- Email: [email protected]
Muscle protein changes post mortem in relation to pork quality traits
PII:
SO309-1740(96)00116-7
Meat Science, Vol. 45, No. 3, 33%352, 1997 0 1997 ElsevierScienceLtd All rights rese&d. Printed in Great B&in 0309-1740/97 $17.00+0.00
ELSEVIER
Muscle Protein Changes Post Mortem Quality Traits
Relation
to
R. D. Warner,* R. G. Kauffman & M. L. Greaser Muscle Biology Laboratory,
University of Wisconsin, Madison, Wisconsin 53706, USA
(Received 15 April 1996; revised version received 21 September 1996; accepted 24 September 1996)
ABSTRACT The relationship between post-mortem traits of muscle proteins and water loss traits was investigated using 84 pork loins representing the four quality traits of PSE, RSE (reddishpink, soft, exudative). RFN (reddish-pink,$rm, non-exudative) and DFD. Protein solubility measurements (sarcoplasmic, myojibrillar and total) were lower and myosin denaturation (quantified by myofbrillar ATPase activity) was higher for PSE samples compared with samples from the other quality classes. RSE samples were similar to RFN samples in protein solubility and myosin denaturation, although RSE had lower values then DFD samples for protein solubility measurements. RFN samples had lower drip, thaw, cook and total water loss than RSE samples and all water loss traits were lowest for DFD samples and highest for PSE samples. Insoluble phosphorylase was the only characteristic that d@erentiated among PSE, RSE and RFN samples. SDS-PAGE and Western blots indicated that in PSE and RSE samples, the myojibrillar protein titin was less degraded and nebulin was more degraded compared with RFN and DFD samples. SDS-PAGE of extracted and unextracted myojibrils showed that the reduced myojibrillar solubility of PSE samples was caused by decreased extractability of the myosin heavy chain in these samples. In conclusion, although RSE samples have unacceptably high water loss, muscle protein denaturation was minimal and did not explain the low water-holding capacity. 0 1997 Elsevier Science Ltd. All rights reserved
INTRODUCTION It is well known that pork which is described as pale, soft and exudative (PSE), has high drip loss and a pale unstable colour and denaturation of muscle proteins as indicated by low solubility of sarcoplasmic and myofibrillar proteins (Wismer-Pederson, 1959; Bendall & Wismer-Pederson, 1962; Sayre & Briskey, 1963). Specifically, myosin (Penny, 1967) and the sarcoplasmic proteins phosphorylase and creatine kinase (Fischer et al., 1979) are denatured in PSE pork. *To whom correspondence should be addressed at: Victorian Agriculture Victoria, Private Bag 7, Werribee, Vie. 3030, Australia. 339
Institute
of Animal
Science,
340
R. D. Warner, R. G. Kauflman, M. L. Greaser
The water-holding capacity (WHC) of pork has been reported to be influenced by a number of factors, including ultimate pH (pH,), protein denaturation, intra- and interfascicular spacing and sarcomere length (Offer & Knight, 1988). Myosin denaturation has been suggested to be the cause of the high rate of drip loss in PSE pork (Offer, 1991). Denaturation of myosin results in shrinkage of the myosin head, drawing the thick and thin filaments closely together. This shrinkage, in addition to the shrinkage of the myofilaments due to the low ultimate pH in PSE pork, results in more fluid being expelled between tibres and fibre bundles (Offer & Knight, 1988; Irving et al., 1989). Offer (1991) assumes that myosin head shrinkage and loss of myosin catalytic activity are identical and decisive events in determining drip loss from pork muscle. However, these events may have a different dependence on pH and temperature in the immediate post-mortem period. Recently, a quality of pork which is acceptable in colour but has excessive exudation has been described and given the name RSE-reddish-pink, soft, exudative (Kauffman et al., 1992; Warner et al., 1993; Van Laack et al., 1996), but the biochemical and physiochemical traits of pork of this quality have not been described. Degradation of structural proteins post mortem is potentially important in determining water-holding capacity (WHC) and the swelling of the myofilament lattice during extraction and solubilisation of myofibrillar proteins (Offer & Trinick, 1983). The degree of myofibrillar fragmentation is lower in PSE muscle (Kang et al., 1978) and titin has been reported to be less degraded in PSE muscle (Boles et al., 1992). The importance of the rate and extent of proteolysis during the pre-rigor period, as well as during the post-rigor period, has only recently been recognised (Koohmaraie, 1992). Titin and nebulin, which are major components of the myofibrillar structure, undergo proteolysis post mortem. The degree of their integrity post mortem may affect WHC capacity (Offer & T&rick, 1983). To understand the basic mechanisms of drip loss from pork, the contribution of protein denaturation to drip loss needs to be established. In the present study, we determined muscle protein denaturation, degradation and extractability in pork samples in relation to pork quality traits. MATERIALS
AND METHODS
Samples
A total of 84 boneless pork loins were selected at 24 h post mortem from a commercial pork plant on fourteen different occasions. Initial selection was based on a visual assessment of colour and exudate, followed by measurement of pH, using an Omega pH50 portable meter and an Orion spear-tipped glass electrode. Expressible exudate from a freshly cut surface was determined by using the filter paper fluid uptake procedure (surface exudate, mg of fluid; Kauffman et al., 1986). The loins were transported in vacuumsealed bags for 2 h and measurements of physical, biochemical and water loss traits were conducted in the laboratory. A IO-cm thick sample cranial to the 10th costae was used for measurements of protein denaturation. Ten centimetres of loin caudal to the 10th costae were used for measurements of meat quality. The remaining loin, caudal to the 10th costae was stored at -20°C for a maximum of 2 months prior to determination of post-freezing water loss (thaw loss) and cooking loss. Due to logistics, not all measurements were conducted on all samples and the number of samples used for each assay is indicated in the Results. Physical measurements
of meat quality
Meat pH, was measured using equipment described above and surface colour (CIE--L’ a’ b’) was measured in triplicate on a freshly cut surface after a 15 min bloom time at 4°C
Muscle proteins andpork quality
341
using a Minolta chromameter 200b (0’ viewing angle, diffuse illumination and 8 mm optical port, standardisation using four calibration plates, Nos 11133101, 11133853, 13033041, 16332111, set to ‘auto-select’). Drip loss was measured on a size-standard&d sample (3 cm thick and 4.1 cm in diameter) and expressed as the weight loss over 48 h at 4°C during suspension of a sample (about 50 g) (Honikel et al., 1986). These measurements were used to confirm the assignment of samples to one of four quality classes as defined below: Pale, soft, exudative (PSE) Reddish-pink, soft, exudative (RSE) Reddish-pink, firm, non-exudative Dark, firm, dry (DFD)
: L’ > 50, drip > 5%, pH, < 6.0; : L’ = 42-50, drip > 5%, pH, < 6.0;
(RFN) : L* = 42-50, drip < 5% and pH, < 6.0 and : L’ < 42, drip < 5% and pH, 2 6.0.
Using these parameters, ten samples were re-classified (once actual percent drip was known) to new classes within the RFN, RSE and PSE classes. There were no changes in classification of DFD samples. For water losses during thawing and cooking, four samples, representing each quality class, were removed from the freezer, cut into 2.5~cm thick chops and trimmed of external fat (but not epimysium), while still frozen, weighed (pre-thaw weight), thawed at 4°C for about 24 h and re-weighed (pre-cook weight). Samples were broiled in an oven to an internal temperature of 68°C and weighed (post-cook weight). Thaw loss was determined as the difference between pre-thawed weight and precooked weight and expressed as a percent of pre-thaw weight. Cooked loss was determined as the difference between preand post-cooked weight expressed as a percent of precook weight. Total fluid loss was defined as the difference between pre-thawed weight and post-cooked weight, and expressed as a percent of pre-thawed weight. Protein solubility The solubility of the sarcoplasmic proteins was determined using duplicate 1 g samples homogenised in 10 ml of ice-cold 0.025 M potassium phosphate buffer (pH 7.2). For total protein solubility, duplicate 1 g samples were added to 20 ml of ice-cold 1.1 M KI/O. 1 M potassium phosphate (pH 7.2). The samples were minced with scissors, homogenised on ice with a Polytron on the lowest setting using 3 x 4 s bursts to minimise protein denaturation through heating, and then left overnight on ice. Samples were then centrifuged at 1500 g at 4°C for 20 min and the supernatants decanted and frozen for a maximum of 2 weeks before determination of protein concentration. Myofibrillar protein solubility was determined as the difference between total and sarcoplasmic protein solubility. Myofibril purification About 2-3 g of muscle were used to isolate myofibrils by adapting the procedure of Swartz et al. (1993). Muscle was excised within 24 h post mortem and added to 15 ml of rigor buffer (RB) (10 mM imidazole, 75 mM KCl, 2 mM EGTA, 2 mM MgCl*, 2 mM NaN3, pH 7.2), cut finely with scissors and Polytron homogenised using the method described above. After Dounce homogenisation with a B-pestle for 75 strokes, the homogenate was centrifuged for 20 min at 4000 g at 4°C. The pellet was re-suspended in 15 ml of RB and Dounce homogenised with the A-pestle for 100 strokes. The solution was then diluted by 30 ml with RB, homogenised with 15 more strokes then filtered through cheesecloth and centrifuged at 4000 g at 4°C. The pellet was re-suspended three times in RB with 0.5% Triton X-100 and once in RB, using a non-shearing agitator.
342
R. D. Warner, R. G. Kauffman, M. L. Greaser
The final suspension was prepared in 15 ml of RB with 50% glycerol and 1 mM DTT, and subsequently stored at -20°C in a non-defrosting freezer. ATPase assay
The myofibril suspension was removed from the freezer, washed free of glycerol by four dilution and centrifugation cycles (11500 g for 15 s) with RB- 1 mg/ml BSA- 1 mM DTT. After the final re-suspension in RB/BSA/DTT, the protein concentration was determined using the biuret assay. Calcium-activated myofibrillar ATPase activity was determined using a total protein concentration of 0.1 mg/ml in RB/BSA/DTT plus 2.2 mM CaC12. In each case, ATPase activity was determined at 22°C in triplicate 0.2 ml samples. The reaction was initiated using 5 ~1 of 0.2M ATP and was terminated after 5 min using 20 ~1 of ice-cold 25% trichloroacetic acid (TCA). A blank was included for each sample which had TCA added before addition of ATP. After centrifuging the samples at 11500 g for 15 s to precipitate the denatured protein, 0.1 ml of sample was used to determine inorganic phosphate production by the malachite green assay of Carter & Karl (1982). Standard curves were prepared using 0.65 mM K2HP04 (Sigma No. 661-9) and results were expressed as pmole of phosphate liberated per mg of protein per min. Sarcomere length
Myofibrils were removed from -20°C storage and re-suspended in RB plus 1 mg/ml BSA (RB-BSA), the concentration determined and 100 ~1 of 1 mg/ml myofibril suspension applied to each of two coverslips (No. 1.5) and left for 1 min. The coverslip was washed five times in RB-BSA, excess fluids carefully removed and the coverslip mounted on a glass slide with a drop of mounting medium (70% glycerol-RB, pH 8.5). The excess mounting medium was removed and the coverslip sealed with clear nail polish. The slides were stored at -20°C in a non-defrosting freezer and removed as required for microscopy measurements of sarcomere length. A Nikon inverted microscope (Model Diaphot) was used with phase-contrast and a 100 x (NA 1.4) oil immersion objective. The microscope was equipped with a cooled charge-coupled device (CCD, Thomson 7883, Photometrics Ltd, Tucson, AZ, USA). The CCD was controlled via a Macintosh IIfx (Apple Computer Inc, Santa Clara, CA, USA) equipped with a Matrox computer board and Nu 200 2.0 software (Photometrics Ltd). Images were captured using exposure times of 1 s and quantitative measurement of sarcomere length on captured images was conducted using the computer program IPlab (version 2.5; Signal Analytic Corp). The average sarcomere length over five sarcomeres within one myofibril was measured and 50 myofibrils were measured per sample. Protein extraction
For the purpose of examining extracted proteins on SDS-PAGE, 1 ml aliqots of suspended myofibrils of PSE, RSE and RFN quality classes were removed from frozen storage, re-suspended in RB, as for the ATPase assay except without BSA and subjected to extraction using 10 vol. (w/v) of 1.1 M KI/O.l M potassium phosphate (pH 7.2). The samples were left overnight on ice and then split into three samples: (i) control myofibril sample; (ii) sample for determining concentration and (iii) sample for SDS-PAGE. Samples (ii) and (iii) were centrifuged at 11500 g for 1 min and the supematant carefully removed. For the pellet in sample (ii), an equal weight of 0.5 M NaOH was added to solubilise the pellet prior to determining protein concentration. Protein concentration was
Muscle proteins and pork quality
343
determined in the myofibril, pellet/NaOH and supernatant samples. One milliltre of methanol was added to the remaining supernatant and pellet samples to remove the salt, carefully mixed, centrifuged at 11500 g for 1 min and subjected to vacuum for 10 min to remove the methanol. Finally, an appropriate mixture of SDS sample buffer and distilled water was added to each sample to obtain a protein concentration of 2 mg/ml prior to subjecting to SDS-PAGE. This procedure was repeated on different PSE, RSE, RFN samples and one representative gel is presented in the results. DFD samples were not included as previous results for the ATPase activity and protein solubility showed that RFN and DFD samples were not different from each other (Warner, 1994). SDS gel electrophoresis One millilitre of suspended myofibrils for all samples was removed from -20°C storage, re-suspended in RB as above and the protein concentration determined. Samples were subjected to SDS-PAGE for: (i) the examination of degradation of the high molecular proteins titin and nebulin; (ii) the quantitation of the phosphorylase and creatine kinase bands, using densitometry and (iii) the examination of myofibril, pellet and supematant fractions from the extraction procedures described above. Gel composition, electrophoresis running conditions and treatment of gels post-electrophoresis were essentially the same as described by Fritz et al. (1989). Gels used for densitometry were treated postelectrophoresis as described by Fritz et al. (1993). For resolution of creatine kinase and phosphorylase bands, the resolving gel conditions were 12% acrylamide (w/v), 0.06% bis-acrylamide (w/v), pH 9.3 and 5.0 pg of protein were applied per lane. Five lanes were used for standard concentrations of creatine kinase (SIGMA No. C3755, San Francisco, CA, USA) and phosphorylase (Boehringer Mannheim No. 108-561, Indianapolis, IN, USA) for each gel. The concentration range for each protein standard was 0.025-125 pg, as this was previously determined to be the appropriate range for the protein load used. For resolution of titin and nebulin, the resolving gel conditions were 8% acrylamide (w/ v), 0.04% bis-acrylamide (w/v), pH 8.6 and 12 pg of protein was applied to each lane. For resolution of proteins in the myosin extraction study, the gel conditions were the same as for resolution of phosphorylase and creatine kinase, but 12 pg of protein were applied per lane and a lane for molecular weight standards (BIORAD No. 161-0317, Cambridge, MA, USA) was included. A Hoefer SE280 Tall Mighty Small 11 cm slab unit (San Francisco, CA, USA) was used for separation of creatine kinase/phosphorylase bands and a Hoefer 250 Mighty Small electrophoresis unit (San Francisco, CA, USA) was used for the other gels. Western blot analysis Proteins were transferred from an unstained gel to 0.45 pm Immobilon-P (Millipore) membrane, using methods described in Fritz & Greaser (1991) except the transfer was run at 300 mA for 2 h and the temperature was maintained at 20°C. Once the transfer was complete, the membrane was stored and processed as described by Fritz & Greaser (1991). Blots were incubated with anti-nebulin monoclonal (Sigma Chemical Company, NB2, diluted 1 : 1200) for 4 h, 1 : 7500 dilution of alkaline phosphatase labelled anti-mouse IgG (H and L) (Promega, No. S372B) for 2 h. The alkaline phosphatase substrate contained 0.0134 mg/ml phenazine methosulfate, 0.16 mg/ml nitro blue tetrazolium and 0.33 mg/ml 5 bromo-4-chloro-3-indolyl phosphate and the colour was developed for 5-10 min. The reaction was stopped by rinsing in distilled water and photographs were taken of the blot within 2 h.
344
R. D. Warner, R. G. Kauffman, M. L. Greaser
Densitometry Gels were placed between two cellulose sheets and dried on a gel dryer for > 48 h prior to subjecting the gels to densitometry. A scanning densitometer (BIORAD G6-670, Cambridge, MA, USA) was used to scan the gels and the program ‘Molecular Analyst’ (BIORAD, Cambridge, MA, USA) attached to an IBM computer, was used to analyse the image and the concentration of phosphorylase in each protein band. Results were expressed as ng of phosphorylase bound to myofibrils (per 5 hg of total protein applied to the gel). Miscellaneous laboratory procedures Protein concentrations were determined by the Biuret method (Gornall et al., 1949) using bovine serum albumin as a standard. statistics
Data were analysed by ANOVA using the General Linear Model (GLM) of SAS (1985) to evaluate the differences between quality classes. Because of unequal numbers of observations among quality classes, individual mean separation was achieved using the least squares mean test.
RESULTS Physical measurements on samples Physical measurements of pH,, surface lightness, drip loss and fluid exudate were used to allocate samples to either PSE, RSE, RFN or DFD quality class and thus, by design, the mean values of all traits (except L* between RSE and RFN) were significantly different (p < 0.05) among quality classes (Table 1). PSE samples had the lowest pH,, lightest surface (L‘), highest drip and most exudate on filter paper (p < 0.05 for all). The samples selected to be RSE were similar in surface lightness (L’) to RFN samples and were, on average, 8.2 units darker than PSE samples (p < 0.05). In comparison with RFN samples, RSE samples lost 3.8% more weight during suspension and expressed 37 mg more surface exudate (p < 0.05 for both). PSE and RSE samples lost similar (p > 0.05) amounts of fluid during thawing and cooking (Table 2) and DFD lost less water (p < 0.05) then all other samples. DFD myofibrils had shorter (p < 0.05) sarcomeres TABLE 1 Meat Quality Measurements Quality clad
N
PSE RSE RFN DFD
26 19 19 20
on Loin Samples Derived From the Four Quality Classes’ Lightness (L*)
PK 5.30 5.44 5.59 6.29
f f f f
0.04” O&lb 0.04’ O&Id
55.5 47.3 45.5 38.3
f f f *
0.6” 0.76 0.76 0.7’
Drip loss (%) 9.6 7.2 3.4 1.3
f f f f
0.3” 0.4b 0.4c 0.4d
Surface exudate (mg of.BW 140 100 63 23
f f f f
5” 6b 6c 6d
‘Least squares means f SE. *PSE: pale soft, exudative; RSE: reddish-pink, soft, exudative; RFN: reddish-pink, firm, nonexudative; DFD: dark, firm, dry. pd Within columns, means with different superscripts are significantly different (p < 0.01).
Muscle proteins and pork quality
345
then RFN, RSE or PSE myofibrils (1.67, 1.79, 1.77 and 1.84 pm, respectively; SE = 0.03). Thus our data did not support previous reports that sarcomere length explains differences in drip loss between samples (Honikel et al., 1986). Protein solubility
Differences between the quality classes in protein solubility for each extraction buffer are shown in Fig. 1. PSE samples exhibited lower (p < 0.05) protein solubility for all three protein fractions (sarcoplasmic, myofibrillar and total protein fractions) compared with RSE, RFN and DFD, although the relative differences were smaller among classes for sarcoplasmic protein solubility. RSE and RFN samples had similar protein solubilities for all three extraction methods and RFN and DFD samples were also similar. RSE samples had lower (p < 0.05) protein solubility for all three methods when compared with DFD samples. Van Laack et al. (1996) reported that RSE samples were intermediate between TABLE 2
Water Loss Measurements Quality class
N (X)
18 13 15 16
PSE RSE RFN DFD
During Thawing and Cooking on Loin Samples Derived From Four Quality Classes’ Thawed losti (%)
10.9 10.3 8.2 3.1
f f f f
0.6” 0.7” 0.7b 0.6’
Cooked 10s~ (% )
29.8 27.4 25.3 16.1
f zt f f
Total fluid lo&
0.9” l.Oub l.Ob 0.9’
37.5 35.0 31.5 18.7
f f i f
0.8” 0.96 0.8’ O.gd
‘Least squares mean f SE. *Thawed loss: difference between pre-thawed weight and precooked weight expressed as a percent of pre-thaw weight. 3Cooked loss: difference between pre- and post-cooked weight expressed as a percent of pre-cook weight. 4Total fluid loss: difference between pre-thawed weight and post-cooked weight expressed as a percent of pre-thawed weight. “within columns, means with different superscripts are significantly different (p < 0.05). c
3 200 r ;160 = $ 120 g .E 80 Q) 3 h 40
sarcoplasmic Fig. 1. Comparison
myofibrillar
total
among the four quality classes in sarcoplasmic, myofibrillar and total protein solubility. Different superscripts within a protein group denote a difference (p < 0.05) between least squares means (vertical bar represents SE).
346
R. D. Warner, R. G. Kauffman, M. L. Greaser
Fig. 2. Comparison among the four quality classes in myofibrillar ATPase activity. Different superscripts denote a difference (p < 0.05) between least squares means (vertical bar represents SE).
12345678
Fig. 3. SDS polyacrylamide gel patterns of porcine iongissimus thoracis myofibrik. Quality classes: lanes 1 and 2, PSE; lanes 3 and 4, RSE; lanes 5 and 6, RFN, lanes 7 and 8, DFD. Protein bands identified are: T, titin; N, nebulin; M, myosin heavy chain; era, alpha-actinin; P, phosphorylase. Sample lanes were loaded with 12 pg of total protein and the gel was 8%. acrylamide (w/v), 0.046% bis-acrylamide (w/v), pH 8.6.
Muscle proteins and pork quality
341
PSE and RFN samples in protein denaturation as measured by the protein transmission test, which essentially measures sarcoplasmic protein denaturation. However, they did note that transmission values are a relatively insensitive measure of protein denaturation. The results obtained for sarcoplasmic, myofibrillar and total protein solubility for PSE, normal (RFN) and DFD pork are similar to those obtained by Lopez-Bote et al. (1989) and are higher then those obtained by Sayre & Briskey (1963). Myofibrillar ATPase activity
The calcium-activated myofibrillar ATPase activity was lower (p < 0.05) for PSE samples, but the activity was similar among RSE, RFN and DFD samples (Fig. 2). A
B
T_ N-
1
2
3
4
1
2
3
4
Fig. 4. Electrophoresis and immunoblots of porcine longissimus thoracis myofibrils. Panel A: Coomassie-stained SDS-polyacrylamide gel; panel B: western blot from an identical gel reacted with anti-nebulin monoclonal antibody. Quality classes: lane 1, DFD; lane 2, RFN; lane 3; RSE; lane 4, PSE. Protein bands identified are: T, titin; N, nebulin; M, myosin heavy chain; crA, alpha-actinin; P, phosphorylase. Sample lanes were loaded with 12 pg of total protein and the gel was 8%, acrylamide (w/v), 0.046% bis-acrylamide (w/v), pH 8.6.
348
R. D. Warner, R. G. KauJinan, M. L. Greaser
The results for calcium-activated myofibrillar ATPase activity are comparable with those obtained by Greaser et al. (1969) for PSE and normal muscle. Bound phosphorylase SDS-PAGE gels of the myofibril samples distinctly show protein bands for PSE and RSE samples which can be identified as the two sarcoplasmic proteins creatine kinase (just below actin), and phosphorylase (just below alpha-actinin) from standards for creatine kinase and phosphorylase run with each gel. The bands for creatine kinase and phosphorylase were clearly visible in PSE samples, fainter in RSE samples, and generally absent in RFN and DFD samples. Only the phosphorylase band was quantitated since the resolution of creatine kinase from the adjacent actin band was often incomplete. Figures 3 and 4 show the phosphorylase band clearly, but in these gels the creatine kinase band is not visible as the conditions of the resolving gel were designed to optimise titin and nebulin resolution. PSE samples had the highest (p < 0.05) amount of bound phosphorylase compared with the other three quality classes (Fig. 5). RSE samples had more (p < 0.05) bound phosphorylase than RFN and DFD samples. DFD and RFN samples had a similar (p > 0.05) and low amount of bound phosphorylase. Protein degradation Protein degradation was assessed in myofibril samples isolated at 24 h post mortem. Gels were run for all samples and representative gels are presented in Figs 3 and 4, which shows typical protein band patterns for myofibril samples of the four quality classes. For the gel shown in Fig. 4, an immunoblot utilising a monoclonal antibody to nebulin is also shown. PSE and RSE samples show evidence of breakdown of nebulin, as indicated by a fainter nebulin band and more protein bands immediately below the nebulin band, in both the gel and the Western blot. In contrast, RFN and DFD samples show less evidence of nebulin
Fig. 5. Comparison among the four quality classes in bound phosphorylase. Different superscripts denote a difference 0, < 0.05) between least squares means (vertical bar represents SE).
349
Muscle proteins and pork quality
breakdown as the nebulin band appears as a singlet, with strong staining intensity in DFD samples and as a doublet in RFN samples. There is also some evidence for more titin breakdown in RFN and DFD samples as Figs 3 and 4 show a singlet for titin in PSE and RSE samples and a doublet for RFN and DFD samples. Protein extraction
Myofibril samples subjected to extraction with 1.l M KI/O.l M potassium phosphate show different protein extractabilities among the PSE, RSE and RFN samples (Fig. 6). PSE samples show a high concentration of myosin and of the sarcoplasmic protein phosphorylase in the unextracted pellet. The supernatant contains a high concentration of alpha-actinin, actin and an unknown band with an approximate weight of 60 kD. Although this weight corresponds to that of BSA, BSA can be discounted as it was not added to or present in the sample. In contrast, RSE and RFN samples show the same RFN
RSE
PSE h4f
P
S
Mf
P
S
Mf
P
S
2
3
4
5
6
7
8
9
10
200 166 97 62
45
31
21 14
1
Fig. 6. SDS polyacrylamide
gel patterns showing the differences in protein bands between samples of PSE, RSE and RFN myofibrils after extraction with 1.1 M KI/O. 1 M potassium phosphate (pH 7.2). Lane 1 is molecular weight standards and respective molecular weights are indicated (kilodaltons). Lanes 24 are PSE samples, lanes 5-7 are RSE samples and lanes 8-10 are RFN samples. The lanes are marked as: Mf, myofibril sample; P, pellet; S, supernatant. Protein bands identified are; T, titin; N, nebulin; M, myosin heavy chain; oa, alpha-actinin; P, phosphorylase; A, actin; TT, troponin-T; TM, tropomyosin. Sample lanes were loaded with 12 pg of total protein and the gel was 12%, acrylamide (w/v), 0.06% bis-acrylamide (w/v), pH 9.3.
350
R. D. Warner, R. G. Kauffinan, M. L. Greaser
ratio of protein bands in the supernatant remain predominantly in the pellet.
and pellet, except for titin which appears to
DISCUSSION RSE is a quality class which has only recently been described (Kauffman et al., 1992), although the existence of pork which is acceptable in colour but has high drip loss has previously been reported (Warriss & Brown, 1987). Data on the protein solubility, protein degradation and specific denaturation characteristics of this class of pork were not previously available. Our study shows that RSE samples had similar protein solubility and myosin denaturation characteristics, but exhibited increased binding of the sarcoplasmic protein phosphorylase compared with RFN samples. Binding of phosphorylase to the myofibril is due to a critical combination of low pH and high temperature in pre-rigor musculature as described by Yamamoto et al. (1979). For RSE and for PSE samples, the high molecular weight protein titin was less degraded in RSE samples, and nebulin was more degraded compared with RFN samples. Reduced degradation of titin in PSE samples has previously been reported by Boles et al. (1992), but differences in nebulin breakdown between quality classes of pork have not been described before. Geesink (1993) reported decreased degradation of titin and increased degradation of nebulin post mortem in beef muscle with a low pH,, compared with beef muscle of high pH,. There was a continuum in surface lightness and drip loss from low to high values, and the samples represented a wide range in quality as already observed in industry by Kauffman et al. (1992). It is interesting to note that, although PSE and RSE samples were different in drip loss and in exudate on filter paper, there was no difference in cook loss or thaw loss and only a 3.5% difference in total water loss. As shown before, measurements of protein denaturation (protein solubility and ATPase activity) differentiated PSE samples from samples of normal (RFN) and DFD quality (Greaser et al., 1969; Sayre & Briskey, 1963; Wismer-Pederson, 1959; Yamamoto et al., 1979). These same measurements showed minimal protein denaturation in RSE samples. PSE samples exhibited a high level of myosin denaturation, supporting the hypothesis that the excessive drip loss of PSE pork is caused by myosin denaturation (Offer, 1991). The differences we found between PSE and RSE samples in myosin denaturation, protein solubility and bound phosphorylase are most likely to be a consequence of the denaturation of phosphorylase and creatine kinase at a higher pH then the denaturation of other sarcoplasmic and myofibrillar proteins such as myosin. Carcasses that have a normal rate of post-mortem glycolysis due to unusually high initial levels of glycogen can produce pork with a low ultimate pH and a high drip loss (Bendall & Swatland, 1988) and this is possibly the cause of RSE pork. RSE pork may be a consequence of low pH,, as reported by Monin & Sellier (1985) for the ‘Hampshire’ breed, although the studies included in this paper did not address this. The ‘Hampshire effect’ has since been described as the RN-gene (Le Roy et al., 1990) and has been identified as being caused by high glycogen content in the muscle. We propose that the high drip loss observed in RSE pork is caused by the lower pH,, with associated deposition of phosphorylase and creatine kinase on to the myofibril, but no major changes in protein solubility or myosin denaturation. The only protein denaturation measurement that differentiated between RSE and RFN samples was bound phosphorylase. Monin & Laborde (1985) suggested that precipitation of sarcoplasmic proteins may cause the increased drip in PSE pork. Our results show that, for RSE samples, phosphorylase binding in the myofibrillar fraction is associated with samples which have an unacceptably high drip loss but acceptable surface lightness. It is
Muscle proteins and pork quality
351
not clear whether precipitation of phosphorylase on to the myofibril is a causative factor in the increased drip loss or purely an indicator of pre-rigor conditions causing some denaturation of the most sensitive sarcoplasmic proteins. In PSE muscles, the reduced extractability of myosin, and the association of phosphorylase and creatine kinase with the unextractable protein in the pellet, suggests that the binding of the two sarcoplasmic proteins is irreversible and potentially inhibits the extraction of myosin. The average mole ratio of bound phosphorylase to myosin in PSE samples is calculated to be 1 : 20, and in RSE samples is 1 : 30. It is possible that this is high enough to be implicated in the reduced extractability of myosin, possibly through preventing the process of myosin tail unravelling. However, it is unclear from this study whether this amount could prevent myosin extraction or whether actual myosin denaturation is more important in reducing protein extractability. The rate of proteolysis by calpain in PSE muscle is severely limited (Dransfield, 1994), which may explain the toughness of PSE samples and the observed lack of ageing in PSE pigmeat (see Buchter & Zeuthen, 1971; as cited by Dransfield, 1994). This would agree with our studies which show reduced degradation of titin in PSE and RSE samples. The degree of integrity of structural constraints within the myofibril is thought to affect significantly both WHC and protein functionality (Offer & Trinick, 1983). In summary, RSE samples were intermediate in drip and pH, between RFN and PSE and similar in colour to RFN. RSE samples were not different in myosin denaturation, extractability of myofibrillar proteins or extractability of sarcoplasmic proteins compared with RFN samples. RSE had a higher proportion of the sarcoplasmic protein phosphorylase remaining with the myofibril fraction after extensive homogenising, washing and centrifugation. This implies that pre-rigor conditions in RSE muscle caused precipitation of the sarcoplasmic proteins which are most sensitive to pH/temperature conditions existing immediately post mortem. Nevertheless, the pre-rigor conditions in RSE samples did not cause extensive denaturation of myofibrillar or sarcoplasmic proteins. RSE pork has unacceptably high drip losses during storage and cooking. The only practical measurement that differentiated RSE from RFN was pH,, but the differences were too small to be used reliably under commercial conditions. Techniques are available to differentiate PSE and DFD quality pork from RFN but further research is needed to develop methods to identify pork of RSE quality at the slaughter plant.
ACKNOWLEDGEMENTS The authors are appreciative of funding provided by the College of Agriculture and Life Sciences, National Pork Producers Council, the Victorian Department of Agriculture in Australia and the Australian Pig Research and Development Corporation. The assistance provided by Mia Hospel, Nina Nusbaum, Daria Jerome, Leslie Braun and Scott Rasch is gratefully acknowledged. The advice of Erik Nordheim, Riette Van Laack, Jeff Fritz and Dar1 Swartz is also appreciated. This is Muscle Biology manuscript number 345.
REFERENCES Bendall, J. R. & Wismer-Pederson, J. (1962). J. Food Sci., 27, 144. Bendall, J. R. & Swatland, H. J. (1988). Meat Sci., 24, 85. Boles, J. A., Parrish, F. C., Jr., Huiatt, T. W. & Robson, R. M. (1992). J. Animal Sci., 70,454. Carter, S. G. & Karl, D. W. (1982). J. Biochem. Biophys. Methods, 7, 7. Dransfield, E. (1994). Meat Sci., 28, 391.
352
R. D. Warner, R. G. Kaufman, M. L. Greaser
Fischer, C., Hamm, R. & Honikel, K. 0. (1979). Meat Sci., 3, 11. Fritz, J. D., Swartz, D. R. & Greaser, M. L. (1989). Anal. Biochem., 180, 205. Fritz, J. D. & Greaser, M. L. (1991). J. Food Sci., 56, 607. Fritz, J. D., Mitchell, M. C., Marsh, B. B. & Greaser, M. L. (1993). Meat Sci., 33, 41. Geesink, G. H. (1993). PhD thesis, University of Utrecht, The Netherlands. Gomall, A. G., Bardawill, C. J. & David, M. M. (1949). J. Biol. Chem., 177, 751. Greaser, M. L., Cassens, R. G., Briskey, E. J. & Hoekstra, W. G. (1969). J. Food Sci., 34, 120. Honikel, K. O., Kim, C. J., Hamm, R. & Roncales, P. (1986). Meat Sci., 16, 267. Irving, T. C., Swatland, H. J. & Millman, B. M. (1989). J. Animal Sci., 67, 152. Kang, C. G., Muguruma, T., Fukazawa, T. & Ito, T. (1978). J. Food Sci., 43, 508. Kauffman, R. G., Eikelenboom, G., van der Waal, P. G., Merkus, G. & Zaar, M. (1986). Meat Sci, 18, 191. Kauffman, R. G., Cassens, R. G., Scherer, A. & Meeker, D. L. (1992). Variations in Pork Quality. National Pork Producers Council Publication, Des Moines, IA, USA. Koohmaraie, M. (1992). Proc. 45th Recipr. Meat Conf., Fort Collins, CO, USA p. 63. Le Roy, P., Naveau, J., Elsen, J. M. & Sellier, P. (1990). Genet. Res., 55, 33. Lopez-Bote, C., Warriss, P. D. & Brown, S. N. (1989). Meat Sci., 26, 167. Monin, G. & Laborde, D. (1985). Science des Aliments, 5, 341. Monin, G. & Sellier, P. (1985). Meat Sci., 13, 49. Offer, G. (1991). Meat Sci., 30, 157. Offer, G. & Knight, P. (1988). In Developments in Meat Science--& ed. R. A. Lawrie, Elsevier Science, London, p. 173. Offer, G. & Trinick, J. (1983). Meat Sci., 8, 245. Penny, I. F. (1967). Biochem. J., 104, 609. SAS (1985). In SAS Users Guide: Statistics, 5th edn, SAS Institute Inc, Cary, NC, USA. Sayre, R. N. & Briskey, E. J. (1963). J. Food Sci., 28, 675. Swartz, D. R., Greaser, M. L. & Marsh, B. B. (1993). Meat Sci., 33, 157. Van Laack, R. L. J. M., Kauffman, R. G., Sybesma, W., Smulders, F. J. M., Eikelenboom, G. & Pinheiro, J. C. (1996). J. Muscle Foods, 7, 149. Warner, R. D. (1994). PhD thesis, University of Wisconsin-Madison, USA. Warner, R. D., Kauffman, R. G. & Russell, R. L. (1993). Meat Sci., 33, 359. Warriss, P. D. & Brown, S. N. (1987). Meat Sci., 20,65. Wismer-Pederson, J. (1959). Food Res., 24, 711. Yamamoto, K., Samejima, K. & Yasui, T. (1979). J. Food Sci., 44, 51.
SO309-1740(96)00116-7
Meat Science, Vol. 45, No. 3, 33%352, 1997 0 1997 ElsevierScienceLtd All rights rese&d. Printed in Great B&in 0309-1740/97 $17.00+0.00
ELSEVIER
Muscle Protein Changes Post Mortem Quality Traits
Relation
to
R. D. Warner,* R. G. Kauffman & M. L. Greaser Muscle Biology Laboratory,
University of Wisconsin, Madison, Wisconsin 53706, USA
(Received 15 April 1996; revised version received 21 September 1996; accepted 24 September 1996)
ABSTRACT The relationship between post-mortem traits of muscle proteins and water loss traits was investigated using 84 pork loins representing the four quality traits of PSE, RSE (reddishpink, soft, exudative). RFN (reddish-pink,$rm, non-exudative) and DFD. Protein solubility measurements (sarcoplasmic, myojibrillar and total) were lower and myosin denaturation (quantified by myofbrillar ATPase activity) was higher for PSE samples compared with samples from the other quality classes. RSE samples were similar to RFN samples in protein solubility and myosin denaturation, although RSE had lower values then DFD samples for protein solubility measurements. RFN samples had lower drip, thaw, cook and total water loss than RSE samples and all water loss traits were lowest for DFD samples and highest for PSE samples. Insoluble phosphorylase was the only characteristic that d@erentiated among PSE, RSE and RFN samples. SDS-PAGE and Western blots indicated that in PSE and RSE samples, the myojibrillar protein titin was less degraded and nebulin was more degraded compared with RFN and DFD samples. SDS-PAGE of extracted and unextracted myojibrils showed that the reduced myojibrillar solubility of PSE samples was caused by decreased extractability of the myosin heavy chain in these samples. In conclusion, although RSE samples have unacceptably high water loss, muscle protein denaturation was minimal and did not explain the low water-holding capacity. 0 1997 Elsevier Science Ltd. All rights reserved
INTRODUCTION It is well known that pork which is described as pale, soft and exudative (PSE), has high drip loss and a pale unstable colour and denaturation of muscle proteins as indicated by low solubility of sarcoplasmic and myofibrillar proteins (Wismer-Pederson, 1959; Bendall & Wismer-Pederson, 1962; Sayre & Briskey, 1963). Specifically, myosin (Penny, 1967) and the sarcoplasmic proteins phosphorylase and creatine kinase (Fischer et al., 1979) are denatured in PSE pork. *To whom correspondence should be addressed at: Victorian Agriculture Victoria, Private Bag 7, Werribee, Vie. 3030, Australia. 339
Institute
of Animal
Science,
340
R. D. Warner, R. G. Kauflman, M. L. Greaser
The water-holding capacity (WHC) of pork has been reported to be influenced by a number of factors, including ultimate pH (pH,), protein denaturation, intra- and interfascicular spacing and sarcomere length (Offer & Knight, 1988). Myosin denaturation has been suggested to be the cause of the high rate of drip loss in PSE pork (Offer, 1991). Denaturation of myosin results in shrinkage of the myosin head, drawing the thick and thin filaments closely together. This shrinkage, in addition to the shrinkage of the myofilaments due to the low ultimate pH in PSE pork, results in more fluid being expelled between tibres and fibre bundles (Offer & Knight, 1988; Irving et al., 1989). Offer (1991) assumes that myosin head shrinkage and loss of myosin catalytic activity are identical and decisive events in determining drip loss from pork muscle. However, these events may have a different dependence on pH and temperature in the immediate post-mortem period. Recently, a quality of pork which is acceptable in colour but has excessive exudation has been described and given the name RSE-reddish-pink, soft, exudative (Kauffman et al., 1992; Warner et al., 1993; Van Laack et al., 1996), but the biochemical and physiochemical traits of pork of this quality have not been described. Degradation of structural proteins post mortem is potentially important in determining water-holding capacity (WHC) and the swelling of the myofilament lattice during extraction and solubilisation of myofibrillar proteins (Offer & Trinick, 1983). The degree of myofibrillar fragmentation is lower in PSE muscle (Kang et al., 1978) and titin has been reported to be less degraded in PSE muscle (Boles et al., 1992). The importance of the rate and extent of proteolysis during the pre-rigor period, as well as during the post-rigor period, has only recently been recognised (Koohmaraie, 1992). Titin and nebulin, which are major components of the myofibrillar structure, undergo proteolysis post mortem. The degree of their integrity post mortem may affect WHC capacity (Offer & T&rick, 1983). To understand the basic mechanisms of drip loss from pork, the contribution of protein denaturation to drip loss needs to be established. In the present study, we determined muscle protein denaturation, degradation and extractability in pork samples in relation to pork quality traits. MATERIALS
AND METHODS
Samples
A total of 84 boneless pork loins were selected at 24 h post mortem from a commercial pork plant on fourteen different occasions. Initial selection was based on a visual assessment of colour and exudate, followed by measurement of pH, using an Omega pH50 portable meter and an Orion spear-tipped glass electrode. Expressible exudate from a freshly cut surface was determined by using the filter paper fluid uptake procedure (surface exudate, mg of fluid; Kauffman et al., 1986). The loins were transported in vacuumsealed bags for 2 h and measurements of physical, biochemical and water loss traits were conducted in the laboratory. A IO-cm thick sample cranial to the 10th costae was used for measurements of protein denaturation. Ten centimetres of loin caudal to the 10th costae were used for measurements of meat quality. The remaining loin, caudal to the 10th costae was stored at -20°C for a maximum of 2 months prior to determination of post-freezing water loss (thaw loss) and cooking loss. Due to logistics, not all measurements were conducted on all samples and the number of samples used for each assay is indicated in the Results. Physical measurements
of meat quality
Meat pH, was measured using equipment described above and surface colour (CIE--L’ a’ b’) was measured in triplicate on a freshly cut surface after a 15 min bloom time at 4°C
Muscle proteins andpork quality
341
using a Minolta chromameter 200b (0’ viewing angle, diffuse illumination and 8 mm optical port, standardisation using four calibration plates, Nos 11133101, 11133853, 13033041, 16332111, set to ‘auto-select’). Drip loss was measured on a size-standard&d sample (3 cm thick and 4.1 cm in diameter) and expressed as the weight loss over 48 h at 4°C during suspension of a sample (about 50 g) (Honikel et al., 1986). These measurements were used to confirm the assignment of samples to one of four quality classes as defined below: Pale, soft, exudative (PSE) Reddish-pink, soft, exudative (RSE) Reddish-pink, firm, non-exudative Dark, firm, dry (DFD)
: L’ > 50, drip > 5%, pH, < 6.0; : L’ = 42-50, drip > 5%, pH, < 6.0;
(RFN) : L* = 42-50, drip < 5% and pH, < 6.0 and : L’ < 42, drip < 5% and pH, 2 6.0.
Using these parameters, ten samples were re-classified (once actual percent drip was known) to new classes within the RFN, RSE and PSE classes. There were no changes in classification of DFD samples. For water losses during thawing and cooking, four samples, representing each quality class, were removed from the freezer, cut into 2.5~cm thick chops and trimmed of external fat (but not epimysium), while still frozen, weighed (pre-thaw weight), thawed at 4°C for about 24 h and re-weighed (pre-cook weight). Samples were broiled in an oven to an internal temperature of 68°C and weighed (post-cook weight). Thaw loss was determined as the difference between pre-thawed weight and precooked weight and expressed as a percent of pre-thaw weight. Cooked loss was determined as the difference between preand post-cooked weight expressed as a percent of precook weight. Total fluid loss was defined as the difference between pre-thawed weight and post-cooked weight, and expressed as a percent of pre-thawed weight. Protein solubility The solubility of the sarcoplasmic proteins was determined using duplicate 1 g samples homogenised in 10 ml of ice-cold 0.025 M potassium phosphate buffer (pH 7.2). For total protein solubility, duplicate 1 g samples were added to 20 ml of ice-cold 1.1 M KI/O. 1 M potassium phosphate (pH 7.2). The samples were minced with scissors, homogenised on ice with a Polytron on the lowest setting using 3 x 4 s bursts to minimise protein denaturation through heating, and then left overnight on ice. Samples were then centrifuged at 1500 g at 4°C for 20 min and the supernatants decanted and frozen for a maximum of 2 weeks before determination of protein concentration. Myofibrillar protein solubility was determined as the difference between total and sarcoplasmic protein solubility. Myofibril purification About 2-3 g of muscle were used to isolate myofibrils by adapting the procedure of Swartz et al. (1993). Muscle was excised within 24 h post mortem and added to 15 ml of rigor buffer (RB) (10 mM imidazole, 75 mM KCl, 2 mM EGTA, 2 mM MgCl*, 2 mM NaN3, pH 7.2), cut finely with scissors and Polytron homogenised using the method described above. After Dounce homogenisation with a B-pestle for 75 strokes, the homogenate was centrifuged for 20 min at 4000 g at 4°C. The pellet was re-suspended in 15 ml of RB and Dounce homogenised with the A-pestle for 100 strokes. The solution was then diluted by 30 ml with RB, homogenised with 15 more strokes then filtered through cheesecloth and centrifuged at 4000 g at 4°C. The pellet was re-suspended three times in RB with 0.5% Triton X-100 and once in RB, using a non-shearing agitator.
342
R. D. Warner, R. G. Kauffman, M. L. Greaser
The final suspension was prepared in 15 ml of RB with 50% glycerol and 1 mM DTT, and subsequently stored at -20°C in a non-defrosting freezer. ATPase assay
The myofibril suspension was removed from the freezer, washed free of glycerol by four dilution and centrifugation cycles (11500 g for 15 s) with RB- 1 mg/ml BSA- 1 mM DTT. After the final re-suspension in RB/BSA/DTT, the protein concentration was determined using the biuret assay. Calcium-activated myofibrillar ATPase activity was determined using a total protein concentration of 0.1 mg/ml in RB/BSA/DTT plus 2.2 mM CaC12. In each case, ATPase activity was determined at 22°C in triplicate 0.2 ml samples. The reaction was initiated using 5 ~1 of 0.2M ATP and was terminated after 5 min using 20 ~1 of ice-cold 25% trichloroacetic acid (TCA). A blank was included for each sample which had TCA added before addition of ATP. After centrifuging the samples at 11500 g for 15 s to precipitate the denatured protein, 0.1 ml of sample was used to determine inorganic phosphate production by the malachite green assay of Carter & Karl (1982). Standard curves were prepared using 0.65 mM K2HP04 (Sigma No. 661-9) and results were expressed as pmole of phosphate liberated per mg of protein per min. Sarcomere length
Myofibrils were removed from -20°C storage and re-suspended in RB plus 1 mg/ml BSA (RB-BSA), the concentration determined and 100 ~1 of 1 mg/ml myofibril suspension applied to each of two coverslips (No. 1.5) and left for 1 min. The coverslip was washed five times in RB-BSA, excess fluids carefully removed and the coverslip mounted on a glass slide with a drop of mounting medium (70% glycerol-RB, pH 8.5). The excess mounting medium was removed and the coverslip sealed with clear nail polish. The slides were stored at -20°C in a non-defrosting freezer and removed as required for microscopy measurements of sarcomere length. A Nikon inverted microscope (Model Diaphot) was used with phase-contrast and a 100 x (NA 1.4) oil immersion objective. The microscope was equipped with a cooled charge-coupled device (CCD, Thomson 7883, Photometrics Ltd, Tucson, AZ, USA). The CCD was controlled via a Macintosh IIfx (Apple Computer Inc, Santa Clara, CA, USA) equipped with a Matrox computer board and Nu 200 2.0 software (Photometrics Ltd). Images were captured using exposure times of 1 s and quantitative measurement of sarcomere length on captured images was conducted using the computer program IPlab (version 2.5; Signal Analytic Corp). The average sarcomere length over five sarcomeres within one myofibril was measured and 50 myofibrils were measured per sample. Protein extraction
For the purpose of examining extracted proteins on SDS-PAGE, 1 ml aliqots of suspended myofibrils of PSE, RSE and RFN quality classes were removed from frozen storage, re-suspended in RB, as for the ATPase assay except without BSA and subjected to extraction using 10 vol. (w/v) of 1.1 M KI/O.l M potassium phosphate (pH 7.2). The samples were left overnight on ice and then split into three samples: (i) control myofibril sample; (ii) sample for determining concentration and (iii) sample for SDS-PAGE. Samples (ii) and (iii) were centrifuged at 11500 g for 1 min and the supematant carefully removed. For the pellet in sample (ii), an equal weight of 0.5 M NaOH was added to solubilise the pellet prior to determining protein concentration. Protein concentration was
Muscle proteins and pork quality
343
determined in the myofibril, pellet/NaOH and supernatant samples. One milliltre of methanol was added to the remaining supernatant and pellet samples to remove the salt, carefully mixed, centrifuged at 11500 g for 1 min and subjected to vacuum for 10 min to remove the methanol. Finally, an appropriate mixture of SDS sample buffer and distilled water was added to each sample to obtain a protein concentration of 2 mg/ml prior to subjecting to SDS-PAGE. This procedure was repeated on different PSE, RSE, RFN samples and one representative gel is presented in the results. DFD samples were not included as previous results for the ATPase activity and protein solubility showed that RFN and DFD samples were not different from each other (Warner, 1994). SDS gel electrophoresis One millilitre of suspended myofibrils for all samples was removed from -20°C storage, re-suspended in RB as above and the protein concentration determined. Samples were subjected to SDS-PAGE for: (i) the examination of degradation of the high molecular proteins titin and nebulin; (ii) the quantitation of the phosphorylase and creatine kinase bands, using densitometry and (iii) the examination of myofibril, pellet and supematant fractions from the extraction procedures described above. Gel composition, electrophoresis running conditions and treatment of gels post-electrophoresis were essentially the same as described by Fritz et al. (1989). Gels used for densitometry were treated postelectrophoresis as described by Fritz et al. (1993). For resolution of creatine kinase and phosphorylase bands, the resolving gel conditions were 12% acrylamide (w/v), 0.06% bis-acrylamide (w/v), pH 9.3 and 5.0 pg of protein were applied per lane. Five lanes were used for standard concentrations of creatine kinase (SIGMA No. C3755, San Francisco, CA, USA) and phosphorylase (Boehringer Mannheim No. 108-561, Indianapolis, IN, USA) for each gel. The concentration range for each protein standard was 0.025-125 pg, as this was previously determined to be the appropriate range for the protein load used. For resolution of titin and nebulin, the resolving gel conditions were 8% acrylamide (w/ v), 0.04% bis-acrylamide (w/v), pH 8.6 and 12 pg of protein was applied to each lane. For resolution of proteins in the myosin extraction study, the gel conditions were the same as for resolution of phosphorylase and creatine kinase, but 12 pg of protein were applied per lane and a lane for molecular weight standards (BIORAD No. 161-0317, Cambridge, MA, USA) was included. A Hoefer SE280 Tall Mighty Small 11 cm slab unit (San Francisco, CA, USA) was used for separation of creatine kinase/phosphorylase bands and a Hoefer 250 Mighty Small electrophoresis unit (San Francisco, CA, USA) was used for the other gels. Western blot analysis Proteins were transferred from an unstained gel to 0.45 pm Immobilon-P (Millipore) membrane, using methods described in Fritz & Greaser (1991) except the transfer was run at 300 mA for 2 h and the temperature was maintained at 20°C. Once the transfer was complete, the membrane was stored and processed as described by Fritz & Greaser (1991). Blots were incubated with anti-nebulin monoclonal (Sigma Chemical Company, NB2, diluted 1 : 1200) for 4 h, 1 : 7500 dilution of alkaline phosphatase labelled anti-mouse IgG (H and L) (Promega, No. S372B) for 2 h. The alkaline phosphatase substrate contained 0.0134 mg/ml phenazine methosulfate, 0.16 mg/ml nitro blue tetrazolium and 0.33 mg/ml 5 bromo-4-chloro-3-indolyl phosphate and the colour was developed for 5-10 min. The reaction was stopped by rinsing in distilled water and photographs were taken of the blot within 2 h.
344
R. D. Warner, R. G. Kauffman, M. L. Greaser
Densitometry Gels were placed between two cellulose sheets and dried on a gel dryer for > 48 h prior to subjecting the gels to densitometry. A scanning densitometer (BIORAD G6-670, Cambridge, MA, USA) was used to scan the gels and the program ‘Molecular Analyst’ (BIORAD, Cambridge, MA, USA) attached to an IBM computer, was used to analyse the image and the concentration of phosphorylase in each protein band. Results were expressed as ng of phosphorylase bound to myofibrils (per 5 hg of total protein applied to the gel). Miscellaneous laboratory procedures Protein concentrations were determined by the Biuret method (Gornall et al., 1949) using bovine serum albumin as a standard. statistics
Data were analysed by ANOVA using the General Linear Model (GLM) of SAS (1985) to evaluate the differences between quality classes. Because of unequal numbers of observations among quality classes, individual mean separation was achieved using the least squares mean test.
RESULTS Physical measurements on samples Physical measurements of pH,, surface lightness, drip loss and fluid exudate were used to allocate samples to either PSE, RSE, RFN or DFD quality class and thus, by design, the mean values of all traits (except L* between RSE and RFN) were significantly different (p < 0.05) among quality classes (Table 1). PSE samples had the lowest pH,, lightest surface (L‘), highest drip and most exudate on filter paper (p < 0.05 for all). The samples selected to be RSE were similar in surface lightness (L’) to RFN samples and were, on average, 8.2 units darker than PSE samples (p < 0.05). In comparison with RFN samples, RSE samples lost 3.8% more weight during suspension and expressed 37 mg more surface exudate (p < 0.05 for both). PSE and RSE samples lost similar (p > 0.05) amounts of fluid during thawing and cooking (Table 2) and DFD lost less water (p < 0.05) then all other samples. DFD myofibrils had shorter (p < 0.05) sarcomeres TABLE 1 Meat Quality Measurements Quality clad
N
PSE RSE RFN DFD
26 19 19 20
on Loin Samples Derived From the Four Quality Classes’ Lightness (L*)
PK 5.30 5.44 5.59 6.29
f f f f
0.04” O&lb 0.04’ O&Id
55.5 47.3 45.5 38.3
f f f *
0.6” 0.76 0.76 0.7’
Drip loss (%) 9.6 7.2 3.4 1.3
f f f f
0.3” 0.4b 0.4c 0.4d
Surface exudate (mg of.BW 140 100 63 23
f f f f
5” 6b 6c 6d
‘Least squares means f SE. *PSE: pale soft, exudative; RSE: reddish-pink, soft, exudative; RFN: reddish-pink, firm, nonexudative; DFD: dark, firm, dry. pd Within columns, means with different superscripts are significantly different (p < 0.01).
Muscle proteins and pork quality
345
then RFN, RSE or PSE myofibrils (1.67, 1.79, 1.77 and 1.84 pm, respectively; SE = 0.03). Thus our data did not support previous reports that sarcomere length explains differences in drip loss between samples (Honikel et al., 1986). Protein solubility
Differences between the quality classes in protein solubility for each extraction buffer are shown in Fig. 1. PSE samples exhibited lower (p < 0.05) protein solubility for all three protein fractions (sarcoplasmic, myofibrillar and total protein fractions) compared with RSE, RFN and DFD, although the relative differences were smaller among classes for sarcoplasmic protein solubility. RSE and RFN samples had similar protein solubilities for all three extraction methods and RFN and DFD samples were also similar. RSE samples had lower (p < 0.05) protein solubility for all three methods when compared with DFD samples. Van Laack et al. (1996) reported that RSE samples were intermediate between TABLE 2
Water Loss Measurements Quality class
N (X)
18 13 15 16
PSE RSE RFN DFD
During Thawing and Cooking on Loin Samples Derived From Four Quality Classes’ Thawed losti (%)
10.9 10.3 8.2 3.1
f f f f
0.6” 0.7” 0.7b 0.6’
Cooked 10s~ (% )
29.8 27.4 25.3 16.1
f zt f f
Total fluid lo&
0.9” l.Oub l.Ob 0.9’
37.5 35.0 31.5 18.7
f f i f
0.8” 0.96 0.8’ O.gd
‘Least squares mean f SE. *Thawed loss: difference between pre-thawed weight and precooked weight expressed as a percent of pre-thaw weight. 3Cooked loss: difference between pre- and post-cooked weight expressed as a percent of pre-cook weight. 4Total fluid loss: difference between pre-thawed weight and post-cooked weight expressed as a percent of pre-thawed weight. “within columns, means with different superscripts are significantly different (p < 0.05). c
3 200 r ;160 = $ 120 g .E 80 Q) 3 h 40
sarcoplasmic Fig. 1. Comparison
myofibrillar
total
among the four quality classes in sarcoplasmic, myofibrillar and total protein solubility. Different superscripts within a protein group denote a difference (p < 0.05) between least squares means (vertical bar represents SE).
346
R. D. Warner, R. G. Kauffman, M. L. Greaser
Fig. 2. Comparison among the four quality classes in myofibrillar ATPase activity. Different superscripts denote a difference (p < 0.05) between least squares means (vertical bar represents SE).
12345678
Fig. 3. SDS polyacrylamide gel patterns of porcine iongissimus thoracis myofibrik. Quality classes: lanes 1 and 2, PSE; lanes 3 and 4, RSE; lanes 5 and 6, RFN, lanes 7 and 8, DFD. Protein bands identified are: T, titin; N, nebulin; M, myosin heavy chain; era, alpha-actinin; P, phosphorylase. Sample lanes were loaded with 12 pg of total protein and the gel was 8%. acrylamide (w/v), 0.046% bis-acrylamide (w/v), pH 8.6.
Muscle proteins and pork quality
341
PSE and RFN samples in protein denaturation as measured by the protein transmission test, which essentially measures sarcoplasmic protein denaturation. However, they did note that transmission values are a relatively insensitive measure of protein denaturation. The results obtained for sarcoplasmic, myofibrillar and total protein solubility for PSE, normal (RFN) and DFD pork are similar to those obtained by Lopez-Bote et al. (1989) and are higher then those obtained by Sayre & Briskey (1963). Myofibrillar ATPase activity
The calcium-activated myofibrillar ATPase activity was lower (p < 0.05) for PSE samples, but the activity was similar among RSE, RFN and DFD samples (Fig. 2). A
B
T_ N-
1
2
3
4
1
2
3
4
Fig. 4. Electrophoresis and immunoblots of porcine longissimus thoracis myofibrils. Panel A: Coomassie-stained SDS-polyacrylamide gel; panel B: western blot from an identical gel reacted with anti-nebulin monoclonal antibody. Quality classes: lane 1, DFD; lane 2, RFN; lane 3; RSE; lane 4, PSE. Protein bands identified are: T, titin; N, nebulin; M, myosin heavy chain; crA, alpha-actinin; P, phosphorylase. Sample lanes were loaded with 12 pg of total protein and the gel was 8%, acrylamide (w/v), 0.046% bis-acrylamide (w/v), pH 8.6.
348
R. D. Warner, R. G. KauJinan, M. L. Greaser
The results for calcium-activated myofibrillar ATPase activity are comparable with those obtained by Greaser et al. (1969) for PSE and normal muscle. Bound phosphorylase SDS-PAGE gels of the myofibril samples distinctly show protein bands for PSE and RSE samples which can be identified as the two sarcoplasmic proteins creatine kinase (just below actin), and phosphorylase (just below alpha-actinin) from standards for creatine kinase and phosphorylase run with each gel. The bands for creatine kinase and phosphorylase were clearly visible in PSE samples, fainter in RSE samples, and generally absent in RFN and DFD samples. Only the phosphorylase band was quantitated since the resolution of creatine kinase from the adjacent actin band was often incomplete. Figures 3 and 4 show the phosphorylase band clearly, but in these gels the creatine kinase band is not visible as the conditions of the resolving gel were designed to optimise titin and nebulin resolution. PSE samples had the highest (p < 0.05) amount of bound phosphorylase compared with the other three quality classes (Fig. 5). RSE samples had more (p < 0.05) bound phosphorylase than RFN and DFD samples. DFD and RFN samples had a similar (p > 0.05) and low amount of bound phosphorylase. Protein degradation Protein degradation was assessed in myofibril samples isolated at 24 h post mortem. Gels were run for all samples and representative gels are presented in Figs 3 and 4, which shows typical protein band patterns for myofibril samples of the four quality classes. For the gel shown in Fig. 4, an immunoblot utilising a monoclonal antibody to nebulin is also shown. PSE and RSE samples show evidence of breakdown of nebulin, as indicated by a fainter nebulin band and more protein bands immediately below the nebulin band, in both the gel and the Western blot. In contrast, RFN and DFD samples show less evidence of nebulin
Fig. 5. Comparison among the four quality classes in bound phosphorylase. Different superscripts denote a difference 0, < 0.05) between least squares means (vertical bar represents SE).
349
Muscle proteins and pork quality
breakdown as the nebulin band appears as a singlet, with strong staining intensity in DFD samples and as a doublet in RFN samples. There is also some evidence for more titin breakdown in RFN and DFD samples as Figs 3 and 4 show a singlet for titin in PSE and RSE samples and a doublet for RFN and DFD samples. Protein extraction
Myofibril samples subjected to extraction with 1.l M KI/O.l M potassium phosphate show different protein extractabilities among the PSE, RSE and RFN samples (Fig. 6). PSE samples show a high concentration of myosin and of the sarcoplasmic protein phosphorylase in the unextracted pellet. The supernatant contains a high concentration of alpha-actinin, actin and an unknown band with an approximate weight of 60 kD. Although this weight corresponds to that of BSA, BSA can be discounted as it was not added to or present in the sample. In contrast, RSE and RFN samples show the same RFN
RSE
PSE h4f
P
S
Mf
P
S
Mf
P
S
2
3
4
5
6
7
8
9
10
200 166 97 62
45
31
21 14
1
Fig. 6. SDS polyacrylamide
gel patterns showing the differences in protein bands between samples of PSE, RSE and RFN myofibrils after extraction with 1.1 M KI/O. 1 M potassium phosphate (pH 7.2). Lane 1 is molecular weight standards and respective molecular weights are indicated (kilodaltons). Lanes 24 are PSE samples, lanes 5-7 are RSE samples and lanes 8-10 are RFN samples. The lanes are marked as: Mf, myofibril sample; P, pellet; S, supernatant. Protein bands identified are; T, titin; N, nebulin; M, myosin heavy chain; oa, alpha-actinin; P, phosphorylase; A, actin; TT, troponin-T; TM, tropomyosin. Sample lanes were loaded with 12 pg of total protein and the gel was 12%, acrylamide (w/v), 0.06% bis-acrylamide (w/v), pH 9.3.
350
R. D. Warner, R. G. Kauffinan, M. L. Greaser
ratio of protein bands in the supernatant remain predominantly in the pellet.
and pellet, except for titin which appears to
DISCUSSION RSE is a quality class which has only recently been described (Kauffman et al., 1992), although the existence of pork which is acceptable in colour but has high drip loss has previously been reported (Warriss & Brown, 1987). Data on the protein solubility, protein degradation and specific denaturation characteristics of this class of pork were not previously available. Our study shows that RSE samples had similar protein solubility and myosin denaturation characteristics, but exhibited increased binding of the sarcoplasmic protein phosphorylase compared with RFN samples. Binding of phosphorylase to the myofibril is due to a critical combination of low pH and high temperature in pre-rigor musculature as described by Yamamoto et al. (1979). For RSE and for PSE samples, the high molecular weight protein titin was less degraded in RSE samples, and nebulin was more degraded compared with RFN samples. Reduced degradation of titin in PSE samples has previously been reported by Boles et al. (1992), but differences in nebulin breakdown between quality classes of pork have not been described before. Geesink (1993) reported decreased degradation of titin and increased degradation of nebulin post mortem in beef muscle with a low pH,, compared with beef muscle of high pH,. There was a continuum in surface lightness and drip loss from low to high values, and the samples represented a wide range in quality as already observed in industry by Kauffman et al. (1992). It is interesting to note that, although PSE and RSE samples were different in drip loss and in exudate on filter paper, there was no difference in cook loss or thaw loss and only a 3.5% difference in total water loss. As shown before, measurements of protein denaturation (protein solubility and ATPase activity) differentiated PSE samples from samples of normal (RFN) and DFD quality (Greaser et al., 1969; Sayre & Briskey, 1963; Wismer-Pederson, 1959; Yamamoto et al., 1979). These same measurements showed minimal protein denaturation in RSE samples. PSE samples exhibited a high level of myosin denaturation, supporting the hypothesis that the excessive drip loss of PSE pork is caused by myosin denaturation (Offer, 1991). The differences we found between PSE and RSE samples in myosin denaturation, protein solubility and bound phosphorylase are most likely to be a consequence of the denaturation of phosphorylase and creatine kinase at a higher pH then the denaturation of other sarcoplasmic and myofibrillar proteins such as myosin. Carcasses that have a normal rate of post-mortem glycolysis due to unusually high initial levels of glycogen can produce pork with a low ultimate pH and a high drip loss (Bendall & Swatland, 1988) and this is possibly the cause of RSE pork. RSE pork may be a consequence of low pH,, as reported by Monin & Sellier (1985) for the ‘Hampshire’ breed, although the studies included in this paper did not address this. The ‘Hampshire effect’ has since been described as the RN-gene (Le Roy et al., 1990) and has been identified as being caused by high glycogen content in the muscle. We propose that the high drip loss observed in RSE pork is caused by the lower pH,, with associated deposition of phosphorylase and creatine kinase on to the myofibril, but no major changes in protein solubility or myosin denaturation. The only protein denaturation measurement that differentiated between RSE and RFN samples was bound phosphorylase. Monin & Laborde (1985) suggested that precipitation of sarcoplasmic proteins may cause the increased drip in PSE pork. Our results show that, for RSE samples, phosphorylase binding in the myofibrillar fraction is associated with samples which have an unacceptably high drip loss but acceptable surface lightness. It is
Muscle proteins and pork quality
351
not clear whether precipitation of phosphorylase on to the myofibril is a causative factor in the increased drip loss or purely an indicator of pre-rigor conditions causing some denaturation of the most sensitive sarcoplasmic proteins. In PSE muscles, the reduced extractability of myosin, and the association of phosphorylase and creatine kinase with the unextractable protein in the pellet, suggests that the binding of the two sarcoplasmic proteins is irreversible and potentially inhibits the extraction of myosin. The average mole ratio of bound phosphorylase to myosin in PSE samples is calculated to be 1 : 20, and in RSE samples is 1 : 30. It is possible that this is high enough to be implicated in the reduced extractability of myosin, possibly through preventing the process of myosin tail unravelling. However, it is unclear from this study whether this amount could prevent myosin extraction or whether actual myosin denaturation is more important in reducing protein extractability. The rate of proteolysis by calpain in PSE muscle is severely limited (Dransfield, 1994), which may explain the toughness of PSE samples and the observed lack of ageing in PSE pigmeat (see Buchter & Zeuthen, 1971; as cited by Dransfield, 1994). This would agree with our studies which show reduced degradation of titin in PSE and RSE samples. The degree of integrity of structural constraints within the myofibril is thought to affect significantly both WHC and protein functionality (Offer & Trinick, 1983). In summary, RSE samples were intermediate in drip and pH, between RFN and PSE and similar in colour to RFN. RSE samples were not different in myosin denaturation, extractability of myofibrillar proteins or extractability of sarcoplasmic proteins compared with RFN samples. RSE had a higher proportion of the sarcoplasmic protein phosphorylase remaining with the myofibril fraction after extensive homogenising, washing and centrifugation. This implies that pre-rigor conditions in RSE muscle caused precipitation of the sarcoplasmic proteins which are most sensitive to pH/temperature conditions existing immediately post mortem. Nevertheless, the pre-rigor conditions in RSE samples did not cause extensive denaturation of myofibrillar or sarcoplasmic proteins. RSE pork has unacceptably high drip losses during storage and cooking. The only practical measurement that differentiated RSE from RFN was pH,, but the differences were too small to be used reliably under commercial conditions. Techniques are available to differentiate PSE and DFD quality pork from RFN but further research is needed to develop methods to identify pork of RSE quality at the slaughter plant.
ACKNOWLEDGEMENTS The authors are appreciative of funding provided by the College of Agriculture and Life Sciences, National Pork Producers Council, the Victorian Department of Agriculture in Australia and the Australian Pig Research and Development Corporation. The assistance provided by Mia Hospel, Nina Nusbaum, Daria Jerome, Leslie Braun and Scott Rasch is gratefully acknowledged. The advice of Erik Nordheim, Riette Van Laack, Jeff Fritz and Dar1 Swartz is also appreciated. This is Muscle Biology manuscript number 345.
REFERENCES Bendall, J. R. & Wismer-Pederson, J. (1962). J. Food Sci., 27, 144. Bendall, J. R. & Swatland, H. J. (1988). Meat Sci., 24, 85. Boles, J. A., Parrish, F. C., Jr., Huiatt, T. W. & Robson, R. M. (1992). J. Animal Sci., 70,454. Carter, S. G. & Karl, D. W. (1982). J. Biochem. Biophys. Methods, 7, 7. Dransfield, E. (1994). Meat Sci., 28, 391.
352
R. D. Warner, R. G. Kaufman, M. L. Greaser
Fischer, C., Hamm, R. & Honikel, K. 0. (1979). Meat Sci., 3, 11. Fritz, J. D., Swartz, D. R. & Greaser, M. L. (1989). Anal. Biochem., 180, 205. Fritz, J. D. & Greaser, M. L. (1991). J. Food Sci., 56, 607. Fritz, J. D., Mitchell, M. C., Marsh, B. B. & Greaser, M. L. (1993). Meat Sci., 33, 41. Geesink, G. H. (1993). PhD thesis, University of Utrecht, The Netherlands. Gomall, A. G., Bardawill, C. J. & David, M. M. (1949). J. Biol. Chem., 177, 751. Greaser, M. L., Cassens, R. G., Briskey, E. J. & Hoekstra, W. G. (1969). J. Food Sci., 34, 120. Honikel, K. O., Kim, C. J., Hamm, R. & Roncales, P. (1986). Meat Sci., 16, 267. Irving, T. C., Swatland, H. J. & Millman, B. M. (1989). J. Animal Sci., 67, 152. Kang, C. G., Muguruma, T., Fukazawa, T. & Ito, T. (1978). J. Food Sci., 43, 508. Kauffman, R. G., Eikelenboom, G., van der Waal, P. G., Merkus, G. & Zaar, M. (1986). Meat Sci, 18, 191. Kauffman, R. G., Cassens, R. G., Scherer, A. & Meeker, D. L. (1992). Variations in Pork Quality. National Pork Producers Council Publication, Des Moines, IA, USA. Koohmaraie, M. (1992). Proc. 45th Recipr. Meat Conf., Fort Collins, CO, USA p. 63. Le Roy, P., Naveau, J., Elsen, J. M. & Sellier, P. (1990). Genet. Res., 55, 33. Lopez-Bote, C., Warriss, P. D. & Brown, S. N. (1989). Meat Sci., 26, 167. Monin, G. & Laborde, D. (1985). Science des Aliments, 5, 341. Monin, G. & Sellier, P. (1985). Meat Sci., 13, 49. Offer, G. (1991). Meat Sci., 30, 157. Offer, G. & Knight, P. (1988). In Developments in Meat Science--& ed. R. A. Lawrie, Elsevier Science, London, p. 173. Offer, G. & Trinick, J. (1983). Meat Sci., 8, 245. Penny, I. F. (1967). Biochem. J., 104, 609. SAS (1985). In SAS Users Guide: Statistics, 5th edn, SAS Institute Inc, Cary, NC, USA. Sayre, R. N. & Briskey, E. J. (1963). J. Food Sci., 28, 675. Swartz, D. R., Greaser, M. L. & Marsh, B. B. (1993). Meat Sci., 33, 157. Van Laack, R. L. J. M., Kauffman, R. G., Sybesma, W., Smulders, F. J. M., Eikelenboom, G. & Pinheiro, J. C. (1996). J. Muscle Foods, 7, 149. Warner, R. D. (1994). PhD thesis, University of Wisconsin-Madison, USA. Warner, R. D., Kauffman, R. G. & Russell, R. L. (1993). Meat Sci., 33, 359. Warriss, P. D. & Brown, S. N. (1987). Meat Sci., 20,65. Wismer-Pederson, J. (1959). Food Res., 24, 711. Yamamoto, K., Samejima, K. & Yasui, T. (1979). J. Food Sci., 44, 51.