Meat Science 66 (2004) 629637 www.elsevier.

com/locate/meatsci

Concentrations in beef and lamb of taurine, carnosine, coenzyme Q10, and creatine
R.W. Purchasa,*, S.M. Rutherfurda, P.D. Pearcea, R. Vathera, B.H.P. Wilkinsona
a

Institute of Food, Nutrition and Human Health, Massey University, Palmerston North, New Zealand Received 4 March 2003; received in revised form 4 July 2003; accepted 4 July 2003

Abstract Levels of taurine, carnosine, coenzyme Q10, and creatine were measured in beef liver and several muscles of beef and lamb and in cooked and uncooked meat. The amino acid taurine has numerous biological functions, the dipeptide carnosine is a buer as well as an antioxidant, coenzyme Q10 is also an antioxidant present within mitochondria, and creatine along with creatine phosphate is involved with energy metabolism in muscle. Large dierences were shown for all compounds between beef cheek muscle (predominantly red bres) and beef semitendinosus muscle (mainly white bres), with cheek muscle containing 9.9 times as much taurine, and 3.2 times as much coenzyme Q10, but only 65% as much creatine and 9% as much carnosine. Levels in lamb relative to beef semitendinosus muscles were higher for taurine but slightly lower for carnosine, coenzyme Q10 and creatine. Values for all the compounds varied signicantly between eight lamb muscles, possibly due in part to dierences in the proportion of muscle bre types. Slow cooking (90 min at 70  C) of lamb longissimus and semimembranosus muscles led to signicant reductions in the content of taurine, carnosine, and creatine (P < 0.001), but a slight increase in coenzyme Q10. There was also a four-fold increase in creatinine, presumably due to its formation from creatine. It is concluded that biologically, and possibly nutritionally, signicant levels of taurine, carnosine, coenzyme Q10, and creatine are present in beef and lamb, but that these levels vary between muscles, between animals, and with cooking. # 2003 Elsevier Ltd. All rights reserved.
Keywords: Beef; Lamb; Bioactives; Taurine; Carnosine; Coenzyme Q10; Creatine

1. Introduction Lean beef or lamb meat is a good source of highquality protein, of several B vitamins, and of minerals such as iron and zinc in a highly bioavailable form (Lawrie, 1991). Less information is available, however, on numerous other compounds in meat that are not generally recognized as nutrients, but that have been reported to have bioactive properties under certain conditions. A bioactive component of a diet, as dened by Lee and Ho (2002), is any food or part of a food that provides medical or health benets, including prevention and treatment of a disease. Examples of such compounds in meat include taurine, carnosine, coenzyme Q10, and creatine. Taurine (2-aminoethane sulphonic acid) is a sulphurcontaining amino acid synthesized from methionine in the liver and found in most mammalian tissues. It is present in muscle as a free acid rather than as a com* Corresponding author. Tel.: +64-6-350-4336; fax: +64-6-3505657. E-mail address: [email protected] (R.W. Purchas). 0309-1740/$ - see front matter # 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0309-1740(03)00181-5

ponent of proteins (Huxtable, 1992) and has been linked with many biological actions (34 are listed by Huxtable, 1992), but no one basic function has been identied. Taurine supplements have been shown to be benecial for infants and some groups of elderly subjects (Cho et al., 2002). It has been reported to play a protective role in exercise-induced muscle injury (Dawson, Biasetti, Messina, & Dominy, 2002), but whether supplements are benecial for normal healthy adults is unclear (Hayes & Trautwein, 1994). Levels of taurine in skeletal muscle vary between species (Hayes & Trautwein, 1994) and between muscles within a species (Cornet & Bousset, 1999). Carnosine is a dipeptide (N-b-alanyl-l -histidine) found in skeletal muscle and other mammalian tissues (Decker & Mei, 1996) where it has a buering role. It also has signicant antioxidant properties (Zhou & Decker, 1999) and is present in higher concentrations in white-type bres where lactic acid accumulation is more likely (Cornet & Bousset, 1999). In addition, carnosine reduces certain proteolytic reactions associated with cell ageing (Hipkiss, Brownson, Bertani, Ruiz, & Ferro, 2002) and, as a result, has been credited with anti-ageing

630

R.W. Purchas et al. / Meat Science 66 (2004) 629637

properties (www.oshorepharmacy.net/info-carnosine. htm). Coenzyme Q10 or ubiquinone (2,3-dimethoxy-5 methyl-6-decaprenyl-1,4-benzoquinone), which makes up part of the mitochondrial electron transport chain, also has antioxidant properties (Decker & Mei, 1996; Overvad et al., 1999). Diet supplementation with this compound has led to a range of benecial eects in some studies (Overvad et al., 1999). Meat is an important dietary source of coenzyme Q10 (Weber, Bysted, & Holmer, 1997), and it might be expected that levels in muscle will correlate with the number of mitochondria present. Creatine and its phosphorylated derivative creatine phosphate (Wyss & Kaddurah-Daouk, 2000) play an important role in muscle energy metabolism and there is extensive evidence that under some circumstances creatine supplements can enhance muscle performance (McKenna, Morton, Selig, & Snow, 1999). Muscle creatine is slowly converted nonenzymatically to creatinine by the removal of water and the formation of a ring structure. The creatinine passes into the blood and is excreted in the urine. The purposes of the current study were to determine the extent to which the concentrations of these four compounds along with creatinine varied between dierent muscles of beef and lamb and to measure the eects of cooking on these concentrations for selected muscles.

weight standard deviation=18.3 0.9 kg). As all three muscle samples were obtained from each lamb, lamb was included as a blocking factor in the analyses of variance. The eight Texel-cross lambs (one ewe and seven rams) used in Experiment 3 were aged approximately 7 months at slaughter, and were raised and nished on a pasture-only diet after being weaned at an age of approximately 13 weeks (mean carcass weight standard deviation=20.6 1.3 kg). For each lamb, samples were obtained for the biceps femoris, quadriceps femoris, gluteus medius, semimembranosus, longissimus lumborum and psoas major muscles. For the semimembranosus and longissimus muscles, uncooked and cooked (25 mm steaks cooked within plastic bags suspended in a water bath at 70  C for 90 min) samples were prepared. For all muscles the outer epimysium layer was removed and a sample of 2535 g from the thickest part of the muscle was diced and frozen before being freezedried and analysed as described below. Data analysis involved two models, with the rst comparing the ve uncooked muscles blocked within animals in a complete block design, and the second, which assessed the eects of cooking and muscle, used data from the longissimus and semimembranosus muscles only. Treatments were blocked within animals and the cooking versus muscle eects were analysed as a factorial design. 2.2. Analytical procedures

2. Methods and materials 2.1. Muscle samples For Experiment 1, ve samples of the semitendinosus muscle, heart muscle, cheek muscle (primarily the masseter muscle), and liver were obtained from prime cattle at a commercial meat plant. The history of the animals involved was not known in detail except that they were less than 2.5 years of age, of bos Taurus beef breeds or crosses, and had been nished on pasture. It is not known whether more than one tissue was sampled from any one animal. The supercial connective tissue sheath was removed from each tissue before dicing a sample in preparation for freeze-drying. Levels of taurine, carnosine, coenzyme Q10, creatine, and creatinine were determined as described below. Data were analysed as a one-way analysis of variance. For Experiment 2 samples of the longisimus lumborum, triceps brachii and semitendinosus muscles from six Romney lambs (three rams and three ewes) were prepared for analysis in the same way as described for the samples of Experiment 1. The lambs had been run together on pasture from birth, were weaned at approximately 12 weeks of age, and were slaughtered at an age of approximately 7 months (mean carcass

The assay of Coenzyme Q10 was based on that of Mattila, Lehtonen, and Kumpulainen (2000). Duplicate 10 mg samples of freeze-dried muscle were mixed thoroughly with a vortex mixer with 200 ml of 0.15 M NaCl and then remixed after adding 200 ml of ethanol. The resulting mixture was extracted as follows: 500 ml of n-hexane was added to the suspension and the tubes were mixed for 10 min on a mixer (Thermolyne Maxi-Mix IIITM, Dubuque, Iowa). The upper (hexane) layer was then removed and the extraction process was repeated twice. The combined hexane extract was dried, taken up in 200 ml of isopropyl alcohol, and the coenzyme Q10 in a 20 ml sample was separated by HPLC using a C18reverse-phase column, and quantied using absorbance at 275 nm. For taurine and carnosine analysis, duplicate 10 mg samples of freeze-dried tissue were extracted with 480 ml of 67 mM sodium citrate buer (pH 2.2) containing 0.1% phenol and 20 ml of 4 mM norleucine as an internal standard and allowed to stand for 4 h with intermittent mixing on a Vortex mixer before being centrifuged at 12,000 g for 20 min. The supernatant was transferred to ultraltration Vivaspin tubes (5000 MW cut-o) and centrifuged at 12,000 g for 30 min. The ltrate was stored in the Vivaspin tube after discarding the lter. Three ml was injected onto a cation exchange

R.W. Purchas et al. / Meat Science 66 (2004) 629637

631

HPLC column and taurine and carnosine were detected with O-phthalaldehyde (OPA). Creatine was assayed in a water extract (typically 2 ml of water and 0.5 g of freeze-dried sample) using an enzyme-based assay in a Cobas-Fara auto analyzer at 37  C. The assay was based on the diagnostic ammonia kit obtained from Sigma (171-C). In this assay, the sample was pre-incubated with the ammonia reagent (NADH, 2-oxoglutarate) to which glutamate dehydrogenase and urease had been added. Any endogenous urea was converted to ammonia and bicarbonate. Creatinase was then added to convert creatine to sarcosine and urea. The urea was then converted to ammonia at a rate that was proportional to the initial concentration of creatine. The pattern of change in optical density at 340 nm in a typical creatine assay is shown in Fig. 1. Creatine standards of 2, 4, 8 and 16 mmol l1. were used, with a typical relationship between predicted (y) and actual (x) creatine levels being: y=1.0005x0.0039 (R2=0.9992). Creatinine was assayed in the same water extract as creatine using a standard assay kit for creatinine obtained from Roche Diagnostic Ltd. The kit used the method for creatinine determination (Bartels, Jae Bohmer, & Heierli, 1972).

3. Results and discussion 3.1. Experiment 1 The results for this experiment set out in Fig. 2 and Table 1 show that there were clear dierences between

Fig. 1. Changes in optical density with time in a typical creatine assay. Glutamate dehydrogenase and urease together with NADH were added at A and after any endogenous urea and ammonia were removed, creatinase was added at B. The subsequent rate of reduction in OD340 is a measure of creatine content and is measured by regression over the range shown by solid circles. In this case the regression coecient was 0.1046 OD340 units per 100 seconds (r2=0.998).

the four tissues for all ve of the compounds measured. Results are shown as the concentrations per 100 g of fresh sample because 100 g is commonly considered as a typical serving of meat. Results for all compounds are given as absolute values in Table 1 because that enables the simple calculation of quantities being consumed when dierent quantities of the tissues are involved, but for Fig. 2 concentrations of taurine, carnosine and coenzyme Q10 are given as log10 values because this trasformation made the variances within tissues more homogeneous. Analysis as absolute values and log10 values gave similar results (Fig. 2, Table 1) but there were some small dierences in the levels of signicance of dierences between pairs of tissues. Taurine levels were lowest in the heart, similar in the semitendinosus muscle and the liver, and appreciably higher in the cheek muscle. The masseter muscle in the cheek has a high proportion of red type-I muscle bres relative to the semitendinosus muscle (Aristoy & Toldra, 1998; Cornet & Bousset, 1999) and has been reported previously to have higher levels of taurine than other muscles more suited to anaerobic metabolism. Cornet and Bousset (1999) reported that taurine levels in porcine masseter muscle were more than seven times greater than those for the corresponding longissimus muscle. The reason for this big dierence is not clear as the levels in heart muscle, which like the masseter muscle is predominantly dependant on aerobic metabolism, were the lowest of the four tissues tested and only 3.9% of those in the masseter muscle. The suggestion of associations between muscle taurine levels and levels of muscle activity may be important (Bakker & Berg, 2002; Dawson, Biasetti, Messina, & Dominy, 2002; Matsuzaki et al., 2002) as a characteristic of red-type muscles such as the masseter is the ability to operate over extended periods as a result of a high emphasis on aerobic metabolism. It is clear from these results that in order to maximize taurine intakes, the muscle that the meat came from should be taken into account. Tissue dierences in carnosine levels paralleled those for taurine in that heart muscle contained the lowest levels and levels were signicantly higher in liver than in heart muscle (Fig. 2, Table 1), but dierences between the two skeletal muscles were opposite to those shown for taurine with lower levels of carnosine in cheek muscle. These dierences are consistent with those reported elsewhere for muscles diering in the proportion of muscle bre types, with white-type muscles such as the semitendinosus having higher levels of carnosine (Aristoy & Toldra, 1998; Cornet & Bousset, 1999; Harris, Dunnett, & Greenha, 1998). This appears to be because the buering power of carnosine is more likely to be required for white-type bres where a greater dependence on anaerobic metabolism will increase the chances of lactic acid accumulation. Higher concentrations of muscle carnosine have been shown to be associated

632

R.W. Purchas et al. / Meat Science 66 (2004) 629637

Fig. 2. Concentrations of taurine, carnosine, coenzyme Q10 and creatine in heart muscle, liver, cheek muscle, and semitendinosus muscle of prime beef cattle. Individual values are shown as well as group means and standard deviation bars. Within a graph, groups with dierent letters dier signicantly (P < 0.05).

with an improved high-intensity exercise performance (Suzuki, Ito, Mukai, Takahashi, & Takamatsu, 2002). Levels of coenzyme Q10 showed more variation within the tissues than taurine or carnosine (Fig. 2, Table 1), with similar average values for heart muscle, cheek muscle, and liver. Values for the semitendinosus muscle, however, were signicantly lower as might be expected for a predominantly white muscle with lower numbers of mitochondria. In human tissues the highest concentrations have been reported in heart, liver, and kidney (Overvad et al., 1999), with the levels of 11 and 6 mg 100 g1 for heart and liver, respectively, being somewhat higher than those given in Table 1 for cattle.

Mattila et al. (2000) reported mean values of 6.3 mg 100 g1 for pork liver, and 1.6 mg 100 g1 for an unspecied beef muscle, while Weber, Bysted, and Holmer (1997) reported a concentration of 3.1 mg 100 g1 in fried beef. Creatine and its breakdown product creatinine are found in highest concentrations in skeletal and heart muscle (Wyss & Kaddurah-Daouk, 2000), so very low levels in liver relative to the other three tissues for cattle (Fig. 2, Table 1) were expected. Levels were slightly higher in heart muscle than in cheek muscle (P < 0.05) and were appreciably higher for the semitendinosus muscle. This is consistent with the greater dependence of the latter muscle on anaerobic metabolism so that the

R.W. Purchas et al. / Meat Science 66 (2004) 629637

633

need for creatine phosphate is greater than for muscles such as those of the cheek where a greater emphasis on aerobic metabolism is associated with more ecient production of ATP (Lawrie, 1991). Tissue dierences in creatinine concentrations paralleled those for creatine but at levels less than 2% of the creatine concentrations for the muscle tissues. These concentrations reect the fact that the conversion of creatine to creatinine takes place at a rate of about 1.7% day1 (Wyss & KaddurahDaouk, 2000). 3.2. Experiment 2 Concentrations of the ve compounds in lamb muscles did not dier signicantly between male and female animals, but this may reect the low number of animals involved (Table 2). There were dierences between the three muscles, however, for taurine, carnosine and coenzyme Q10, but not for creatine or creatinine (Table 2). Taurine levels were lowest in the longissimus muscle and similar in the semitendinosus and triceps brachii muscles. Relative to the triceps brachii, the

longissimus muscle contains more white type bres and a more glycolytic metabolism (Aristoy & Toldra, 1998; Briand, Talmant, Briand, Monin, & Durand, 1981), so these results are consistent with those for cattle (Table 1), but the semitendinosus muscle has also generally been shown to contain a higher proportion of white bres than the triceps brachii, so the similar levels for those two muscles does not t the expected pattern. Dierences in carnosine were more clear-cut (Table 2) with lowest levels in the reddest muscle (triceps brachii) and higher levels in the two whiter muscles (semitendinosus and longissimus). Concentrations of coenzyme Q10 were low and did not seem to parallel the expected proportion of white-type bres (and therefore mitochondria) as the highest concentration was present in the longissimus muscle and the lowest in the semitendinosus. Levels in the triceps brachii muscle did not dier signicantly from either of the other muscles but were closer to those of the longissimus muscle. The animal factor in the statistical model was signicant for carnosine and coenzyme Q10 (Table 2), which indicates that higher levels of these compounds in

Table 1 Means for the concentrations of several compounds relative to fresh weight in four tissues from beef cattle determined in Experiment 1; values are given in absolute terms here and as log10 values in Fig. 2a Tissue Cheek Number of samples Taurine (mg 100 g1) Carnosine (mg 100 g1) Coenzyme Q10 (mg 100 g1) Creatine (mg 100 g1) Creatinine (mg 100 g1) 5 382.4b 42.9a 6.79c 263b 2.34b Heart 5 22.3a 32.6a 6.05bc 298c 2.16b Liver 5 45.8a 77.5a 4.60b 16a 0.54a ST muscleb 5 38.6a 452.6b 2.18a 401d 5.82c Eectc < 0.0001 < 0.0001 0.0008 < 0.0001 < 0.0001 95, 94, 64, 98, 89, 39.4 49.2 1.47 21 0.77 R2(%), RSD

a Measures of goodness of t of the model are given by coecients of determination [R2(%)] and residual standard deviations (RSD). Means within a row dier signicantly (P < 0.05) if they do not have a common letter. b Semitendinosus muscle. c Probability that there were no real dierences between the tissues.

Table 2 Means for the concentrations of several compounds relative to fresh weight in three muscles from male and female lambs determined in Experiment 2a Sex Ram Number of samples Taurine (mg 100 g1) Carnosine (mg 100 g1) Coenzyme Q10 (mg 100 g1) Creatine (mg 100 g1) Creatinine (mg 100 g1) 3 79.1 333.5 1.21 307 4.69 Ewe 3 83.7 399.1 1.73 333 5.25 Eectc ns ns ns ns ns Muscleb LL 6 31.0a 491.1c 1.71b 346 5.90 ST 6 108.7b 356.7b 1.07a 335 4.69 TB 6 104.5b 251.1a 1.64ab 278 4.31 Eect *** *** * + ns Animal Eect R2(%), RSD

+ *** ** ns ns

97, 98, 80, 67, 66,

10.2 22.2 0.36 43 1.46

a Measures of goodness of t of the model are given by coecients of determination [R2(%)] and residual standard deviations (RSD). Means for muscles within a row do not dier signicantly (P > 0.05) if they have a common letter or if they have no letters. Interactions between sex and muscle were not signicant (P > 0.10). b LL, ST, and TB refer to the longissimus lumborum, the semitendinosus, and the triceps brachii muscles, respectively. c ns, P > 0.10; +, P < 0.10; *, P < 0.05; **, P < 0.01; ***, P < 0.001.

634

R.W. Purchas et al. / Meat Science 66 (2004) 629637

one muscle for an individual lamb tended to be associated with higher levels in the other two muscles. There was also a suggestion that this may have been the case for taurine levels, but not for creatine or creatinine. Comparisons of levels of the ve compounds between the cattle of Experiment 1 and the lambs of Experiment 2 can only be made for the semitendinosus muscle, and these need to be made with care because they were separate experiments and the cattle were appreciably older that the lambs. Relative to concentrations in the semitendinosus muscle of cattle, the same muscle from lambs had 82% more taurine, but only 79% as much carnosine, 49% as much coenzyme Q10, and 42% as much creatine. More data is required to determine whether these are consistent species dierences. 3.3. Experiment 3 Results for levels of the four compounds in six muscles of the Texel-cross lambs are shown in Table 3, and the eects of cooking on the levels for two of those muscles are shown in Table 4. Levels for the longissimus muscle, which was the only muscle common to Experiments 2 and 3, were generally similar (Tables 2 and 3) despite the fact that the lambs of Experiment 3 were older, of a dierent breed and from dierent farms. Muscle dierences were generally consistent with those shown for Experiments 1 and 2 with some exceptions. The longissimus muscle, which has a relatively high proportion of white bres, had the lowest levels of taurine and coenzyme Q10, the highest level of carnosine along with the semimembranosus muscle, and one of the higher levels of creatine and creatinine (Table 3). The quadriceps femoris muscle, in contrast, had the highest levels of taurine, a medium level of coenzyme Q10, and the lowest levels of carnosine, creatine and creatinine. For each of the compounds apart from taurine, the concentration in the muscle where it was highest was less than 1.6 times greater than in the muscle with the lowest mean concentration. For taurine the

dierence was more than 2.5-fold. The animal eect that was included in the model as a blocking factor was highly signicant for all the compounds (Table 3) indicating that there was a tendancy for a compound to be at a higher level in all the muscles if it was higher in one of them. Cooking at 70  C for 90 min had a signicant eect on all ve of the compounds measured, and this eect was similar for the two muscles evaluated (Table 4). For the three situations where the interaction between muscle and cooking was signicant (detailed in the footnote to Table 4) the ranking of the cooked and uncooked samples did not dier between the two muscles, but the size of the eect was slightly dierent from one muscle to the other. The cooking results were similar when expressed on a dry weight basis (following freeze drying) or on the basis of the uncooked weight. In calculating the concentrations on the basis of the uncooked weight, the losses during freeze drying were taken into account for the uncooked samples, and the losses during cooking as well as freeze drying were taken into account for the cooked samples. Thus the change in concentrations on an uncooked basis from before to after cooking are an indication of what is commonly referred to as retention values (Ono et al., 1984) when expressed as a percentage of the initial value as shown in Table 4. Under the cooking method used here, only about half the taurine and creatine was retained, while retention for carnosine was over 60%. The low retention values for these compounds may be due either to the compounds changing during cooking or to their loss in cooking juices, the latter explanation seeming more likely as all three compounds are water soluble. The retention of coenzyme Q10 was greater than 100% for both muscles, which reects the fact that higher levels were detected in the cooked samples, although this dierence was signicant at only the 5% level when expressed per unit of uncooked sample (Table 4). This was unexpected and suggests that either

Table 3 Means for the concentrations of several compounds relative to fresh weight in six uncooked muscles from lambs determined in Experiment 3a Muscleb LL Number of samples Taurine (mg 100 g1) Carnosine (mg 100 g1) Coenzyme Q10 (mg 100 g1) Creatine (mg 100 g1) Creatinine (mg 100 g1) 8 57.3a 458c 1.87a 489b 3.86bc BF 8 138.5c 380ab 3.05c 468a 3.67b GM 8 116.2b 430b 2.48b 472ab 4.18c PM 8 151.5cd 346a 2.45b 511c 4.00bc QF 8 160.6d 317a 2.40b 456a 3.02a SM 8 99.8b 470c 2.63bc 464a 4.08c Eectc *** *** ** *** *** Animal Eectc R2(%), RSD

*** ** *** *** ***

90, 17.6 64, 65 66, 0.48 78, 18 78, 0.35

a Measures of goodness of t of the model are given by coecients of determination [R2(%)] and residual standard deviations (RSD). Means for muscles within a row do not dier signicantly (P > 0.05) if they have a common letter or if they have no letters. b Muscle abbreviations are LL (longissimus lumborum), BF (biceps femoris), GM (gluteus medius), PM (psoas major), QF (quadriceps femoris), and SM (semimembranosus). c ns, P > 0.10; +, P < 0.10; *, P < 0.05; **, P < 0.01; ***, P < 0.001.

R.W. Purchas et al. / Meat Science 66 (2004) 629637 Table 4 Means for the concentrations of several compounds in two muscles before and after cooking at 70  C for 90 min from lambs of Experiment 3a Cooking eect Before Number of samples 8 Concentrations on a dry weight basis: 309.7 Taurine (mg 100 g1) Carnosine (mg 100 g1) 1816 8.85 Coenzyme Q10 (mg 100 g1) Creatine (mg 100 g1) 1866 Creatinine (mg 100 g1) 15.6 Concentrations on an uncooked weight basis: Taurine (mg 100 g1) 78.5 Carnosine (mg 100 g1) 464 Coenzyme Q10 (mg 100 g1) 2.25 476 Creatine (mg 100 g1) Creatinine (mg 100 g1) 4.0 Percentage retention (%) Taurine Carnosine Coenzyme Q10 Creatine Creatinine After 8 170.6 1279 11.65 1108 73.4 40.3 306 2.76 265 17.5 *** *** ** *** *** *** *** * *** *** Eectc Muscleb LL 8 172.2 1539 8.26 1500 44.3 44.0 393 2.09 384 11.1 54.7 75.5 126.5 57.2 478 SM 8 308 1555 12.24 1474 44.7 74.9 376 2.92 358 10.4 48.6 60.8 122.4 54.1 413 ns ns *** ns ns *** ns *** ** + * * ns + * *** + + + + *** + * * + ns * ns * ns 91, 46.2 71, 243 71, 2.51 97, 82 99, 3.3 91, 11.9 75, 63 66, 0.60 98, 20 99, 0.9 67, 9.7 87, 9.5 49, 26.5 82, 3.2 71, 45 Eect Animal Eect

635

R2(%), RSD

a Measures of goodness of t of the model are given by coecients of determination [R2(%)] and residual standard deviations (RSD). Interactions between cooking and muscle were not signicant (P > 0.10) except for taurine on a dry matter basis and taurine and creatinine on an uncooked weight basis (all P < 0.05). b LL, and SM refer to the longissimus lumborum, and the semimembranosus muscles, respectively. c ns, P > 0.10; +, P < 0.10; *, P < 0.05; **, P < 0.01; ***, P < 0.001.

it was extracted more eectively from the cooked meat than the raw meat or that any dry matter lost during cooking contained negligible amounts of coenzyme Q10. Cooking losses were 37.4% (SD=1.2) and 31.5% (SD=1.3) for the semimembranosus and longissimus muscles, respectively, and the dry matter content of the cooking juices for four samples tested was 5.55% (SD=0.05). Loss of this small amount of dry matter (about 2% of the initial weight) would aect the concentration of coenzyme Q10 by only about 2% if it contained no coenzyme Q10. Therefore this could account for only a small part of the retentions of over 120% shown in Table 4. Weber et al. (1997) reported losses of from 15 to 32% of coenzyme Q10 from pork chops on cooking but details of the cooking method were not given. It is possible that the extended cooking time of 90 min (at 70  C) made some previously inaccessible coenzyme Q10 available for the extraction procedure used. It is considered unlikely that the cooking brought about the formation of additional coenzyme Q10. The more than four-fold increases in creatinine with cooking can be attributed largely to the conversion of creatine to creatinine with heat (Macy, Naumann, & Bailey, 1970; Harris, Lowe, Warnes, & Orme, 1997) although this would only account for a small proportion of the drop in creatine shown. Macy et al. (1970) reported that heating of ground lamb at 60  C for up to an hour resulted in little change in creatine levels, but creatinine levels increased to approximately double their

initial levels. The cooking method used here was longer and slower than many methods in common use, so it is not appropriate to extrapolate from these results to other faster dry-heat cooking methods, about which more information is required. 3.4. General discussion In a typical serving of beef or lamb (100 g of meat) the quantities of the four compounds reported here are generally lower than quantities available commercially as supplements and/or than quantities recommended for certain groups of people on the basis of experimental evidence. Recommended levels of taurine intake do not appear to have been made as physiological requirements are usually met by synthesis from methionine within the tissues (Hayes & Trautwein, 1994) although supplementary taurine is commonly supplied for babies who are not breast-fed. Taurine intakes reported by Rana and Sanders (1986) averaged 58 mg day1 for omnivorous subjects, but were not detectable for vegans in the same study. This dierence in intake between omnivores and vegans was reected in much higher levels of taurine in the urine and breast milk of the omnivores but plasma levels were only 16% higher leading the authors to conclude that the absence of preformed taurine in the diet had no apparent harmful eect. Despite such ndings, taurine supplements are readily available (e.g. as

636

R.W. Purchas et al. / Meat Science 66 (2004) 629637

500 mg capsules at www.vitaminshoppe.com). The levels reported here in 100 g of meat are much less than 500 mg (the closest was 382 mg in beef cheek muscle), but are similar to the daily intake of 58 mg reported by Rana and Sanders (1986). Carnosine, like taurine, is synthesized by the human body from its constituent amino acids and is not recognized as a dietary requirement. It is also readily available as a supplement (e.g. as 500 mg capsules at www.vitaminshoppe.com) and has been reported to have a number of desirable eects as reviewed in the introduction. The large dierences between muscles reported in the current study for both taurine and carnosine may be a mixed blessing. On the one hand they enable consumers seeking higher levels of these compounds to seek out the appropriate muscles that may therefore be sold at a premium, but on the other hand beef or lamb as a whole cannot easily be promoted as having a certain level of the compound. The latter problem is exacerbated by the fact that signicant animal-to-animal dierences also exist (Tables 2 and 3). Coenzyme Q10 has been available as a dietary supplement for longer than taurine and carnosine, and although evidence of the benets of taking such supplements is not clear-cut (Overvad et al., 1999), it has been recommended for use as an adjuvant therapy with some forms of heart failure (Tran, Mitchell, Kennedy, & Giles, 2001). Overvad et al. (1999) noted that taking daily supplements of 200 mg day1 for 612 months had no adverse eects and according to Bucci (1998) daily doses are typically 60100 mg (e.g. 30, 60, and 100 mg capsules at www.vitaminshoppe.com). Daily intake levels may not be indicative of the eectiveness of coenzyme Q10, however, as available preparations have been shown to vary in bioavailability (Miles et al., 2002). The bioavailability of coenzyme Q10 in cooked meat is not known. Levels of Coenzyme Q10 in a 100 g serving of meat reported here were less than 8 mg for beef tissues and less than 4 mg for lamb muscles, making them appreciably lower than quantities being taken as supplements. Whether long-term consumption of such daily levels of coenzyme Q10 is benecial is not known. Creatine has been reported to improve short-term, intense exercise performance in some but not all studies (Bucci, 1998; Feldman, 1999; McKenna et al., 1999), with a typical recommended protocol of 5 g day1 for four days to increase muscle creatine levels and then a maintenance dose of 2 g day1. Even this maintenance level is almost four times higher than the highest level reported in 100 g of meat in the current study (511 mg for lamb psoas major muscle). Levels reported by Klont and Lambooy (1995) for pig longissimus and semimembranosus muscles were similar to this at over 500 mg 100 g1, while levels of 633 mg 100 g1 have been reported in beef top round steak (Pais, Salmon, Knize,

& Felton, 1999). Creatine levels in human vastus lateralis muscle were about 420 mg 100 g1 without creatine supplementation and over 500 mg 100 g1 following supplementation in the study of McKenna et al. (1999). Thus, normal levels of meat consumption will lead to intakes of creatine appreciably lower than those suggested for athletes, but the extent to which the longterm consumption of such lower levels are benecial is not known. For some athletes daily levels of meat consumption will be appreciably higher than 100 g. The presence of creatine in meat may also have negative aspects as it has been implicated in the formation of potentially carcinogenic heterocyclic amines on the surface of meats if the surface reaches high temperatures ( > about 200  C) during dry-heat cooking methods such as roasting and grilling (Pais et al., 1999). The principal ndings from the research reported here are that there are signicant levels of the potentially bioactive compounds taurine, carnosine, coenzyme Q10, and creatine in beef and lamb meat, but that the amounts present in a typical serving of meat (100 g) are generally lower than those available as dietary supplements, particularly if the cooking eects reported here are similar for other cooking methods. In addition, the concentrations vary appreciably between muscle types, and to a signicant extent between similar animals, at least for lambs.

Acknowledgements This research was carried out under contract to Meat New Zealand.

References
Aristoy, M.-C., & Toldra, F. (1998). Concentration of free amino acids and dipeptides in porcine skeletal muscles with dierent oxidative patterns. Meat Science, 50, 327332. Bakker, A. J., & Berg, H. M. (2002). Eect of taurine on sarcoplasmic reticulum function and force in skinned fast-twitch skeletal muscle bres of the rat. Journal of Physiology, 538, 185194. Bartels, H., Bohmer, M., & Heierli, C. (1972). Serum kreatininbestimmung ohne enteiweissen. Clinica Chimica Acta, 37, 193197. Briand, M., Talmant, A., Briand, Y., Monin, G., & Durand, R. (1981). Metabolic types of muscle in the sheep: I Myosin ATPase, glycolytic, and mitochondrial enzyme activities. European Journal of Applied Physiology, 46, 347358. Bucci, L. R. (1998). Dietary supplements as ergogenic aids. In I. Wolinsky (Ed.), Nutrition in exercise and sport (pp. 315368). Baton Rouge: CRC Press. Cho, K. H., Kim, E. S., Chen, J. D., Zhang, S., Kim, H., Kim, S. Y., Kang, S. A., Lee, Y. N., & Leem, K. (2002). Serum and urine taurine levels in elderly patients undergoing long-term enteral nutrition are reduced over time. Nutrition Research, 22, 10171025. Cornet, M., & Bousset, J. (1999). Free amino acids and dipeptides in porcine muscles: dierences between red and white muscles. Meat Science, 51, 215219. Dawson, R., Biasetti, M., Messina, S., & Dominy, J. (2002). The

R.W. Purchas et al. / Meat Science 66 (2004) 629637 cytoprotective role of taurine in exercise-induced muscle injury. Amino Acids, 22, 309324. Decker, E. A., & Mei, L. (1996). Antioxidant mechanisms and applications in muscle foods. Reciprocal Meat Conference Proceedings, 49, 6472. Feldman, E. B. (1999). Creatine: a dietary supplement and ergogenic aid. Nutrition Reviews, 57, 4550. Harris, R. C., Lowe, J. A., Warnes, K., & Orme, J. A. (1997). The concentration of creatine in meat, oal and commercial dog food. Research in Veterinary Science, 62, 5862. Harris, R. C., Dunnett, M., & Greenha, P. L. (1998). Carnosine and taurine contents in individual bres of human vastus lateralis muscle. Journal of Sports Sciences, 16, 639643. Hayes, K. C., & Trautwein, E. A. (1994). Taurine. In M. E. Shils, J. A. Olson, & M. Shike (Eds.), Modern nutrition in health and disease (pp. 477485). Malvern PA: Lea & Febiger. Hipkiss, A. R., Brownson, C., Bertani, M. F., Ruiz, E., & Ferro, A. (2002). Reaction of carnosine with aged proteins. Another protective process? Annals of the New York Academy of Science, 959, 285 294. Huxtable, R. J. (1992). Physiological actions of taurine. Physiological Reviews, 72, 101163. Klont, R. E., & Lambooy, E. (1995). Eects of preslaughter muscle exercise on muscle metabolism and meat quality studied in anaesthetized pigs of dierent halothane genotypes. Journal of Animal Science, 73, 108117. Lawrie, R. A. (1991). Meat science (5th ed). Oxford: Pergamon Press. Lee, T -C., & Ho, C -T. (2002). Bioactive compounds in foods. Eects of processing and storage. American Chemical Society, Symposium Series 816. Macy, R. L., Naumann, H. D., & Bailey, M. E. (1970). Water-soluble avor and odor precursors of meat. 5. Inuence of heating on acidextractable non-nucleotide chemical constituents of beef, lamb and pork. Journal of Food Science, 35, 8387. Matsuzaki, Y., Miyazaki, T., Miyakawa, S., Bouscarel, B., Ikegami, T., & Tanaka, N. (2002). Decreased taurine concentration in skeletal muscles after exercise for various durations. Medicine and Science in Sports and Exercise, 34, 793797. Mattila, P., Lehtonen, M., & Kumpulainen, J. (2000). Comparison of in-line connected diode array and electrochemical detectors in the

637

high-performance liquid chromatographic analysis of coenzymes Q9 and Q10 in food materials. Journal of Agricultural and Food Chemistry, 48, 12291233. McKenna, M. J., Morton, J., Selig, S. E., & Snow, R. J. (1999). Creatine supplementation increases muscle total creatine but not maximal intermittent exercise performance. Journal of Applied Physiology, 87, 22442252. Miles, M. V., Horn, P., Miles, L., Tang, P., Steele, P., & DeGrauw, T. (2002). Bioeqivalence of coenzyme Q10 from over-the-counter supplements. Nutrition Research, 22, 919929. Ono, K., Berry, B. W., Johnson, H. K., Russek, E., Parker, C. F., Cahill, V. R., & Althouse, P. G. (1984). Nutrient composition of lamb of two age groups. Journal of Food Science, 49, 12331239 & 1257. Overvad, K., Diamant, B., Holm, L., Holmer, G., Mortensen, S. A., & Stender, S. (1999). Coenzyme Q10 in health and disease. European Journal of Clinical Nutrition, 53, 764770. Pais, P., Salmon, C. P., Knize, M. G., & Felton, J. S. (1999). Formation of mutagenic/carcinogenic heterocyclic amines in dry-heated model systems, meats, and meat drippings. Journal of Agricultural and Food Chemistry, 47, 10981108. Rana, S. K., & Sanders, T. A. B. (1986). Taurine concentrations in the diet, plasma, urine and breast milk of vegans compared with omnivores. British Journal of Nutrition, 56, 1727. Suzuki, Y., Ito, O., Mukai, N., Takahashi, H., & Takamatsu, K. (2002). High level of skeletal muscle carnosine contributes to the latter half of exercise performance during 30-s maximal cycle ergometer sprinting. Japanese Journal of Physiology, 52, 199205. Tran, M. T., Mitchell, T. M., Kennedy, D. T., & Giles, J. T. (2001). Role of coenzyme Q10 in chronic heart failure, angina, and hypertension. Pharmacotherapy, 21, 797806. Weber, C., Bysted, A., & Holmer, G. (1997). The coenzyme Q10 content of the average Danish diet. International Journal of Vitamin and Nutritional Research, 67, 123129. Wyss, M., & Kaddurah-Daouk, R. (2000). Creatine and creatinine metabolism. Physiological Reviews, 80, 11071213. Zhou, S., & Decker, E. A. (1999). Ability of carnosine and other skeletal muscle components to quench unsaturated aldehydic lipid oxidation products. Journal of Agricultural and Food Chemistry, 47, 5155.