig. 1). This temporary increase in pH is illustrated in Fig. 2, showing observations from two individuals with PSE. When the analyses were performed on the measured pH-values instead, the relationship between the decrease in pH and time was linear for PSE pH 7.2OOO DFD xxx Normal *** PSE
7.06.86.6-
O X
6.4-"
X X
X
0
(~ ~
6.2-
x
*
6.o-
*
5.8-
*
*
5.65"4~, 0
o
* . . . . . . . . .
,
10
. . . . . . . . .
~
. . . . . . . . .
2O
,
. . . . . . . . .
3O
~
* . . . . . . . . .
4O
i
50 Min.
Fig. 1. Observed pH-values in M. longissimus dorsi of normal (n = 95), PSE (n = 13) and DFD (n = 8) quality during the first 50 min post mortem. Means for every 5th minute were calculated on the hydrogen ion concentrations and transformed back to pH.
136
Ann-Charlotte Enfait, Kerstin LundstrOm, Ulla Engstrand pH
7.2~ 7.06.86.66.4-
6.26,0-" 5.85.65.4.........
;o . . . . . . . . .
~o . . . . . . . . .
~o . . . . . . . . .
~o . . . . . . . . .
;o
Mln.
Fig. 2. Observed pH-values in M. longissimus dorsi of PSE quality from two individuals illustrating the temporary increase in pH.
quality and nonlinear for the normal and D F D quality. Regression curves obtained from the linear regressions were as follows: Normal: F r o m [H+]: F r o m pH:
Y = - l o g (2 • 21 x 10 -7 + 6-18 × 10 -9 X At) Y = 6-74 - 0.0086 × X
PSE: F r o m [H÷]: F r o m pH:
Y = - l o g (2.61 X 10 -7 + 3-67 X 10-8 X X) Y = 6-52 - 0.0190 X X
DFD: F r o m [H+]: F r o m pH:
Y = - l o g (1.29 x 10 -7 + 7.28 x lO -9 X JO Y=6.90-0.0117 x X
where Y = pH; X = time. Both analyses on measured pH-values and on hydrogen ion concentrations showed a significant difference in the slopes o f the linear regression curves between PSE and the other two quality classes (PSE vs normal p < 0.001; PSE vs D F D p < 0.001), while the slopes did not differ between normal and D F D quality (p > 0-05). Figure 3 shows the regression curves from the analysis on hydrogen ion concentration. F o r comparison, both curves from the linear (Fig. 3a) and the quadratic analyses are shown (Fig. 3b). Only when calculations were m a d e on measured pH, was the intercept for muscles with PSE quality significantly lower than normal, and D F D qualities (PSE vs D F D p < 0.001; PSE vs normal p < 0-01). The differ-
Early post mortem pH decrease in porcine M. longissimus dorsi
137
ooo DFD xxx Normal
7.06.8-
6.66.46.26.05.8" 5.65.4-
.
.
.
.
.
.
.
.
.
i
. . . . . . . . .
i
10
. . . . . . . . .
i
20
.
.
.
.
.
.
.
.
.
i
30
. . . . . . . . .
t
40
50 Min.
(a) pH
7.27.06.8.
~ ~
ooo DFD
xxx Normal PSE
6.6~ 6.4S 6.2-: 6.0-" 5.85.65.4-
.
.
.
.
.
.
.
.
.
i
.
10
.
.
.
.
.
.
.
.
i
.
.
.
.
.
20
.
.
.
.
i
30
.
.
.
.
.
.
.
.
.
i
40
.
.
.
.
.
.
.
.
.
!
50 Min.
(b) Fig. 3. The regression curves for the decrease in pH in M. longissimus dorsi of normal, PSE and DFD quality during the first 50 rain post mortem (for number of animals, see Fig. 1). Calculations were made on hydrogen ion concentration and transformed back to pH for illustration. (a) Curves from the linear regression. (b) Curves from the quadratic regression. ence between normal and D F D quality was not significant in any o f the analyses (p = 0.15 for [H÷]; p = 0-07 for measured pH; refer to Table 1). When the first registered values, pHi, were compared, the order between the three quality classes was similar, but not significantly different. The same tendency was seen both when the calculations were made with measured pH-values and with hydrogen ion concentrations, but the levels differed (Table 1). To be able to compare our results with other
138
Ann-Charlotte Enfalt, Kerstin Lundstr6m, Ulla Engstrand
TABLE 1 Initial p H (PHi), p H at 45 m i n post mortem (PH45), U l t i m a t e p H (pHt0, Intercepts and b-values from the Regression of pH on Time in LD-muscles with Normal, PSE and D F D Quality (from the Linear Regressions)
Quality class PSE
Normal
DFD
(n = 13)
(n =95)
(n = 8)
6.61 a 6.58 a 5.72 a 5.49 a
6.78 ab 6.66 a 6.30 b 5.52 a
7.05 b 6.89 a 6.34 b 5.96 b
6.78 a 6.52 a 5-67 ~ -0.0190 a 5.50 ~
6.90 ab 6.74 b 6.35 b -0.0086 b 5.53 a
7.08 b 6.90 b 6.37 b -0.0117 b 5-97 b
Calculation from [H +] pH i
Intercept pH45 pHu
Calculation from pH pH i
Intercept pH45
b-value (units/min) pHu
Means within row with the same letter are not significantly different (p > 0.05).
investigations, the regression-coefficients from the linear regression of the measured pH-values without transformation are shown (Table 1).
DISCUSSION When calculations such as the mean are performed on pH values, it is important to consider that pH is a logarithmic value, defined as the negative logarithm of the hydrogen ion activity. Usually the activity is set equal to the concentration, giving pH = - l o g [H+]. According to Murphy (1982) and Hofmann (1987), errors may arise when making calculations with the pH-values. In order to get means of pH-values, one should use the concentrations of hydrogen ions [H÷], and then take the negative logarithm of this mean. In the present study we found differences between the calculations made on hydrogen ion concentrations and those made on measured pH-values. In general, means calculated from measured pH-values were higher than those calculated via the hydrogen ion concentration. For example, calculating means for every fifth minute gave a difference of 0.13 pH units for the first calculated values, and 0.07 pH units at 30 min post mortem for the PSE quality. The difference between the two methods of calculation was influenced by the number of
Early post mortempH decrease in porcine M. longissimusdorsi
139
observations included and the variation at each time. If the variation was large, the difference became more pronounced, which could lead to somewhat different conclusions. The variation within a certain time was large in this study, due to individual variation between animals within each quality class. In the present study, we used a probe for direct pH measurement. This method can be compared with the sampling method, where a small piece of meat is taken out and then homogenized in iodoacetate or KC1 before the pH-measurement. The sampling method leads to the measurement being from a greater area than the probe, which measures only at the area where the aperture diaphragm of the probe is in contact with the meat. Bager & Petersen (1983) compared the sampling method with the probe method and found that the two methods gave the same results for ultimate pH measurements. During the pre-rigor state, probe measurements gave higher within-muscle variation and thus proved less precise when the pH is still declining. The repeated measurements adopted in this study should, however, reduce the error caused by the within-muscle variation. Making pH measurements in the ordinary slaughterhouse environment often involves practical problems in comparison with laboratory studies where a muscle is taken out of the carcass at slaughter. One disadvantage of direct measurements is the difficulty of standardizing the times of measurement because of the practical problems involved. In addition, it is impossible to make registrations during some periods of the post mortem glycolysis because of scalding and other operations undergone by the carcasses. On the other hand, results obtained by direct measurement can be applied in practice. The time from exsanguination to scalding differed between the carcasses in this study. This gave us the opportunity to take pH measurements in some carcasses during the time-period when they would normally have been scalded. For the last 15 years pig breeding in Sweden has been working on elimination of the halothane-gene. The work has been successful, and today the frequency of the gene in the breeding population is very low (Petterson & Gahne, 1988). In fact, in this study, it was even difficult to find carcasses with PSE and D F D loins, giving the small number of muscles with deviating quality. In a simulation study, Offer (1991) divided PSE muscles into three classes depending on the rate of the pH decrease, where a decrease of 0.10 units/min corresponded to serious PSE, and decreases of 0.03 units/min and 0-02 units/min to moderate and marginal PSE, respectively. In the present study, the rate of pH decrease in PSE quality was 0.019 units/min and would consequently be called marginal PSE (according to Offer, 1991). It should be mentioned that
140
Ann-Charlotte Enfalt, Kerstin Lundstr6m, Ulla Engstrand
Offer calculated the rates under the assumption that muscle pH was 7.0 at slaughter and that the decrease was linear. This was not the case in the present study. Instead, the initial pH was well below 7-0 in the PSE quality leading to a lower calculated rate for the pH decrease. During stunning and exsanguination of pigs, large changes occur in the muscles. The term 'initial pH' is often used in the literature to describe the pH-value registered first, independent of time. This may be quite confusing when results from different investigations are compared. Wismer-Pedersen (1959) observed values around 6.7 in muscle samples taken just before the pigs were slaughtered and Tarrant et al. (1972) observed a change in pH from 7.0 in the living muscle to 6.3 at 5 min post mortem. The results of the present study indicate that the muscle pH at exsanguination is somewhat lower in loins in which PSE will develop compared with those of normal or DFD quality. The same tendency was observed both when comparing intercepts from the regression equations and the means for the value registered first. It should be noted however, that the significance of the differences between quality classes depends on the method of calculation (pH vs. H÷-concentration). Bendall et al. (1963) also reported significantly lower pH in PSE carcasses than in normal carcasses (6.66 vs. 6.78) at 10 minutes post mortem. As the PSE carcasses had a higher rate of decrease in pH between 10 and 200 min post mortem, they concluded that there could not have been any difference in the pH between normal and PSE muscles when extrapolating to the time at exsanguination. Bendall (1966) reported a wide variation in post mortem pH (30 min), both between animals and within the LD muscle, indicating the importance of a well-defined site of measurement, and the necessity of using many animals to get representative results. When calculating the rate of decrease in pH, one should bear in mind that the unavoidable damage to muscle tissue caused by insertion of the electrode may stimulate post mortem glycolysis and decrease of pH (Hofmann, 1987). By keeping the pH electrode in the muscle between registrations, we found that a faster decrease of pH occurred compared to inserting the electrode only at the registrations (Enf'alt, unpublished). This may be due to a stimulation of glycolysis when the electrode remains in the muscle. The rate of the pH decrease seems to change with time and actual pH-level (Bendall et al., 1963; Bendall, 1966; Kivikari & Poulanne, 1989). Bendall et al. (1963) studied the rate of fall in pH post mortem, keeping muscle samples under constant temperature (37°C). Normal muscles had a biphasic decrease in pH, with a lower rate when pH was higher than 6.5 (approximately the first 90 min post mortem). In contrast, Bendall (1966) reported a maximum rate of fall in pH in the region of pH 7 and
Early post mortem pH decrease in porcine M. longissimusdorsi
141
6 in normal pigs, and Kivikari & Poulanne (1989) also found a higher rate at the early part of the post mortem period. This was also the case for normal and D F D muscles in the present study when the calculations were made on measured pH-values. Bendall et al. (1963) found a constant rate in PSE muscles during the first 150 min post mortem. A linear pH decrease for PSE muscles was also found in the present study when the calculations were made on measured pH-values. Earlier studies concerning the rate of decrease in pH have been performed over longer periods than in this study, and the regressions obtained have all been linear (e.g. Bendall et aL, 1963; Hallund & Bendall, 1965; Tarrant et al., 1972). Comparison of our results with others is possible only with the non-transformed linear regressions. Tarrant et al. (1972) reported an overall decrease of 0.003 units/min during the first 7 h post mortem in normal muscles, and a rate for the first hour post mortem of 0.0117 units/min. The latter result is in good agreement with the rate of normal and D F D muscles in the present study, which were 0.0086 and 0.012, respectively, using the linear regression of non-transformed pH-values. The rates found here also agree with results reported by Bendall (1966) in normal pigs. The decrease for muscles with normal and D F D quality in the present study seems to be rather similar during the measured time period and this was also the case for rabbit psoas muscles reported by Bendall (1973). The rate for PSE carcasses in our study was the same as found by Bendall et al. (1963). The difference in the slopes of the regression found between the analyses with transformed pH-values and measured pH-values, respectively, in the present study, is as expected from a mathematical point of view, since logarithmic transformation is a way to get a linear relation from a curvilinear. This could lead to some bias when using the pH-values without transformation. It should also be pointed out that the variation between carcasses in the pH changes post mortem makes it difficult to calculate general regression curves. Even if the quadratic component was significant for PSE quality in the analysis of hydrogen ion concentration, predicted values from the quadratic curve did not correspond to true values in the first part of the post mortem period. The disagreement was caused by a partial increase in pH during the measured time period in some of the PSE muscles, influencing the regression in a way that made the intercept lower than the true values. If the increase had occurred in a single animal it could have been taken as an error in the measurements, but since it was found in several animals there seems to be a real increase. A temporary increase was also seen in some of the muscles within the normal quality but as this quality class had many animals without any obvious increase no
142
Ann-Charlotte Enfalt, Kerstin Lundstr6m, Ulla Engstrand
influence was found on the regression curve. At the moment, no explanation to this p h e n o m e n o n can be found, but it is of great interest and needs to be further investigated. The post mortem rate of fall in pH is probably somewhat faster in PSE muscles than in muscles of normal or D F D quality. The development of muscles with PSE characteristics thus seems to be initiated by a combination of a lower muscle-pH already at exsanguination due to lactate accumulation before slaughter, and a faster pH-decrease post mortem. The variation in pH level at the start of the post mortem process, as well as the temporary increase in pH during the post mortem process, is of special interest.
ACKNOWLEDGEMENTS The authors wish to thank Mr Anders Karlsson for valuable comment on the text. Financial support for this work was provided by the Farmers' Research Council for Information and Development, and the Swedish Council for Forestry and Agricultural Research.
REFERENCES Bager, F1. & Petersen, G. V. (1983). Nord. Vet.-Med., 35, 86. Barton-Gade, P. A, (1979). Acta Agric. Scand. Suppl., 21, 61. Bendall, J. R. (1966). J. Sci. Food Agric., 17, 333. Bendall, J. R. (1973). In The structure and function of muscle, ed. G. H. Bourne. Academic Press, New York and London, p. 244. Bendall, J. R., Hallund, O. & Wismer-Pedersen, J. (1963). J. Food Sci., 28, 156. Bendall, J. R. & Swatland, H. J. (1988). Meat Sci., 24, 85. Briskey, E. J. & Wismer-Pedersen, J. (1961). J. Food Sci., 26, 297. Eskridge, K. M. & Stevens, E. J. (1987). Agron. J., 79, 291. Hallund, O. & Bendall, J. R. (1965). J. Food Sci., 30, 296. Hofmann, K. (1987). Proc. 33rd Int. Congr. Meat Sci. Technol. Helsinki, p. 359. Honikel, K. O. & Kim, C-J. (1986). Fleischwirts., 66(3), 349. Jensen, P. & Barton-Gade, P. A. (1985). In Stress susceptibility and meat quality in pigs, ed. J. B. Ludvigsen. EAAP publication No. 33, p. 80. Kim, C. J. (1984). Ver~inderungen im Schweinemuskel nach dem Schlachten und deren Bedeutung for WasserbindungsvermSgen und Verarbetungseigenschaften des Fleisches. Dissertation, Giessen. Kivikari, R. & Poulanne, E. (1989). Proc. 35th Int. Congr. Meat Sci. Technol. Copenhagen, p. 201. LSwe, G., Steinhardt, M, Beutling, D., Lyhs, L. & Fachmin, G. (1977). Arch. exper. Vet. med. (Leipzig), 31,643. Lundstr0m, K., Essen-Gustavsson, B., Rundgren, M, Edfors-Lilja, I. & Malmfors, G. (1989). Meat Sci., 25, 251.
Early post mortem pH decrease in porcine M. longissimus dorsi
143
Martin, A. H., Fredeen, H. T. & L'Hirondelle, P. J. (1975). Can. J. Anim. Sci., 55, 527. Murphy, M. R. (1982). J. Dairy Sci., 65, 161. Offer, G. (1991). Meat Sci., 30, 157. Offer, G., Puslow, P., Almond, R., Cousins, T., Elsey, J., Lewis, G., Parsons, N. & Sharp, A. (1988). Proc. 34th Int. Congr. Meat Sci. Technol. Brisbane, p. 161. Penny, I. F. (1967). Biochem. J., 104, 609. Petterson, H. & Gahne, B. (1988). SvinskOtsel, 78(10), 18. SAS Institute Inc. (1985). SAS User's guide: Statistics Version 5 Edition. SAS Institute Inc., Cary, NC. Scheper, J. (1971). Proc. 2nd int. Symp. Condition Meat Quality Pigs. Zeist, p. 271. Tarrant, P. J. V., McLoughlin, J. V. & Harrington, M. G. (1972). Proc. Roy. Irish Acad., 72, 55. Warris, P. D. (1987). In Evaluation and control of meat quality in pigs, ed. P. V. Tarrant, G. Eikelenboom & G. Monin. Martinus Nijhoff Publishers, Dordrecht, The Netherlands, p. 245. Wheeler, T. J. & Lowenstein, J. M. (1979). J. Biol Chem., 254, 8894. Wismer-Pedersen, J. (1959). Food Res., 24, 711. Wismer-Pedersen, J. & Briskey, E. J. (1961). J. Food Technol., 15, 232.
Early Post Mortem p H D e c r e a s e in Porcine
M. Longissimus dorsi of PSE, Normal and DFD Quality Anne-Charlotte Enf/ilt, a Kerstin Lundstr6m ~ & Ulla Engstrand b a Department of Food Science, b Department of Statistics Data Processing and Agricultural Extension Swedish University of Agricultural Sciences, S-750 07 Uppsala, Sweden (Received 3 December 1991; revised version received 18 February 1992; accepted 22 February 1992)
A BS TRA CT The purpose of this investigation was to compare the muscle p H at exsanguination and the rate of p H changes in porcine M. longissimus dorsi (LD) of normal DFD (Dark, Firm, Dry) and PSE (Pale, Soft, Exudative) quality. The p H was continuously measured in the LD in 116 carcasses during the first 50 min post mortem. Calculations were made both on measured pH-values and on pH-values transformed to hydrogen ion concentrations. A regression of p H or hydrogen ion concentration on time was made for each animal These individual regressions were then combined, using a multivariate analysis to estimate regression curves for each meat quality class. The two methods for expressing p H gave somewhat different results. The relationship between the hydrogen ion concentration and time was found to be linear for normal and DFD muscles and quadratic for PSE muscles. As a consequence of the mathematical properties of pH, the relationship between p H and time was found to be quadratic for the normal and DFD qualities, and linear for the PSE quality. For both methods of calculations the slopes for the regression curves were significantly different between PSE and the other two quality classes with both calculating methods, while the slopes did not differ between normal and DFD muscle qualities. The intercepts of the regression curves differed significantly between PSE and the other two quality 131
Meat Science 0309-1740/93/$06.00 © 1993 Elsevier Science Publishers Ltd, England. Printed in Great Britain
132
Ann-Charlotte Enfalt, Kerstin Lundstrrm, Ulla Engstrand classes only when the calculations were made on measured pH-values without transformation. A temporary increase in p H was seen in some normal and PSE carcasses during the measured time period. Development of muscles with PSE characteristics thus seems to be initiated by a combination of a lower muscle-pH already at exsanguination and a faster pH decrease. It is also of importance to consider the special mathematical properties of the pH-value.
INTRODUCTION The occurrence of muscles with PSE (Pale, Soft, Exudative) and DFD (Dark, Firm, Dry) characteristics is unfavourable since both these deviations cause meat with less good quality. PSE meat has pale colour, soft consistency and less good water-holding capacity. DFD meat has darker colour and higher ultimate pH than normal meat. Many investigations have been performed to find out the causes of these deviations. Pigs carrying the halothane-gene are generally more stress-susceptible, which may result in a higher incidence of PSE and less good meat quality than pigs without this gene (Jensen & Barton-Gade, 1985; Lundstrrm et al., 1989). The handling of the animals prior to slaughter, and too short a resting period at the slaughterhouse also has an influence on the development of both PSE and DFD as reviewed by Warris (1987). The development of PSE is usually attributed to increased glycolysis rate post mortem (Wismer-Pedersen, 1959; Wismer-Pedersen & Briskey, 1961). In muscles which develop DFD, the muscle glycogen is already depleted before slaughter. This gives less substrate for the post mortem glycolysis and the ultimate pH becomes higher than normal, as reviewed by Bendall & Swatland (1988). When PSE develops in a muscle, pH drops to values lower than 5.8 at 45 min post mortem (Wismer-Pedersen, 1959; Scheper, 1971). This should be compared with normal muscles in which the pH decreases from approximately 7 in living muscles (range 6-9-7.3) (Tarrant et al., 1972; Bendall, 1973; L/Swe et al., 1977) to values between 5-3 and 5-8 at 24 h post mortem (Wismer-Pedersen, 1959; Briskey & Wismer-Pedersen, 1961). The lower pH in PSE muscles, combined with a high carcass temperature within the first hour post mortem, causes the proteins in the muscles to denature (Wismer-Pedersen, 1959; Penny, 1967; Honikel & Kim, 1986; Offer, 1991). This contributes to the pale colour of PSE meat (Wismer-Pedersen & Briskey, 1961; Martin et al., 1975; Honikel & Kim, 1986) and also to the reduced water-holding capacity (Wismer-Pedersen, 1959; Offer et al., 1988). Offer (1991) suggested that denaturation of
Early post mortem pH decrease in porcine M. longissimusdorsi
133
sarcoplasmic proteins in the PSE muscle had a major influence on the increase in paleness, while denaturation of the myofibrillar proteins was responsible for the decrease in water-holding capacity. If pigs are stressed prior to slaughter there may be an increased metabolic activity in the muscles. This could continue after slaughter, producing the low pH at 45 min post mortem which characterizes carcasses with PSE (Bendall, 1966; L6we et aL, 1977). The higher metabolic rate in the muscles before slaughter may result in an accumulation of lactate and a low pH in the muscles before post mortem glycolysis starts. It is thus not clear if the lower pH in PSE carcasses seen at, e.g 45 min post mortem, is mainly the result of an increased rate of pH decrease or is due to an accumulation of lactate in the muscle before slaughter. Since pH has a major influence on the denaturation of the muscle proteins when PSE develops, it is important to study changes in pH which occur in the muscles due to the post mortem process immediately after slaughter. The buffering capacity of the muscle will also influence the slope as well as the rate of the pH decrease (Bendall, 1973). Little is known about the importance of the buffering systems in the muscles post mortem. Earlier studies have mostly dealt with the decrease in pH during a longer period after slaughter, and only a few repeated measurements have been made within the first hour post mortem. As the most critical time for protein denaturation is soon after slaughter, when the temperature is still high (Penny, 1967), it would be of interest to have repeated registrations of pH within the first hour post mortem. The purpose of this study was to compare normal, D F D and PSE longissimus dorsi muscles concerning the muscle pH at exsanguination and the rate of the pH changes during the first 50 min post mortem.
MATERIALS A N D METHODS
pH measurements The animals used in this study were of purebred Swedish Yorkshire (n = 83) from the Experimental station at the Department of Animal Breeding and Genetics, Uppsala, and commercial pigs (n = 33) from the routine slaughter. The pigs were electrically stunned (90 V, 0.8-1.0 A, 15 s) and shackled by one hind leg at stunning. The carcasses remained shackled during the bleeding procedure. The scalding (63°C, 5 min) started approximately 10 min after exsanguination. After exsanguination, a cut was made with a scalpel through the skin and subcutaneous fat and the electrode was inserted into the longissimus
134
Ann-Charlotte Enftilt, Kerstin LundstrOm, Ulla Engstrand
dorsi (LD) muscle at the last rib. pH was measured using a KnickPortamess pH-meter (651-2, Berlin, BRD) with a Xerolyt® electrode (Dr. W. Ingold, Ztirich, Switzerland). Measurements started immediately after exsanguination in the shackled side, and continued after scalding until 50 min post mortem, when the carcasses were placed in a chillingroom (-17°C) for 60 min. The same measurement site was used from exsanguination to scalding, but after scalding a new cut was made approximately 10 mm cranially of the earlier cut. The carcasses were split along the spine at approximately 30 min post mortem, and measurements were thereafter made from the side of the split vertebral column in the same area as before, using the same cut for all measurements. The difference in the number of observations and times of measurements between the carcasses was the result of practical circumstances. The carcasses were kept at +4°C until cutting (approx. 24 h post mortem) and the ultimate pH (PHu) was measured using a PHM62 Standard pH-meter (Radiometer, Copenhagen, Denmark) equipped with a Xerolyt® electrode. The carcasses were divided into three quality classes, normal, PSE and D F D muscles, with PSE defined as pH at 45 min post mortem (pH45) lower than or equal to 5-8 which is often used in the literature (e.g. Scheper, 1971) and D F D defined as pH u greater than or equal to 5.87. This threshold was chosen as the mean for pHu +2 times the standard deviation in the present material (Barton-Gade, 1979).
Statistical analyses Statistical analyses of the data were carried out using the Statistical Analysis System (SAS Institute Inc., 1985). For comparison, analyses were made both on pH-values transformed to hydrogen ion concentrations and on measured pH-values. Observations for each animal were fitted to polynoms of the first and second orders using the REG-procedure. A multivariate analysis of variance (MANOVA in the GLM-procedure) was carried out, with the intercept and the regression coefficient from each individual as dependent variables and quality class as the independent variable. This gave a regression for the decrease in pH on time for each quality class, and made it possible to compare the slopes of the different quality classes. To test the differences in initial pH between quality classes, the intercepts as well as the initial pH-values were compared. The same multivariate approach was also used by Eskridge & Stevens (1987), where a detailed appendix shows the calculation and programming procedure. The multivariate approach for this analysis was favourable compared with general least-squares analysis of variance, since the variances were somewhat influenced by the time of measurement.
Early post mortem pH decrease in porcine M. longissimus dorsi
135
RESULTS Division o f the longissimus dorsi muscles into three quality classes gave 95 with normal, 13 with PSE and 8 with D F D quality. The n u m b e r o f pH measurements per carcass varied from 3 to 16 (mean = 9-1; SD = 3.5% ). Registered pH-values for the three quality classes are shown in Fig. 1. Observations are presented as means for every 5th minute. Values based on only one observation were excluded due to large individual variation. The calculations are made on the hydrogen ion concentrations and then transformed back to pH for illustration. W h e n the analyses were performed on hydrogen ion concentrations, the relationship between the decrease in pH (i.e. the increase in hydrogen ion concentration) and time during the first 50 min post mortem was found to be linear for normal and D F D quality, and nonlinear for the PSE quality. However, the linear regression curve for the PSE quality corresponded better to the measured pH-values. Therefore, a polynom of the first order was fitted for all quality classes in order to be able to make a simultaneous solution for all classes. In the PSE quality, we noticed a temporary increase in pH at 10 to 15 min post mortem in some animals. A similar increase was also seen at 20 min post mortem in the normal quality (Fig. 1). This temporary increase in pH is illustrated in Fig. 2, showing observations from two individuals with PSE. When the analyses were performed on the measured pH-values instead, the relationship between the decrease in pH and time was linear for PSE pH 7.2OOO DFD xxx Normal *** PSE
7.06.86.6-
O X
6.4-"
X X
X
0
(~ ~
6.2-
x
*
6.o-
*
5.8-
*
*
5.65"4~, 0
o
* . . . . . . . . .
,
10
. . . . . . . . .
~
. . . . . . . . .
2O
,
. . . . . . . . .
3O
~
* . . . . . . . . .
4O
i
50 Min.
Fig. 1. Observed pH-values in M. longissimus dorsi of normal (n = 95), PSE (n = 13) and DFD (n = 8) quality during the first 50 min post mortem. Means for every 5th minute were calculated on the hydrogen ion concentrations and transformed back to pH.
136
Ann-Charlotte Enfait, Kerstin LundstrOm, Ulla Engstrand pH
7.2~ 7.06.86.66.4-
6.26,0-" 5.85.65.4.........
;o . . . . . . . . .
~o . . . . . . . . .
~o . . . . . . . . .
~o . . . . . . . . .
;o
Mln.
Fig. 2. Observed pH-values in M. longissimus dorsi of PSE quality from two individuals illustrating the temporary increase in pH.
quality and nonlinear for the normal and D F D quality. Regression curves obtained from the linear regressions were as follows: Normal: F r o m [H+]: F r o m pH:
Y = - l o g (2 • 21 x 10 -7 + 6-18 × 10 -9 X At) Y = 6-74 - 0.0086 × X
PSE: F r o m [H÷]: F r o m pH:
Y = - l o g (2.61 X 10 -7 + 3-67 X 10-8 X X) Y = 6-52 - 0.0190 X X
DFD: F r o m [H+]: F r o m pH:
Y = - l o g (1.29 x 10 -7 + 7.28 x lO -9 X JO Y=6.90-0.0117 x X
where Y = pH; X = time. Both analyses on measured pH-values and on hydrogen ion concentrations showed a significant difference in the slopes o f the linear regression curves between PSE and the other two quality classes (PSE vs normal p < 0.001; PSE vs D F D p < 0.001), while the slopes did not differ between normal and D F D quality (p > 0-05). Figure 3 shows the regression curves from the analysis on hydrogen ion concentration. F o r comparison, both curves from the linear (Fig. 3a) and the quadratic analyses are shown (Fig. 3b). Only when calculations were m a d e on measured pH, was the intercept for muscles with PSE quality significantly lower than normal, and D F D qualities (PSE vs D F D p < 0.001; PSE vs normal p < 0-01). The differ-
Early post mortem pH decrease in porcine M. longissimus dorsi
137
ooo DFD xxx Normal
7.06.8-
6.66.46.26.05.8" 5.65.4-
.
.
.
.
.
.
.
.
.
i
. . . . . . . . .
i
10
. . . . . . . . .
i
20
.
.
.
.
.
.
.
.
.
i
30
. . . . . . . . .
t
40
50 Min.
(a) pH
7.27.06.8.
~ ~
ooo DFD
xxx Normal PSE
6.6~ 6.4S 6.2-: 6.0-" 5.85.65.4-
.
.
.
.
.
.
.
.
.
i
.
10
.
.
.
.
.
.
.
.
i
.
.
.
.
.
20
.
.
.
.
i
30
.
.
.
.
.
.
.
.
.
i
40
.
.
.
.
.
.
.
.
.
!
50 Min.
(b) Fig. 3. The regression curves for the decrease in pH in M. longissimus dorsi of normal, PSE and DFD quality during the first 50 rain post mortem (for number of animals, see Fig. 1). Calculations were made on hydrogen ion concentration and transformed back to pH for illustration. (a) Curves from the linear regression. (b) Curves from the quadratic regression. ence between normal and D F D quality was not significant in any o f the analyses (p = 0.15 for [H÷]; p = 0-07 for measured pH; refer to Table 1). When the first registered values, pHi, were compared, the order between the three quality classes was similar, but not significantly different. The same tendency was seen both when the calculations were made with measured pH-values and with hydrogen ion concentrations, but the levels differed (Table 1). To be able to compare our results with other
138
Ann-Charlotte Enfalt, Kerstin Lundstr6m, Ulla Engstrand
TABLE 1 Initial p H (PHi), p H at 45 m i n post mortem (PH45), U l t i m a t e p H (pHt0, Intercepts and b-values from the Regression of pH on Time in LD-muscles with Normal, PSE and D F D Quality (from the Linear Regressions)
Quality class PSE
Normal
DFD
(n = 13)
(n =95)
(n = 8)
6.61 a 6.58 a 5.72 a 5.49 a
6.78 ab 6.66 a 6.30 b 5.52 a
7.05 b 6.89 a 6.34 b 5.96 b
6.78 a 6.52 a 5-67 ~ -0.0190 a 5.50 ~
6.90 ab 6.74 b 6.35 b -0.0086 b 5.53 a
7.08 b 6.90 b 6.37 b -0.0117 b 5-97 b
Calculation from [H +] pH i
Intercept pH45 pHu
Calculation from pH pH i
Intercept pH45
b-value (units/min) pHu
Means within row with the same letter are not significantly different (p > 0.05).
investigations, the regression-coefficients from the linear regression of the measured pH-values without transformation are shown (Table 1).
DISCUSSION When calculations such as the mean are performed on pH values, it is important to consider that pH is a logarithmic value, defined as the negative logarithm of the hydrogen ion activity. Usually the activity is set equal to the concentration, giving pH = - l o g [H+]. According to Murphy (1982) and Hofmann (1987), errors may arise when making calculations with the pH-values. In order to get means of pH-values, one should use the concentrations of hydrogen ions [H÷], and then take the negative logarithm of this mean. In the present study we found differences between the calculations made on hydrogen ion concentrations and those made on measured pH-values. In general, means calculated from measured pH-values were higher than those calculated via the hydrogen ion concentration. For example, calculating means for every fifth minute gave a difference of 0.13 pH units for the first calculated values, and 0.07 pH units at 30 min post mortem for the PSE quality. The difference between the two methods of calculation was influenced by the number of
Early post mortempH decrease in porcine M. longissimusdorsi
139
observations included and the variation at each time. If the variation was large, the difference became more pronounced, which could lead to somewhat different conclusions. The variation within a certain time was large in this study, due to individual variation between animals within each quality class. In the present study, we used a probe for direct pH measurement. This method can be compared with the sampling method, where a small piece of meat is taken out and then homogenized in iodoacetate or KC1 before the pH-measurement. The sampling method leads to the measurement being from a greater area than the probe, which measures only at the area where the aperture diaphragm of the probe is in contact with the meat. Bager & Petersen (1983) compared the sampling method with the probe method and found that the two methods gave the same results for ultimate pH measurements. During the pre-rigor state, probe measurements gave higher within-muscle variation and thus proved less precise when the pH is still declining. The repeated measurements adopted in this study should, however, reduce the error caused by the within-muscle variation. Making pH measurements in the ordinary slaughterhouse environment often involves practical problems in comparison with laboratory studies where a muscle is taken out of the carcass at slaughter. One disadvantage of direct measurements is the difficulty of standardizing the times of measurement because of the practical problems involved. In addition, it is impossible to make registrations during some periods of the post mortem glycolysis because of scalding and other operations undergone by the carcasses. On the other hand, results obtained by direct measurement can be applied in practice. The time from exsanguination to scalding differed between the carcasses in this study. This gave us the opportunity to take pH measurements in some carcasses during the time-period when they would normally have been scalded. For the last 15 years pig breeding in Sweden has been working on elimination of the halothane-gene. The work has been successful, and today the frequency of the gene in the breeding population is very low (Petterson & Gahne, 1988). In fact, in this study, it was even difficult to find carcasses with PSE and D F D loins, giving the small number of muscles with deviating quality. In a simulation study, Offer (1991) divided PSE muscles into three classes depending on the rate of the pH decrease, where a decrease of 0.10 units/min corresponded to serious PSE, and decreases of 0.03 units/min and 0-02 units/min to moderate and marginal PSE, respectively. In the present study, the rate of pH decrease in PSE quality was 0.019 units/min and would consequently be called marginal PSE (according to Offer, 1991). It should be mentioned that
140
Ann-Charlotte Enfalt, Kerstin Lundstr6m, Ulla Engstrand
Offer calculated the rates under the assumption that muscle pH was 7.0 at slaughter and that the decrease was linear. This was not the case in the present study. Instead, the initial pH was well below 7-0 in the PSE quality leading to a lower calculated rate for the pH decrease. During stunning and exsanguination of pigs, large changes occur in the muscles. The term 'initial pH' is often used in the literature to describe the pH-value registered first, independent of time. This may be quite confusing when results from different investigations are compared. Wismer-Pedersen (1959) observed values around 6.7 in muscle samples taken just before the pigs were slaughtered and Tarrant et al. (1972) observed a change in pH from 7.0 in the living muscle to 6.3 at 5 min post mortem. The results of the present study indicate that the muscle pH at exsanguination is somewhat lower in loins in which PSE will develop compared with those of normal or DFD quality. The same tendency was observed both when comparing intercepts from the regression equations and the means for the value registered first. It should be noted however, that the significance of the differences between quality classes depends on the method of calculation (pH vs. H÷-concentration). Bendall et al. (1963) also reported significantly lower pH in PSE carcasses than in normal carcasses (6.66 vs. 6.78) at 10 minutes post mortem. As the PSE carcasses had a higher rate of decrease in pH between 10 and 200 min post mortem, they concluded that there could not have been any difference in the pH between normal and PSE muscles when extrapolating to the time at exsanguination. Bendall (1966) reported a wide variation in post mortem pH (30 min), both between animals and within the LD muscle, indicating the importance of a well-defined site of measurement, and the necessity of using many animals to get representative results. When calculating the rate of decrease in pH, one should bear in mind that the unavoidable damage to muscle tissue caused by insertion of the electrode may stimulate post mortem glycolysis and decrease of pH (Hofmann, 1987). By keeping the pH electrode in the muscle between registrations, we found that a faster decrease of pH occurred compared to inserting the electrode only at the registrations (Enf'alt, unpublished). This may be due to a stimulation of glycolysis when the electrode remains in the muscle. The rate of the pH decrease seems to change with time and actual pH-level (Bendall et al., 1963; Bendall, 1966; Kivikari & Poulanne, 1989). Bendall et al. (1963) studied the rate of fall in pH post mortem, keeping muscle samples under constant temperature (37°C). Normal muscles had a biphasic decrease in pH, with a lower rate when pH was higher than 6.5 (approximately the first 90 min post mortem). In contrast, Bendall (1966) reported a maximum rate of fall in pH in the region of pH 7 and
Early post mortem pH decrease in porcine M. longissimusdorsi
141
6 in normal pigs, and Kivikari & Poulanne (1989) also found a higher rate at the early part of the post mortem period. This was also the case for normal and D F D muscles in the present study when the calculations were made on measured pH-values. Bendall et al. (1963) found a constant rate in PSE muscles during the first 150 min post mortem. A linear pH decrease for PSE muscles was also found in the present study when the calculations were made on measured pH-values. Earlier studies concerning the rate of decrease in pH have been performed over longer periods than in this study, and the regressions obtained have all been linear (e.g. Bendall et aL, 1963; Hallund & Bendall, 1965; Tarrant et al., 1972). Comparison of our results with others is possible only with the non-transformed linear regressions. Tarrant et al. (1972) reported an overall decrease of 0.003 units/min during the first 7 h post mortem in normal muscles, and a rate for the first hour post mortem of 0.0117 units/min. The latter result is in good agreement with the rate of normal and D F D muscles in the present study, which were 0.0086 and 0.012, respectively, using the linear regression of non-transformed pH-values. The rates found here also agree with results reported by Bendall (1966) in normal pigs. The decrease for muscles with normal and D F D quality in the present study seems to be rather similar during the measured time period and this was also the case for rabbit psoas muscles reported by Bendall (1973). The rate for PSE carcasses in our study was the same as found by Bendall et al. (1963). The difference in the slopes of the regression found between the analyses with transformed pH-values and measured pH-values, respectively, in the present study, is as expected from a mathematical point of view, since logarithmic transformation is a way to get a linear relation from a curvilinear. This could lead to some bias when using the pH-values without transformation. It should also be pointed out that the variation between carcasses in the pH changes post mortem makes it difficult to calculate general regression curves. Even if the quadratic component was significant for PSE quality in the analysis of hydrogen ion concentration, predicted values from the quadratic curve did not correspond to true values in the first part of the post mortem period. The disagreement was caused by a partial increase in pH during the measured time period in some of the PSE muscles, influencing the regression in a way that made the intercept lower than the true values. If the increase had occurred in a single animal it could have been taken as an error in the measurements, but since it was found in several animals there seems to be a real increase. A temporary increase was also seen in some of the muscles within the normal quality but as this quality class had many animals without any obvious increase no
142
Ann-Charlotte Enfalt, Kerstin Lundstr6m, Ulla Engstrand
influence was found on the regression curve. At the moment, no explanation to this p h e n o m e n o n can be found, but it is of great interest and needs to be further investigated. The post mortem rate of fall in pH is probably somewhat faster in PSE muscles than in muscles of normal or D F D quality. The development of muscles with PSE characteristics thus seems to be initiated by a combination of a lower muscle-pH already at exsanguination due to lactate accumulation before slaughter, and a faster pH-decrease post mortem. The variation in pH level at the start of the post mortem process, as well as the temporary increase in pH during the post mortem process, is of special interest.
ACKNOWLEDGEMENTS The authors wish to thank Mr Anders Karlsson for valuable comment on the text. Financial support for this work was provided by the Farmers' Research Council for Information and Development, and the Swedish Council for Forestry and Agricultural Research.
REFERENCES Bager, F1. & Petersen, G. V. (1983). Nord. Vet.-Med., 35, 86. Barton-Gade, P. A, (1979). Acta Agric. Scand. Suppl., 21, 61. Bendall, J. R. (1966). J. Sci. Food Agric., 17, 333. Bendall, J. R. (1973). In The structure and function of muscle, ed. G. H. Bourne. Academic Press, New York and London, p. 244. Bendall, J. R., Hallund, O. & Wismer-Pedersen, J. (1963). J. Food Sci., 28, 156. Bendall, J. R. & Swatland, H. J. (1988). Meat Sci., 24, 85. Briskey, E. J. & Wismer-Pedersen, J. (1961). J. Food Sci., 26, 297. Eskridge, K. M. & Stevens, E. J. (1987). Agron. J., 79, 291. Hallund, O. & Bendall, J. R. (1965). J. Food Sci., 30, 296. Hofmann, K. (1987). Proc. 33rd Int. Congr. Meat Sci. Technol. Helsinki, p. 359. Honikel, K. O. & Kim, C-J. (1986). Fleischwirts., 66(3), 349. Jensen, P. & Barton-Gade, P. A. (1985). In Stress susceptibility and meat quality in pigs, ed. J. B. Ludvigsen. EAAP publication No. 33, p. 80. Kim, C. J. (1984). Ver~inderungen im Schweinemuskel nach dem Schlachten und deren Bedeutung for WasserbindungsvermSgen und Verarbetungseigenschaften des Fleisches. Dissertation, Giessen. Kivikari, R. & Poulanne, E. (1989). Proc. 35th Int. Congr. Meat Sci. Technol. Copenhagen, p. 201. LSwe, G., Steinhardt, M, Beutling, D., Lyhs, L. & Fachmin, G. (1977). Arch. exper. Vet. med. (Leipzig), 31,643. Lundstr0m, K., Essen-Gustavsson, B., Rundgren, M, Edfors-Lilja, I. & Malmfors, G. (1989). Meat Sci., 25, 251.
Early post mortem pH decrease in porcine M. longissimus dorsi
143
Martin, A. H., Fredeen, H. T. & L'Hirondelle, P. J. (1975). Can. J. Anim. Sci., 55, 527. Murphy, M. R. (1982). J. Dairy Sci., 65, 161. Offer, G. (1991). Meat Sci., 30, 157. Offer, G., Puslow, P., Almond, R., Cousins, T., Elsey, J., Lewis, G., Parsons, N. & Sharp, A. (1988). Proc. 34th Int. Congr. Meat Sci. Technol. Brisbane, p. 161. Penny, I. F. (1967). Biochem. J., 104, 609. Petterson, H. & Gahne, B. (1988). SvinskOtsel, 78(10), 18. SAS Institute Inc. (1985). SAS User's guide: Statistics Version 5 Edition. SAS Institute Inc., Cary, NC. Scheper, J. (1971). Proc. 2nd int. Symp. Condition Meat Quality Pigs. Zeist, p. 271. Tarrant, P. J. V., McLoughlin, J. V. & Harrington, M. G. (1972). Proc. Roy. Irish Acad., 72, 55. Warris, P. D. (1987). In Evaluation and control of meat quality in pigs, ed. P. V. Tarrant, G. Eikelenboom & G. Monin. Martinus Nijhoff Publishers, Dordrecht, The Netherlands, p. 245. Wheeler, T. J. & Lowenstein, J. M. (1979). J. Biol Chem., 254, 8894. Wismer-Pedersen, J. (1959). Food Res., 24, 711. Wismer-Pedersen, J. & Briskey, E. J. (1961). J. Food Technol., 15, 232.