| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ANIMAL PRODUCTS |
* Institute of Food, Nutrition and Human Health, Massey University, Palmerston North, New Zealand; and
New Zealand Institute for Crop and Food Research, Food Industry Science Centre, Palmerston North, New Zealand
 |
Abstract |
|---|
 Top Abstract INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
Key Words: pork quality • flavor • odor • pork color • water-holding capacity • tenderness
|
INTRODUCTION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
; Purchas et al., 2004). However, whether or not meat products with particular strengths in these areas will be acceptable to consumers will depend on the extent to which other aspects of meat quality such as appearance and palatability are affected (Verbeke, 2006). Morel et al. (2008) reported that by dietary manipulation, pork from female Duroc-cross pigs could be produced with enhanced concentrations of CLA, selenium, and vitamin E, and that this could be achieved after all animal products in the diet had been replaced with plant products. These changes had no significant effects on the rate or efficiency with which the animals grew. The current paper reports the results of laboratory measures of physical characteristics of meat from 2 muscles of the pigs described by Morel et al. (2008), together with sensory assessments of the flavor and odor of pork from the loin muscle. Other studies that have tested the components of the supplement separately have generally not shown effects on aspects of pork quality (reviewed in the discussion section of this paper), so the hypothesis here was that the combination would not have detrimental effects either. |
MATERIALS AND METHODS |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
All procedures involving animals were approved by the Massey University Animal Ethics Committee and the New Zealand Code of Practice for the Care and Use of Animals for Scientific Purposes.
Details of the experiment in terms of the animals used, the on-farm nutritional treatments, and the slaughter procedures were reported by Morel et al. (2008). Briefly, 64 Duroc-crossbred gilts were grown from weaning at 23 d until they reached a target BW of 100 kg at a mean age of 144 d on diets that contained either animal and plant products (the animal group), or plant products only (the plant group). The gilts were from 4 sire groups, with an equal number within each sire group being allocated to each of the 4 treatment groups. Within each of these groups half the animals received a nutritional supplement (Sanovite), containing CLA, selenium, and vitamin E. The mean dietary concentrations (as-fed basis) in the control and supplemented diets, respectively, were 0.3 and 0.7 mg·kg–1 for selenium, 38.5 and 164.4 mg·kg–1 for vitamin E, and 0.22 and 1.64% of total fatty acids for CLA. Slaughter procedures followed normal commercial practice, and, after an overnight chilling period (~18 h), a 300-mm length of the caudal end of each left-side LM (including the associated bone, fat, and skin) and all of the left-side semimembranosus muscle (SM) were removed and vacuum-packaged for refrigerated shipping to the Massey University meat laboratory, where the samples were processed at approximately 48 h postmortem. A full set of meat quality and composition data was obtained for 59 animals, with 15 for 3 groups and 14 for one group (Table 1).
|
Laboratory Analyses
For color evaluation, meat samples were cut open to expose an internal surface and then, after a 30 min bloom period at room temperature (17°C), the CIE L*a*b* color space was evaluated by surface reflectance (ChromaMeter CR-200, Minolta Co., Ltd., Osaka, Japan), calibrated against a white tile using illuminant C. Three measurements of each attribute were taken across the surface of each sample and averaged. Chroma (color intensity) and hue angle (0° = red; 90° = yellow) were calculated as: Chroma = 
[(a*)2 + (b*)2] and Hue = tan–1(b*/a*) (Minolta, 1998).
For the determination of drip loss (Honikel, 1998), a cube (approximately 40 x 40 x 40 mm) was prepared from each LM and SM sample, with all 6 surfaces freshly cut and free of epimysium. Each cube was weighed and suspended on a wire hook inside a pre-inflated plastic bag. Bags were sealed with a twist tie and hung in a cooler (0 to 1°C). Cubes were reweighed after 48 and 96 h to determine drip loss as a percentage of initial weight.
To determine expressible moisture, a 500 (±20)-mg sample from each LM was weighed onto a piece of What-man No. 1 filter paper that was previously stored at room temperature over a saturated KCl solution. The sample was then compressed on the paper between 2 Plexiglas plates under a 10-kg weight for 5 min. The wetted area, including the compressed meat area, was measured using a digital planimeter, and expressible moisture was calculated as the ratio of this area to sample weight (cm2·g–1; Purchas, 1990).
From the interior of a slice of each LM and SM, 2.0 to 2.5 g was removed, scalpel-minced into several smaller pieces, and placed in a 25 x 60-mm plastic vial with 10 mL of chilled, distilled water for ultimate pH (pHu) determination after homogenization (Ultra-Turrax, Janke-Kunkel, John Morris Scientific Ltd., Auckland, New Zealand) for three 5-s bursts. The pH of the suspension was measured with a benchtop pH meter (Jenway Model No. 3020, Global Science and Technology Ltd., Auckland, New Zealand).
At approximately 48 h postmortem, one 25-mm-thick slice of each LM and SM was weighed, placed in a polythene bag, and cooked for 60 min in a water bath maintained at 70°C. After cooking, the fluid was poured from the bags and, after being held overnight at 0 to 1°C, the samples were patted dry with paper toweling and reweighed to determine cooking loss, which was expressed as a percentage of uncooked sample weight. From each cooked steak, six 13 x 13-mm samples were prepared parallel to the fiber grain and each was sheared twice using a square blade on a WBSF machine (G-R Electric Mfg., Manhattan, KS; crosshead speed of 230 mm·min–1) to give an initial yield force and a peak shear force, as described by Purchas and Aungsupakorn (1993).
Sensory Analysis
A 13-member descriptive sensory panel, each member having previous evaluation experience, was trained over the course of 12 sessions. Training sessions involved development of both a sampling procedure and an odor/flavor lexicon. With the use of reference materials, derived from Sensory Spectrum methodology (Civille and Lyon, 1996) and panel discussion, the group moved toward consensus for scoring 8 odors (tangy, pork, meat, broth, grain, milky, grass, rancid) and 11 flavors (tangy, sweet, salty, pork, meat, broth, grain, milky, grass, rancid, bitter) in pork samples.
Sensory evaluation was conducted on LM samples from 8 animals within each diet treatment group, balanced with respect to sire group. Eight evaluation sessions were conducted on 8 d, with 4 samples (one from each treatment group) evaluated per session, along with a warm-up sample at the beginning of each session that was included to remove any first-sample bias. Samples (previously aged at 0 to 2°C for 48 h, then frozen at –30°C for up to 5 mo) were thawed overnight at 4°C and then cut into small cubes before adding subcutaneous fat tissue from the same animal at 5% (wt/wt) and mincing through a plate with 8-mm holes. The mince was thoroughly mixed manually, and 15-g (±0.5 g) lots were heat-sealed into plastic bags without vacuum and cooked for 5 min in a 75°C water bath. They were then held in a 65°C water bath for at least 4 min to bring them to a serving temperature.
Panelists were seated in individual booths under red light. Samples were served in a sequential, monadic manner, and the presentation order for the 4 treatments was according to a Latin square design. Using 10-point scales (0 = "none", 10 = "strong"), the panelists were asked to assess the odor immediately after opening the sample bags and then to assess the flavor. Soda water and unsalted water crackers were used to cleanse the palate before evaluating each sample.
Statistical Evaluation
Statistical analyses were performed using the GLM procedure (SAS Inst. Inc., Cary, NC). The statistical model for all variables included a sire effect (n = 4), a treatment effect (n = 4), and a set of 3 orthogonal contrasts between treatments that included a comparison between the animal and plant groups, a comparison between the control and supplemented diets within the animal group, and a comparison between control and supplemented diets within the plant group. For laboratory measures of the meat quality characteristics, the model was fitted with and without ultimate pH as a covariate before testing for sire and treatment effects. The interaction between treatment and sire was investigated but was not retained in the models because it was seldom significant and did not affect the nature of the treatment or covariate effects. Furthermore, the implications of any significant treatment by sire interaction could not be interpreted in any useful way. Statistical analysis of the sensory data followed the same procedures after scores for individual odor and flavor notes for each animal had been derived using the REML procedure in Genstat (VSN International Ltd., Hemel Hempstead, UK, 2005). This procedure produced means for each pig that had been adjusted for panelist differences and presentation order.
Paired t-tests were used to compare characteristics between the LM and SM across all 59 animals.
|
RESULTS AND DISCUSSION |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
included level of fat cover at P2 (65 mm off the midline over the last rib), which averaged about 12 mm and was not affected by the treatments, and LM intramuscular fat levels, with an overall mean of 1.39% (SD = 0.50) and greater levels in the animal group (P < 0.036) and in the supplemented group within both the animal (P = 0.10) and plant (P = 0.032) groups. Items in the supplement were at greater concentrations in the LM of pigs receiving that supplement. Concentrations for the control and supplemented groups, respectively, were not detectable and 0.18 µg·g–1 for selenium, 2.24 and 4.40 µg·g–1 for vitamin E (P < 0.0001), and 0.30 and 0.85% of total fatty acids for CLA (P < 0.0001). Laboratory Measures of Physical Characteristics of Meat
Across all treatment groups mean LM 45-min pH ranged from 6.39 to 6.51 with no significant treatment effects (Table 1
) indicating a normal progression of postmortem muscle metabolism with no carcasses at risk of developing the pale, soft, exudative condition that is more likely with pH values in the range of 5.5 to 6.0 (Warriss and Brown, 1987; Kauffman et al., 1993; Joo et al., 2000). Mean pHu values for both the LM and SM were all less than 5.75 (Table 1), suggesting that preslaughter stress levels were low, but the SM pHu was significantly greater than that for the LM (P < 0.001). Warner et al. (1993) also reported nonsignificantly greater pHu values for SM, with the smaller difference possibly attributable to the known variation in pHu between pale and dark portions within the SM (Purchas et al., 1988). Lower pHu values for the supplemented animal group were shown for both muscles (P< 0.04, Table 1). The supplemented plant group also had nonsignificantly lower pHu values, but these differences were small and of limited biological importance. In their comparison of pig diets supplemented with 2% of CLA or sunflower oil, Dugan et al. (1999) reported a lack of effect of oil addition on postmortem LM pH, whereas Mahan et al. (1999) indicated a lack of effect of selenium supplementation on pork pHu. Furthermore, no effect on pHu was reported by Cannon et al. (1996) or Swigert et al. (2004) following vitamin E feeding to pigs for 84 and 45 d, respectively. Lahucky et al. (2007) found no effects of feeding supplementary vitamin E for 30 d on muscle pH at 1 or 24 h.
Measures of water-holding capacity, meat tenderness in terms of WBSF, and color, were all affected (P < 0.05) by pHu, so results in Tables 1 to 3 show LS means both before and after adjustment to a common pHu.
|
|
. The pHu effect on drip loss was highly significant for both muscles with R2 values more than doubling when pHu was included as a covariate, but diet and supplement effects were not significant either before or after adjustment for differences in pHu (Table 1). For SM drip loss, however, a significantly greater value for the supplemented animal group without pHu-adjustment (P < 0.006), disappeared following pHu-adjustment (P < 0.067). Whereas cooking loss was not affected in either muscle by dietary treatment either before or after pHu-adjustment (Table 1), LM expressible moisture paralleled drip loss in being greater in the supplemented group, but only for those animals receiving the animal diet (P < 0.05). Lettner et al. (2001) fed grower and finisher diets with varying amounts of meat meal versus soybean meal and observed no effect of diet on drip loss. Joo et al. (2002), likewise, reported no effect on drip loss after supplementing a commercial finisher diet with 1.0, 2.5, or 5.0% CLA. Dunshea et al. (2002) and Dugan et al. (2003) also showed no change with CLA supplementation, and Mahan et al. (1999) reported no effect of an organic selenium supplement on drip losses from pork. No effect of vitamin E feeding was reported by Swigert et al. (2004), and Cannon et al. (1996) showed no effect on either drip or cooking loss.
The highly significant pHu effects on all measures of water-holding capacity (Table 1) were reflected in significant negative correlations within both muscles across all treatments (Table 2), and are consistent with the well-known positive effect of pH on this characteristic (Lawrie, 1991). These different measures of water-holding capacity, although positively correlated (Table 2), do not appear to measure the same characteristic because drip loss was significantly greater in LM than SM (P < 0.001), but cooking losses did not differ between the 2 muscles. Honikel (1987) noted that drip loss differences were not necessarily matched by differences in cooking losses because the former is mainly a measure of free water in fresh meat, while the latter is a measure of the extent to which bound water is lost following heat denaturation of meat proteins. Warner et al. (1993) found no difference in muscle exudate between LM and SM of pork, and Maddock et al. (2002) reported greater drip losses for LM than SM, as was the case for the current trial.
Sire effects, with 2 to 4 animals from each sire group in each treatment group, were observed for several of the characteristics shown in Table 1, which suggested that those characteristics were controlled genetically to some extent. However, with only 4 sires this aspect would require further investigation before conclusions can be drawn.
There were no dietary treatment effects on WBSF of SM before or after pHu adjustment, but values for LM were greater for the plant group in terms of the initial yield as well as the peak force both before and after pHu adjustment (P < 0.04, Table 3). The supplement decreased WBSF of LM within the animal group only (P = 0.022), but this effect was not observed after pHu adjustment, suggesting that it may have been an indirect effect through pHu differences (Table 1). The pH effect on WBSF was highly significant for LM (P = 0.005) but not SM (Table 3) and inclusion of pHu as a covariate led to an appreciable increase in R2 values. Although shear force values were not reported, Lettner et al. (2001) reported that feeding meat meal versus soybean meal did not affect sensory scores for tenderness. Dugan et al. (2003) reported no effect of CLA feeding on shear force values or sensory tenderness scores in the longissimus, nor did Waylan et al. (2002) who fed modified tall oil, a novel source of CLA. Teye et al. (2006), however, found that sensory assessments of tenderness were better for pork from pigs supplemented with palm kernel oil than that from pigs receiving soyabean oil. They attributed this difference to greater levels of lauric and myristic acids. Vitamin E supplementation has also been reported to not affect pork shear values (Swigert et al., 2004; Lahucky et al., 2007).
Overall WBSF values for initial yield as well as peak force were greater for LM than SM (P < 0.001), but the difference between peak force and initial yield was greater for SM than LM (P = 0.013), which is probably a reflection of the slightly greater connective tissue content of SM (Wheeler et al., 2000). The greater shear force values for LM than SM contrasts with the results of Melody et al. (2004) for pork, and also with reports for beef (Belew et al., 2003) and lamb (Johnson et al., 2005) where shear values for LM were significantly lower than for SM. These contrasting results may be due to gradients in shear force values within the SM (Purchas et al., 1988).
Dietary treatments had a small influence on instrumental color parameters in both the LM and SM (Table 4), with pork from the animal group being slightly more red (a* values) and yellow (b* values), which led to greater chroma values (P < 0.06), this is unlikely to be of practical importance. The supplement, however, had essentially no effect on color, and this pattern was very similar after pHu adjustment. Lightness was the color parameter most affected by pHu (P < 0.001, Table 4). Wiegand et al. (2002) and Dugan et al. (2003) also reported no effects of CLA supplementation on either L* or a* values. Vitamin E supplementation was found not to affect pork color by Lahucky et al. (2007).
|
; Maddock et al., 2002). Sensory Evaluation
There were very few dietary treatment effects on evaluations by the trained sensory panel of pork odor or flavor characteristics after data had been adjusted for the effects of panellist and presentation order, and the proportion of the variation accounted for (R2 values) were generally low (Table 5). Five of the flavor and odor notes were analyzed in terms of the percentage of times that they were detected at all rather than as average scores because 63% or more of the scores for these characteristics were zero. The frequency with which these notes were given scores from 0 to 10 are shown in Table 6. The percentage of time a rancid odor was noted was greater for the supplemented plant diet than the control plant diet (25 vs. 12%; P < 0.009, Table 5), which is consistent with the greater ratio of polyunsaturated to saturated fatty acids in lipid of the LM for that group (Morel et al., 2008). Within the animal group, supplementation tended to decrease the percentage of times that a bitter flavor was detected (P < 0.04, Table 5). Teye et al. (2006) reported that quite big differences in the fatty acid composition of intramuscular lipid (e.g., a 23% increase in the proportion of polyunsaturated fatty acids) had no detectable effect on pork flavor or overall liking. In contrast to other measures of meat quality, sire effects were not apparent for the sensory attributes shown in Table 5, except for the percentage of time a bitter flavor was detected. Both tangy odor (r = –0.37; P < 0.05) and tangy flavor (r = –0.69; P < 0.0001) scores were negatively related to pHu, but none of the other sensory attributes were related to pHu, which is why it was not included as a covariate in the model.
|
|
In this study, increasing the value of pork for the consumer by dietary manipulation to increase levels of CLA, selenium, and vitamin E, and by replacing animal products in the diet with plant items, was found to have no substantial effects on laboratory measures of water-holding capacity, color, or shear-force values of pork, and minimal detrimental effects on the odor or flavor of the product. A small increase in the frequency with which rancid odors were detected in pork from pigs on a plant-based diet with dietary supplements was an outcome that may need attention, but generally the results show that including these supplements in the diets of pigs to enhance the nutritive value and health-giving properties of pork will not significantly compromise other aspects of meat quality.
|
Footnotes |
|---|
2 Corresponding author: R.Purchas{at}massey.ac.nz
Received for publication July 19, 2007. Accepted for publication February 20, 2008.
|
LITERATURE CITED |
|---|
|
TopAbstractINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONLITERATURE CITED |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
