Influences of muscle composition and structure of pork from different breeds on stability and textural properties of cooked meat emulsion

Food Chemistry 138 (2013) 1892–1901

Contents lists available at SciVerse ScienceDirect

Food Chemistry journal homepage: www.elsevier.com/locate/foodchem

Influences of muscle composition and structure of pork from different breeds on stability and textural properties of cooked meat emulsion Supaluk Sorapukdee a, Chananya Kongtasorn b, Soottawat Benjakul a, Wonnop Visessanguan c,⇑ a

Department of Food Technology, Faculty of Agro-Industry, Prince of Songkla University, 15 Kanchanawanich Road, Hat Yai, Songkhla 90112, Thailand Department of Research and Development, Betagro Hybrid International Co., Ltd., Bangkok 10210, Thailand c Food Biotechnology Research Unit, National Center for Genetic Engineering and Biotechnology (BIOTEC), 113 Phaholyothin Road, Klong Luang, Pathumthani 12120, Thailand b

a r t i c l e

i n f o

Article history: Received 26 August 2012 Received in revised form 5 October 2012 Accepted 17 October 2012 Available online 12 November 2012 Keywords: Pig Breed Meat emulsion Meat composition Muscle structure Partial least square regression

a b s t r a c t The influences of composition and structure of meats from different pig breeds, including Duroc (D), Large White (LW), Landrace (LR), two-way cross (LR  LW) and three-way cross (D  [LR  LW]) on stability and textural characteristics of cooked meat emulsions were studied by using partial least squares (PLS) regression. Compared to other pig breeds, cooked meat emulsion from LW exhibited superior properties as indicated by lower water and fat released as well as higher chewiness, gumminess, cohesiveness, resilience, springiness and hardness. The univariate analyses of those selected properties indicated a significant correlation with higher contents of myofibrillar and sarcoplasmic proteins, smaller muscle fibre diameter and lower myofibril fragmentation of LW meat, as compared to other breeds. Therefore properties of cooked pork emulsion were influenced by composition and structure of meat, which varied according to the pig breeds. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction Cooked emulsified meat products or sausages are widely consumed around the world. Basically, emulsion-type sausages are made from a mixture of finely chopped meat, fatty tissue, water, ice and additives (i.e. salt, nitrate, phosphate, seasoning, flavourings) (Hernández-Hernández & Guerrero-Legarreta, 2010; Ugalde-Benítez, 2012). Hundreds of different products are available for consumers, e.g., frankfurters, bologna and mortadella. In general, a protein matrix with high fat stability and water-holding capacity is desirable upon processing. A failure to form the gel can produce an excessive loss of water and fat, producing a mushy and mealy texture. Meat proteins serve as the emulsifying agent in a meat emulsion. To form a stable meat emulsion, these proteins must surround the finely chopped fat particles before cooking. Myosin, the major structural protein of meat, is the most important of the proteins for fat emulsification and water-holding capacity of processed meats. It is believed that myosin may bridge the oil– water interface, as the non-polar amino acid residues of the myosin tail would be attracted to the fat cell surface, whilst polar amino acid residues of the myosin head would be associated with the water phase. In addition to meat protein, fat is also an essential

⇑ Corresponding author. Tel.: +66 2 5646700x3747; fax: +66 2 5646707. E-mail address: [email protected] (W. Visessanguan). 0308-8146/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodchem.2012.10.121

component of formulated meat products, contributing to tenderness, juiciness and overall palatability (Foegeding, Clark, & Xiong, 1991). Technologically, emulsion-type sausages are mainly dependent upon the state of meat proteins and their water-binding and emulsifying properties (Ker & Toledo, 1992; Smith, 1988; Ziegler & Acton, 1984). There are a number of factors governing the quality of sausages, e.g., species of meat (beef, pork or chicken), the basic condition of the meat at the time of use, fat source, fat particle size, non-meat additives, pH of emulsions, chopping or emulsifying temperature and time, and cooking method, etc. Although pork has been often utilised for sausage production, the information regarding the effects of pig breeds on quality of cooked pork emulsion is scare. Therefore, the objective of this investigation was to study the quality of cooked pork emulsion in terms of emulsion stability and textural characteristics as influenced by the variation of muscle composition and structure among meats derived from different pig breeds using partial least squares (PLS) regression.

2. Material and methods 2.1. Chemicals Sodium tripolyphosphate (STPP) was obtained from Aditya Birla Chemicals (Samutprakarn, Thailand). Sodium nitrite, ethylenediaminetetraacetic acid (EDTA), trichloroacetic acid (TCA), sodium

S. Sorapukdee et al. / Food Chemistry 138 (2013) 1892–1901

azide, sodium hydroxide (NaOH), potassium chloride (KCl) and Nile Blue A were obtained from Sigma (St. Louis, MO). Glutaraldehyde was purchased from Fluka (Buchs, Switzerland). Ethanol and nitric acid were procured from Lab-Scan (Bangkok, Thailand).

1893

the supernatant was the alkaline-soluble fraction. The final residue was the alkaline-insoluble fraction. For each sample, the extraction was carried out in duplicate. The nitrogen content of all protein and non-protein fractions was determined by the Kjeldahl method (AOAC, 2000).

2.2. Pig selection and preparation of pork samples Ten (female) pigs from three purebred and two crossbred pigs, including Duroc (D), Landrace (LR), Large White (LW), two-way cross from LR and LW (LR  LW), and three-way cross from D as sire line and LR  LW as dam line (D  [LR  LW]) were randomly selected from the Betagro Hybrid International Co., Ltd. (BHI), Prachinburi, Thailand. All pigs were reared in the same production system, space allowance, diet and handling prior to slaughtering. Pigs with approximately 97–108 kg live weight were transported to a commercial abattoir, Betagro Safety Meat Packing Co., Ltd. (BSM), Lopburi, Thailand. Electrical stunning, exsanguination, scalding, dehairing, evisceration and conditioning were performed according to standard commercial procedures. Following 18 h post-mortem (pm) of chilling, M. longissimus thoracis et lumborum (LTL) between the 10th of thoracic to the 6th lumbar vertebrae from left sides of each carcass were subsequently removed, packed in a polyethylene bag, heat-sealed and transported at 4 °C to the Food Biotechnology Research Unit, National Center for Genetic Engineering and Biotechnology (BIOTEC), Pathumthani, Thailand, within 6 h, for further analyses. At 24 h post mortem, the loin muscles were cut into meat chop samples (2.5 cm) and separately vacuum packed for each analysis. Sampling of meat for each analysis was made to maintain consistency. Samples were frozen and stored at 20 °C until used for preparation of pork emulsion and analyses. Samples were kept at 4 °C for nitrogenous and protein composition and muscle microstructure analyses, which were carried out within 1 day. Samples for the determination of myofibril fragmentation index were cut into 0.5  0.5  1.0 cm pieces, snap-frozen in liquid nitrogen and stored at 80 °C until analysis. 2.3. Determinations of muscle compositions and microstructure 2.3.1. Proximate compositions Moisture, protein, and fat contents were analysed according to the methods of AOAC (2000); 950.46, 928.08 and 960.39, respectively. The values were expressed as % (wet weight basis). Determinations were done in triplicate for each animal. 2.3.2. Muscle protein compositions The protein components were fractionated according to the method of Hashimoto, Watabe, Kono, and Shiro (1979) with slight modification. All procedures were performed at 4 °C. Sample (10 g) was extracted with 10 volumes of solution A (15.6 mM Na2HPO4, 3.5 mM KH2PO4, pH 7.5) using an Ultra Turrax T25 homogeniser (IKA Labortechnik, Selangor, Malaysia) at a speed of 9,500 rpm for 1 min. The homogenate was centrifuged at 5000g for 15 min using an AvantiÒ J–E Centrifuge (Beckman Coulter, Palo Alto, CA). The extraction was repeated twice. The supernatants were combined and mixed with cold 50% (w/v) trichloroacetic acid (TCA) to a final concentration of 10% (w/v). The resulting precipitate was collected by centrifugation at 5000g for 15 min (the water-soluble fraction). The filtrate was the non-protein nitrogen (NPN) fraction. The pellet fraction was extracted with 10 volumes of solution B (0.45 M NaCl, 15.6 mM Na2HPO4, 3.5 mM KH2PO4, pH 7.5) using a homogeniser at 9,500 rpm for 1 min and then centrifuged at 5000g for 15 min. The extraction was repeated twice. The supernatants were combined (the salt-soluble fraction). The pellet obtained was extracted with 10 volumes of 0.1 M NaOH with continuous stirring overnight. The mixture was centrifuged and

2.3.3. Muscle microstructure Muscle microstructures of samples were determined by a scanning electron microscope (SEM), according to the procedure of Palka and Daun (1999) with slight modifications. Meat samples (1  1  0.5 cm) were cut and fixed in 2.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.3, for 2 h at room temperature. The specimens were rinsed with distilled water and dehydrated in graded ethanol solutions with a serial concentration of 25%, 50%, 75%, 95% and 100% (v/v). The dehydration was conducted for 1 h in each solution. Dried samples were mounted on a bronze stub and sputter-coated with gold (SPI-Module sputter coater; Structure Probe Inc., West Chester, PA). The specimens were observed with an SEM JSM-5800 LV (JEOL, Tokyo, Japan) at an acceleration voltage of 15 or 20 kV and a magnification of 350 or 10,000 for transverse and longitudinal sections, respectively. The crosssectional area of the muscle fibre and the length of sarcomere were measured on the SEM image files, using UTHSCSA Image ToolÒ Version 3.00 software (The University of Texas Health Science Center, San Antonio, TX,). Fibre diameter was calculated from the crosssectional area of the fibre (A) using the following equation:

Fibre diameter ðlmÞ ¼ 2  ðA=pÞ1=2 2.3.4. Myofibril fragmentation index Myofibril fragmentation index (MFI) was determined as per the method of Hopkins, Littlefield, and Thompson (2000), with slight modifications. A portion of thawed sample (0.5 g) was cut into small pieces and extracted with 30 ml of ice-cold MFI buffer (25 mM potassium phosphate buffer, pH 7.0 containing 0.1 M KCl, 1 mM EDTA and 1 mM sodium azide) using an Ultra-Turrax T25 homogeniser at 13,500 rpm for two bursts of 30 s with a 30 s break. Myofibril suspension was filtered through two layers of gauze to remove connective tissue and rinsed with 10 ml cold MFI buffer. The filtrates were centrifuged at 1000g for 10 min at 2 °C using an Eppendorf refrigerated centrifuge (Model 5403) and the supernatant was decanted. The pellets of myofibrils were resuspended in 10 ml of cold MFI buffer and centrifuged at 5000g for 15 min. The washing was repeated twice and the pellet was finally re-suspended in 10 ml of cold MFI buffer. The protein concentration of the suspensions was determined using the Bradford method (Bradford, 1976). Aliquots of the suspensions were diluted in MFI buffer to a final protein concentration of 0.5 mg/ml with a total volume of 2 ml. Absorbance at 540 nm of the diluted protein suspensions was read using a spectrophotometer and MFI buffer was used as blank. MFI was calculated by multiplying the average absorbance with 200. 2.4. Study on properties of pork emulsion sausage from different breeds 2.4.1. Preparation of meat emulsion Cold pork emulsion was performed as described by FernándezMartín, López-López, Cofrades, and Colmenero (2009), with slight modifications. Frozen pork and back fat were thawed with running tap water and ground through an 8-mm plate. The formulation was 54.6% pork, 9.6% pork back fat, 1.5% sodium chloride (NaCl), 0.3% sodium tripolyphosphate (STPP), 0.015% sodium nitrite, 6.49% water and 27.5% ice. The salts (NaCl, STPP, and sodium nitrite) were dissolved in the water and added to the ground meat and half

1894

S. Sorapukdee et al. / Food Chemistry 138 (2013) 1892–1901

of the ice. The mixture was chopped in a large size food processor, OsterizerÒ blender 6874B (Oster, South Shelton, CT) at high speed for 120 s. Then, ground back fat and the rest of ice were added and chopped at high speed for 90 s. Finally the whole pork batter was continuously chopped at low speed for 90 s. The final batter temperature was maintained below 12 °C. The batters were stuffed into a 50-ml centrifuge tube and centrifuged at 2500g for 15 min at 4 °C to eliminate any air bubbles. Finally six tubes of each sample were hermetically closed and heated in a temperature-controlled water bath (Memmert, Schwabach, Germany) at 70 °C for 30 min. 2.4.2. Determination of pH Direct pH measurement of the ground pork, uncooked and cooked pork emulsion samples was performed in triplicate using a standard pH meter, Mettler Toledo 320 (Mettler Toledo, Greifensee, Switzerland). 2.4.3. Measurement of surface colour The surface colour of six cylinder-shaped samples (25 mm diameter  20 mm height) was measured in the L⁄a⁄b⁄ mode of CIE by a colourmeter CR300 (Minolta Camera Ltd., Osaka, Japan). The average values were expressed as L⁄ (lightness), a⁄ (redness/ greenness) and b⁄ (yellowness/blueness). 2.4.4. Determination of emulsion stability Emulsion stability was determined according to the method of Colmenero, Ayo, and Carballo (2005). After heating at 70 °C for 30 min, the six tubes of each sample were opened and left to stand upside-down at room temperature for 50 min to release the exudates onto a plate. Total fluid released (TFR) was recorded and expressed as percentage of initial sample weight. Water released (WR) was determined from weight loss after heating the TFR at 105 °C in an oven (Memmert, Schwabach, Germany) for 16 h and expressed as percentage of initial sample weight. Fat released (FR) was calculated as the difference between TFR and WR. 2.4.5. Confocal laser scanning microscopy (CLSM) Lipid distribution of uncooked and cooked pork emulsion was examined with CLSM, FV300 (Olympus, Tokyo, Japan). Uncooked pork emulsion added with 0.001 % (w/v) of Nile Blue A solution was smeared on the microscopy slide. Cooked pork emulsion (1– 2 mm thick) was stained with 0.001 % (w/v) of Nile Blue A solution. The CLSM was operated in the fluorescence mode at an excitation wavelength of 533 nm and emission wavelength of 630 nm using a red helium–neon laser. A magnification of 100 was used. 2.4.6. Texture profile analysis (TPA) TPA was performed using a TA-XT2i texture analyser (Stable Micro Systems, Godalming, England) with cylindrical aluminium probe (50 mm diameter). Six cylinder-shaped samples (25 mm diameter  20 mm height) were prepared and placed on the instrument’s base. TPA textural parameters were measured at room temperature with the following testing conditions: crosshead speed was 1 mm/s, time interval between the first and the second compressions was 1.0 s, working distance was 40% strain and trigger force was 0.2 N. Texture Expert version 1.0 software (Stable Micro Systems) was used to collect and process the data. TPA analyses were defined and calculated as previously described by Bourne (1978). Hardness (N), fracturability (N), adhesiveness (N.mm), springiness (ratio), cohesiveness (ratio), gumminess (N), chewiness (N) and resilience (ratio) were calculated from the force–time curves generated for each sample.

2.5. Statistical analysis Data were subjected to analysis of variance (ANOVA) and mean comparison was performed by Duncan’s multiple-range test (Steel & Torrie, 1980). Pearson’s correlation coefficients were evaluated to describe the relationship between emulsion stability parameters (TFR, WR and FR) and textural parameters as measured by TPA. These statistical analyses were performed by using the Statistical Package for Social Sciences (SPSS for Windows Version 11.5; SPSS Inc., Chicago, IL). Since variables regarding the quality of cooked meat emulsion were the main focus, the relationships between the quality of cooked pork emulsion and the muscle composition and structure variables were determined by PLS regression with PLS2 algorithm (Esbensen, Guyot, Westad, & Houmøller, 2002) for explanation or prediction of the obtained quality of cooked pork emulsion. The PLS2 method extracts a few linear combinations (PLS-components or latent variables) of the muscle composition and structural data (explanatory matrix) that was used to predict the systematic variation in quality of cooked pork emulsion (dependent matrix). PLS2 was used to establish a linear regression model, which calculates the dependent variables (Y) using a combination of the independent (or explanatory) variables (X). The regression model can be written according to the following equation:

Y n ¼ b0 þ b1 X n1 þ    þ bk X nk þ en where, Y and the X values are the measured variables for object n; the b0 is the so-called intercept; the other b-values are the regression coefficients for the X-variables; and e is the random error term (Esbensen et al., 2002). In matrix notation, the same formula can be written as the following equation:

Y ¼ Xb þ e where, Y is the column of Y-values for the n objects, X is the matrix of X-variables and b is the estimated regression coefficients which are estimated from the data (Esbensen et al., 2002). In the present study, the qualities of cooked meat emulsion sausage in terms of emulsion stability or texture were set as dependent variables (Y-variables). Five categorical variables of breeds (D, LW, LR, LR  LW and D  [LR  LW]) and 12 variables of raw meat pH, meat composition (moisture, protein and fat), protein composition (non-protein nitrogen, sarcoplasmic protein, myofibrillar protein, alkaline-soluble protein and stromal protein contents) and muscle structure (fibre diameter, sarcomere length and MFI) were set as explanatory variables (X-variables). Since the values of a categorical variable were not numeric information, each value of the categorical variable was changed as indicator or dummy variable. A dummy variable including the values 1 and 0 in which a value of 1 was set corresponding to effect of each breed. The regression model was assessed by full cross validation (leaveone-out) and all of variables were weighted by a factor of 1/(standard deviation) to equal variance. Marten’s uncertainty test was also applied to show the significant X-variables (Martens & Martens, 2000). After performing PLS2, loading plots were graphed to study overview of the relationship. Additionally the estimated regression coefficients (b-coefficients) were also presented to determine the impacts of explanatory variables in predicting dependent variables, where the optimal number of PLS components were selected based on the minimum value of root mean square error of prediction (RMSEP). The RMSEP represents the average prediction error expected for new samples based on the same units of measurements as the original response variables. PLS regression was taken with Unscrambler Version 9.8 software (Camo AS, Oslo, Norway), but numerical data regarding loading plot were extracted and were new plotted by using the Statistical

1895

S. Sorapukdee et al. / Food Chemistry 138 (2013) 1892–1901

Package for Social Sciences (SPSS for Windows version 11.5; SPSS Inc.). 3. Results and discussion 3.1. Muscle composition, microstructure, pH of pork from different breeds Different moisture, intramuscular fat (IMF) and ash contents were observed among meats from different breeds (p < 0.05), whilst no differences in protein content were observed (p > 0.05) (Table 1). LW meat showed higher moisture content than D and LR  LW counterparts, whereas LR and D  [LR  LW] had intermediate values. Meat from D had the highest IMF content (p < 0.05), followed by that from D  [LR  LW], LW, LR  LW and LR, respectively. Pure or crossbred Duroc pigs were reported to have higher IMF content, compared to white European breeds, including Large White and Landrace (McGloughlin et al., 1988; Oliver, Gispert, & Diestre, 1993; Oliver et al., 1994). Duroc crossing resulted in an increase in IMF content (Channon, Kerr, & Walker, 2004). An increased IMF level was associated with higher tenderness, juiciness of cooked meat, flavour scores and consumer acceptability (Fernández, Monin, Talmant, Mourot, & Lebret, 1999a, 1999b). Based on solubility, meat proteins were classified into five fractions, in which the amount of each fraction varied among pig breeds (Table 1). Meat from LW, a lean-type breed, had the highest myofibrillar and sarcoplasmic protein contents (p < 0.05). Myofibrillar proteins play the most critical role during meat processing as they are responsible for cohesive structure and the firm texture of meat products (Xiong, 1997). In sausages and restructured meats, gel formation of myofibrillar proteins is responsible for retaining

and stabilising water and fat in comminuted meats and for binding meat pieces. Sarcoplasmic proteins include the water-soluble proteins and the proteins in subcellular organelles, such as mitochondria and lysosomes. Myoglobin, which is probably the single most important sarcoplasmic protein in meat, imparts a desirable pinkish-red colour to meat (Xiong, 1997). LR meat had the highest level of non-protein nitrogen (NPN) constituents (p < 0.05). NPN compounds consist of free amino acids, short peptides, creatine, creatine phosphate, creatinine, some vitamins, nucleosides and nucleotides (Aberle et al., 2001, chap. 2). A higher content of alkaline-soluble protein was observed in meats from LW, LR and LR  LW, compared with those from D and D  [LR  LW] (p < 0.05). In contrast, meats from D and D  [LR  LW] contained higher amount of stromal proteins than those from LW, LR and LR  LW (p < 0.05). Fibre diameter and sarcomere length obtained from microstructural images of muscles from different pig breeds are shown in Table 1. The fibre diameters of LR and D  [LR  LW] muscles were largest, followed by those of LR  LW muscles, whereas the smallest diameter of fibre was observed in D and LW muscles (p < 0.05). In contrast, the mean sarcomere lengths of the raw muscles from all breeds and crossbreds were similar, although those of LR muscle were slightly shorter (p < 0.05). MFI indicates the extent of proteolysis caused by the rupture of the I-band and breakage of intermyofibril linkages (Taylor, Geesink, Thompson, Koohmaraie, & Goll, 1995). The result indicated that meat from D had the highest MFI, whereas meats from LR and LR  LW showed the lowest MFI (p < 0.05) (Table 1). This might be caused by the different proteolytic activities in different breeds. Calpains and cathepsins have been reported as the proteases that are involved in the post-mortem proteolysis of porks (Rosell & Toldrá, 1998; Toldrá & Flores, 2000).

Table 1 Chemical compositions and characteristics of raw meats and pork emulsions derived from different pig breeds. Parameters

D

LW

LR

LR  LW

D  [LR  LW]

Characteristics of raw meats Moisture (% wet basis) Protein (% wet basis) Fat (% wet basis) Non-protein nitrogen§ Sarcoplasmic protein§ Myofibrillar protein§ Alkaline-soluble protein§ Stromal protein§ Fibre diameter (lm) Sarcomere length (lm) MFI pH of ground meat

72.30 ± 0.85b, ,à 23.90 ± 0.57a 3.54 ± 0.18a 2.47 ± 0.21d 13.12 ± 0.33b 16.91 ± 0.30b 0.32 ± 0.00c 0.62 ± 0.03a 44.92 ± 0.93c 1.43 ± 0.01a 142.15 ± 0.64a 5.79 ± 0.02a

74.70 ± 0.28a 24.20 ± 0.42a 0.58 ± 0.00c 3.36 ± 0.20b 14.09 ± 0.12a 18.75 ± 0.08a 0.65 ± 0.04ab 0.39 ± 0.02c 43.05 ± 0.56c 1.44 ± 0.03a 121.24 ± 10.88b 5.64 ± 0.01bc

73.90 ± 0.56a 24.00 ± 0.85a 0.25 ± 0.11d 4.11 ± 0.13a 13.55 ± 0.48ab 16.47 ± 0.33b 0.56 ± 0.01abc 0.40 ± 0.01c 49.42 ± 0.56ab 1.33 ± 0.06b 118.70 ± 0.81b 5.58 ± 0.03d

74.90 ± 0.28a 24.15 ± 0.07a 0.47 ± 0.04cd 3.38 ± 0.12b 13.69 ± 0.16ab 16.98 ± 0.45b 0.69 ± 0.21a 0.40 ± 0.02c 46.74 ± 0.88b 1.41 ± 0.01ab 121.70 ± 5.05b 5.59 ± 0.00cd

74.30 ± 0.71a 22.95 ± 0.49a 1.30 ± 0.01b 2.99 ± 0.14c 13.17 ± 0.09b 17.04 ± 0.14b 0.43 ± 0.04bc 0.47 ± 0.01b 51.17 ± 1.41a 1.43 ± 0.01a 133.79 ± 7.66ab 5.68 ± 0.03b

5.94 ± 0.02b 6.12 ± 0.02b 78.18 ± 0.24a 3.30 ± 0.31a 7.81 ± 0.04a 1.63 ± 0.03c 1.54 ± 0.04c 0.09 ± 0.00c 19.75 ± 1.80a 0.19 ± 0.00a 0.37 ± 0.06a 0.90 ± 0.00a 0.49 ± 0.00a 9.77 ± 0.99a 8.81 ± 0.93a 0.32 ± 0.01a

5.93 ± 0.03b 6.12 ± 0.03b 80.00 ± 1.02a 2.89 ± 0.31b 7.99 ± 0.13a 2.71 ± 0.02b 2.56 ± 0.03b 0.15 ± 0.00b 16.63 ± 2.06ab 0.19 ± 0.01a 0.34 ± 0.06a 0.86 ± 0.00bc 0.42 ± 0.02b 7.00 ± 1.18b 6.05 ± 1.01b 0.25 ± 0.02b

5.93 ± 0.01b 6.11 ± 0.01b 77.41 ± 1.35a 3.63 ± 0.03a 7.83 ± 0.37a 2.40 ± 0.17b 2.27 ± 0.15b 0.13 ± 0.02b 15.39 ± 0.32bc 0.19 ± 0.00a 0.53 ± 0.20a 0.87 ± 0.01b 0.42 ± 0.01b 6.46 ± 0.74b 5.66 ± 0.12b 0.24 ± 0.01b

5.96 ± 0.02b 6.14 ± 0.01b 79.24 ± 1.22a 3.47 ± 0.31a 7.91 ± 0.33a 3.23 ± 0.06a 3.04 ± 0.05a 0.18 ± 0.01a 13.06 ± 0.04cd 0.19 ± 0.01a 0.49 ± 0.03a 0.85 ± 0.01bc 0.40 ± 0.01b 5.20 ± 0.15bc 4.45 ± 0.19bc 0.22 ± 0.01b

Characteristics of meat emulsion pHuncooked meat emulsion pHcooked meat emulsion Lightness (L⁄ value) Redness (a⁄ value) Yellowness (b⁄ value) Total fluid released (%) Water released (%) Fat released (%) Hardness (N) Fracturability (N) Adhesiveness (N.mm) Springiness (ratio) Cohesiveness (ratio) Gumminess (N) Chewiness (N) Resilience (ratio)

6.03 ± 0.01a 6.22 ± 0.02a 77.91 ± 0.09a 3.66 ± 0.33a 8.19 ± 0.01a 3.39 ± 0.25a 3.20 ± 0.24a 0.19 ± 0.02a 10.91 ± 1.16d 0.19 ± 0.00a 0.41 ± 0.04a 0.84 ± 0.01c 0.38 ± 0.02b 4.19 ± 0.73c 3.53 ± 0.68c 0.21 ± 0.03b

D: Duroc; LW: Large White; LR: Landrace; LR  LW: Two-way cross; D  [LR  LW]: Three-way cross.   Values are given as means ± SD of each animal. à Different superscripts in the same row indicate significant differences (p < 0.05). § The value are expressed as mg N/g meat.

1896

S. Sorapukdee et al. / Food Chemistry 138 (2013) 1892–1901

Ground meat from D exhibited the highest pH, followed by those form D  [LR  LW], LW, LR  LW and LR, respectively (p < 0.05) (Table 1). Plastow et al. (2005) also found that pH at 24 h post-mortem (pH24h) was higher in Duroc, compared to those of Landrace, whereas Large White and Meishan showed intermediate pH24h values. Meats from different pig breeds showed different muscle composition and structure as assessed by the above analysis. Therefore, pig breed could be considered as the indicator or categorical variables in the explanatory matrix to improve the fitting of the predicted model in the PLS regression analysis (Martens & Martens, 1986). 3.2. Effect of pig breed on properties of pork emulsions Uncooked and cooked pork emulsions made from D showed higher pH, compared to those from other porks (p < 0.05) (Table 1). Due to the addition of STPP, the pH of uncooked and cooked meat emulsions increased in the ranges of 0.24–0.35 and 0.43–0.54 pH units, respectively. Alkaline phosphates increased the pH of the meat batter, usually by about 0.2–0.5 of a pH unit (Smith, 2001). The dominant negative charges contributed by alkaline phosphates cause a dramatic increase of the protein net charge to be more basic and increase the solubilisation and water-binding ability of meat protein, leading to promoted functional performance (Smith, 2001). Cooked pork emulsion made from LR had the lowest redness (a⁄ value) (p < 0.05), whereas the other colour parameters including lightness (L* value) and yellowness (b* value) were not influenced by breed (p > 0.05) (Table 1). Pork from LR possessed the lowest redness as compared to those from other pig breeds. The result was in agreement with Oliver et al. (1993), who reported that meat from LR exhibited lower redness than meat from D and LW. The highest emulsion stability, defined as the lowest percentage of TFR, WR, and FR after heat treatment, was found in meat emulsion made from LW, followed by LR or LR  LW. Emulsions made from D and D  [LR  LW] exhibited the lowest emulsion stability (p < 0.05) (Table 1). Water- and fat-holding properties of comminuted meat products such as emulsion sausages are vitally dependent upon the formation of a protein matrix (Allais, 2010; Lavelle & Foegeding, 1993). The gel matrix from protein–protein interactions has the ability to bind water and hold fat globules (Xiong, 1997). The fat distribution in pork emulsion after staining with Nile Blue A was monitored by a CLSM as illustrated in Fig. 1. For uncooked pork emulsion, the fat particles were homogeneously distributed as illustrated by red droplets against a dark background of the continuous phase that was mainly the mixture of water and proteins. However, there were no detectable differences between uncooked pork emulsions made from different pig breeds (Fig. 1a). After heating (Fig. 1b), cooked pork emulsion made from LW exhibited a more compact and finer network of aggregated protein matrix, which was contrasted with a coarser-stranded network with more empty spaces observed in those made from D  [LR  LW] and D. Generally, gels having a fine network are rigid and elastic and can effectively bind water and fat, whilst gels with a coarse texture are soft and brittle and exhibit poor binding capacity (Ishioroshi, Samejima, & Yasui, 1983; Xiong, Blanchard, Ooizumi, & Ma, 2010). This observation indicated that cooked pork emulsion made from LW showed superior emulsion stability than those made from D. For textural characteristics, cooked pork emulsion made from LW showed the highest hardness, springiness, cohesiveness, gumminess, chewiness and resilience, followed by those made from LR, LR  LW, D  [LR  LW] and D, respectively (p < 0.05) (Table 1). In contrast, fracturability and adhesiveness of cooked pork emulsion were not influenced by breed (p > 0.05).

Fig. 1. Confocal laser scanning micrographs of fat distribution in uncooked pork emulsion (a) and cooked pork emulsion at 70 °C for 30 min (b) made from different pig breeds. Magnification: 100. Scale bar = 200 lm.

The correlation coefficients between properties of cooked pork emulsion and the chemical constituents and microstructural parameters of pork from different breeds were determined (Table 2). It can be observed that TFR, WR, FR, hardness, springiness, cohesiveness, gumminess, chewiness and resilience showed a significant correlation with chemical constituents and microstructural parameters (p < 0.05). However, there were no significant correlations found for fracturability and adhesiveness (p < 0.05). The results suggested that these attributes were much less affected by the explanatory matrix, and therefore could be neglected from the dependent matrix. As a result, TFR, WR, FR, hardness, springiness, cohesiveness, gumminess, chewiness and resilience were se-

1897

S. Sorapukdee et al. / Food Chemistry 138 (2013) 1892–1901 Table 2 Correlation coefficients between muscle composition and structure and properties of cooked pork emulsions from different pig breeds.

TRF WR FR Hardness Fracturability Adhesiveness Springiness Cohesiveness Gumminess Chewiness Resilience * **

pH of meat

Moisture

Protein

Fat

Nonprotein nitrogen

Sarcoplasmic protein

Myofibrillar protein

Alkalinesoluble protein

Stromal protein

Fibre diameter

Sarcomere length

MFI

0.583 0.582 0.581 0.666* 0.160 0.032 0.561 0.409 0.568 0.557 0.435

0.625 0.624 0.593 0.753* 0.423 0.058 0.747* 0.720* 0.736* 0.731* 0.721*

0.534 0.533 0.547 0.362 0.012 0.006 0.394 0.347 0.366 0.373 0.343

0.664* 0.661* 0.671* 0.770** 0.098 0.026 0.666* 0.549 0.682* 0.670* 0.569

0.480 0.477 0.498 0.654* 0.081 0.232 0.400 0.364 0.540 0.520 0.426

0.836** 0.837** 0.847** 0.830** 0.161 0.207 0.767** 0.776** 0.834** 0.835** 0.756*

0.695* 0.698* 0.657* 0.547 0.406 0.253 0.714* 0.779** 0.664* 0.683* 0.745*

0.724* 0.725* 0.708* 0.717* 0.241 0.237 0.694* 0.630 0.688* 0.681* 0.633*

0.743* 0.740* 0.757* 0.799** 0.072 0.006 0.724* 0.601 0.729* 0.722* 0.606

0.470 0.473 0.453 0.247 0.232 0.232 0.378 0.471 0.359 0.377 0.431

0.089 0.088 0.090 0.330 0.028 0.366 0.014 0.015 0.208 0.183 0.111

0.652* 0.651* 0.647* 0.797** 0.297 0.134 0.648* 0.614 0.729* 0.717* 0.627

Correlation is significant at p < 0.05. Correlation is significant at p < 0.01.

lected in the dependent matrix, and the explanatory matrix was formed by the chemical constituents and microstructural parameters and categorical variables of pig breeds. 3.3. PLS regression analysis 3.3.1. Multivariate analysis of stability and texture of cooked pork emulsion In order to analyse the relationships among all selected properties of cooked pork emulsion and the chemical constituents and microstructural parameters of porks from different breeds, the PLS loading plot after applying PLS2 of the first two PLS components was shown in Fig. 2. Overall 92% variation of properties of cooked pork emulsions was explained by 67% variation of chemical constituents and microstructural parameters. The first PLS component explained 42% of the explanatory matrix (chemical constituents and microstructural parameters) and 82% of dependent matrix (quality of cooked meat emulsion). The second component

explained 25% of the explanatory matrix, and 10% of the dependent matrix. For the explanatory matrix (Fig. 2), the first PLS component was significantly positively defined by the categorical variable of LW, myofibrillar protein, sarcoplasmic protein and alkaline-soluble protein, and was negatively defined by pork from D, pH of ground meat, fat, MFI and stromal protein. The second PLS component was mainly significantly defined by the categorical variable of LW, myofibrillar protein with positive correlation and fibre diameter with negative correlation. For the dependent matrix, cohesiveness, springiness, resilience, chewiness, gumminess and hardness were positively correlated on the first PLS component, but TFR, WR and FR were negatively correlated on this component. Likewise, on the second PLS component, cohesiveness, springiness, resilience, chewiness, gumminess and hardness were negatively related with TFR, WR and FR. Regarding the prediction of dependent variables by explanatory variables in the first PLS component, the variations in cohesiveness, springiness, resilience, chewiness, gumminess and hardness were significantly positively

Fig. 2. PLS loading plot for the first two PLS components.   Significant explanatory (X) variables after applying Marten’s uncertainty test. Explanatory (X) variables were moisture (M), protein (P), fat (F), non-protein nitrogen (NPN), sarcoplasmic protein (SP), myofibrillar protein (MP), alkali-soluble protein (AP), stromal protein (ST), sarcomere length (SL), fibre diameter (FD), myofibril fragmentation index (MFI) and pH of ground pork (pH). Dependent (Y) variables were total fluid released (TFR), water released (WR), fat released (FR), hardness (HAR), springiness (SPR), cohesiveness (COH), gumminess (GUM), chewiness (CHE) and resilience (RES). Categorical variables were Duroc (D), Large White (LW), Landrace (LR), two-way cross (LR  LW) and three-way cross (D  [LR  LW]).

1898

S. Sorapukdee et al. / Food Chemistry 138 (2013) 1892–1901

correlated with pork from LW, myofibrillar protein, sarcoplasmic protein and alkaline-soluble protein, but were significantly negatively correlated with meat from D, pH of ground meat, fat, MFI and stromal protein. Moreover, the variation in these textural parameters was further explained by the second PLS component, which were significantly positively related with meat from LW and myofibrillar protein, but were significantly negatively related with fibre diameter. The variations of WR, TFR and FR were interpreted as the opposite way of variation in those textural characteristics. It was suggested that the superior texture of cooked pork emulsion made from LW was concomitant with higher waterand fat-binding properties. 3.3.2. Univariate analysis Univariate analysis was performed to estimate the contribution of each explanatory variable on prediction of each dependent variable. Considering the percentage of variation explained with the set of various explanatory variables when each of the dependent variables was analysed individually, WR and TFR were the best explained attributes, with about 90% of their variations explained by two PLS components, followed by FR (86% by 4 PLS components),

chewiness (84% by 4 PLS components), gumminess (82% by 4 PLS components), cohesiveness (80% by 4 PLS components), resilience (79% by 4 PLS components), springiness (79% by 2 PLS components) and hardness (73% by 4 PLS components), respectively. 3.3.2.1. Emulsion stability of cooked pork emulsion. The estimated regression coefficients of explanatory variables for predicting WR, TFR and FR exhibited a similar pattern (Fig. 3a–c). It was shown that the categorical variable of LW was the most important factor for the variation of WR, TFR and FR and was negatively correlated with those from D. Nevertheless, other breeds exhibited a minor impact and were not significant in predicting these properties. WR, TFR and FR of cooked meat emulsion made from LW were significantly positively related with the higher contents of myofibrillar protein, sarcoplasmic protein and alkaline-soluble protein and were significantly negatively related with fibre diameter, stromal protein content, MFI value and fat content. Except for categorical variable of LW, the amount of myofibrillar protein was the most important factor for predicting variation in WR, TFR and FR, followed by the amount of sarcoplasmic protein. Among the meat proteins, myofibrillar proteins that are extracted

Fig. 3. Regression coefficients of explanatory (X) variables for predicting variation in WR (a) and TFR (b) and FR (c). Significant explanatory (X) variables after applying Marten’s uncertainty test were shown as striped column. Categorical variables were Duroc (D), Large White (LW), Landrace (LR), two-way cross (LR  LW) and three-way cross (D  [LR  LW]). Explanatory (X) variables were pH of ground pork (pH), moisture (M), protein (P), fat (F), non-protein nitrogen (NPN), sarcoplasmic protein (SP), myofibrillar protein (MP), alkali-soluble protein (AP), stromal protein (ST), fibre diameter (FD), sarcomere length (SL) and myofibril fragmentation index (MFI).

S. Sorapukdee et al. / Food Chemistry 138 (2013) 1892–1901

into the water phase during comminution and blending are generally considered to be the most important factor for emulsification and the quality of meat network (Tornberg, 2005; Zorba & Kurt, 2006; Zorba, Kurt, & Gençcelep, 1995). Due to its amphiphilic character, a myofibrillar protein, specifically myosin, forms a surface monolayer film at the water–fat interface with its hydrophilic part toward water and its hydrophobic part oriented toward fat (Hernández-Hernández & Guerrero-Legarreta, 2010; Jones, 1984). Additionally, myofibrillar proteins extracted during chopping and emulsification could form a dense protein network, which holds water and fat efficiently inside the protein matrix (Tornberg, 2005). Sarcoplasmic proteins can act as emulsifiers, but myofibrillar proteins are preferably adsorbed to the water–fat interface (Zorba & Kurt, 2006). However, Galluzzo and Regenstein (1978) and Hegarty, Bratzler, and Pearson (1963) reported that a higher emulsion capacity of myofibrillar protein, particularly myosin and actomyosin, rather than sarcoplasmic protein only showed in a diluted protein concentration (5 mg/ml in 0.3 or 0.6 M NaCl solution). Increases in ionic strength and protein concentration diminish the differences between different proteins. At a protein concentration of 12 mg/ml in 0.5 M KCl, myofibrillar protein, including myosin, actin and tropomyosin–troponin, and sarcoplasmic proteins had equal emulsifying capacities (Tsai, Cassens, & Briskey, 1972). In contrast to myofibrillar protein gel, that produced from sarcoplasmic protein is very weak (Ugalde-Benítez, 2012). However, heat-induced aggregated sarcoplasmic proteins can form a gel between the structural gel elements, thereby linking them together (Davey & Gilbert, 1974). The next important factor for predicting a variation of WR, TFR and FR was fibre diameter, where a small fibre diameter could lower WR, TFR and FR of cooked pork emulsion. The physical properties of protein, such as conformation, size and shape, also

1899

influence the protein performance (Liu, Zhao, Xiong, Qiu, & Xie, 2008). As muscle fibre diameter decreased, the gel of cooked surimi-like materials from chicken, pork and beef contained smaller particles of amorphous proteins along with hard and strong gel structure (Kang, Park, Joo, Lee, & Lee, 2010). Furthermore, a lower extent of myofibril fragmentation as indicated by MFI of raw meat also contributed to a lower WR, TFR and FR of cooked pork emulsion. Severe proteolysis of myofibrillar protein caused by the endogenous proteases in muscle is directly associated with poor gel quality (An, Peters, & Seymour, 1996; Katayama, Chin, Yoshihara, & Muguruma, 2006). In sausages and restructured meats, gel formation of myofibrillar proteins is responsible for network formation (Xiong, 1997). Intramuscular fat and stromal protein had also significantly positive impact on WR, TFR and FR. The complex meat system consists of not only dissolved proteins but also insoluble components, e.g., connective tissue and intramuscular fat. The amount and state of these components have a large impact on gelation (Tornberg, 2005), which might interfere or dilute the ability of the myofibrillar proteins to form a strong gel but could enhance WR, TFR and FR. The contribution of stromal proteins to emulsification of fats and water-holding capacity in sausage is very low (Pearson & Gillett, 1996, chap. 9). Stromal proteins can be broken down into three main categories: collagen, elastin, and reticulin; commonly referred to as connective tissue (Aberle et al., 2001, chap. 2). Collagen when present at high levels in poultry formulations may interfere with the functionality of myofibrillar proteins (Smith, 2001). The results suggested that although meat from LW possessed a high amount of alkali-soluble protein, there are several important factors for promoting a low WR, TFR and FR of cooked meat emulsion made from LW. These results was supported by high amount of myofibrillar protein, sarcoplasmic protein with small fibre diameter and low MFI value as well as a low amount

Fig. 4. Regression coefficients of explanatory (X) variables for predicting variation in hardness (a), chewiness (b), gumminess (c), cohesiveness (d), resilience (e), and springiness (f). Significant explanatory (X) variables after applying Marten’s uncertainty test are shown as striped columns. Categorical variables were Duroc (D), Large White (LW), Landrace (LR), two-way cross (LR  LW) and three-way cross (D  [LR  LW]). Explanatory (X) variables were pH of ground pork (pH), moisture (M), protein (P), fat (F), non-protein nitrogen (NPN), sarcoplasmic protein (SP), myofibrillar protein (MP), alkali-soluble protein (AP), stromal protein (ST), fibre diameter (FD), sarcomere length (SL) and myofibril fragmentation index (MFI).

1900

S. Sorapukdee et al. / Food Chemistry 138 (2013) 1892–1901

of insoluble component, such as stromal protein and intramuscular fat of meat from LW. 3.3.2.2. Textural properties of cooked pork emulsion. Selected textural parameters of cooked pork emulsion including hardness, chewiness, gumminess, cohesiveness, resilience and springiness are illustrated in Fig. 4a–f. It was observed that these regression models tended to be similar and closely interrelated with those of emulsifying properties as previously discussed. In common, all textural attributes of cooked meat emulsion were related positively with the categorical variable of LW, the contents of myofibrillar and sarcoplasmic proteins were related negatively with MFI value. However, individual textural attributes of cooked pork emulsion were also significantly affected by other variables associated with raw meat. Hardness was also negatively impacted by the categorical variable of D, the contents of fat, stromal protein and positively affected by the content of alkaline-soluble protein (Fig. 4a). Similar to hardness, chewiness and gumminess were negatively impacted not only by the contents of fat and stromal protein but also muscle fibre diameter (Fig. 4b and c). Cohesiveness and resilience were negatively influenced by the muscle fibre diameter (Fig. 4d and e). Lastly, springiness was positively impacted by the content of alkaline-soluble protein and negatively impacted by the muscle fibre diameter (Fig. 4f). Thus, the results not only emphasise the significant contributions of pig breed, protein compositions (Tornberg, 2005), muscle structure (Kang et al., 2010) and proteolytic potential (An et al., 1996; Katayama et al., 2006) of meat on textural quality of cooked pork emulsion but also suggest how to select meat when a particular textural attribute of cooked meat emulsion is of concern. Since the textural attributes of emulsified sausage are dependent upon the formation of a protein gel matrix (Zogbi & Benejam, 2010), the superior textural properties of cooked pork emulsion made from LW meat could be attributed to (1) higher amounts of gelling components, such as myofibrillar, sarcoplasmic, alkali-soluble proteins, (2) lower amounts of insoluble components, such as stromal protein and intramuscular fat, (3) smaller muscle fibre diameter and (4) lower degradation of myofibril-associated proteins, which resulted in the reduction in molecular size and a loss of structural domains which are probably essential for molecular interaction and binding (Visessanguan & An, 2000). Even though the crosslink and deposition of small protein fragments may occur upon heating, the resulting gel structures are much weaker than those formed from the intact molecules. 4. Conclusion Cooked pork emulsion made from LW exhibited superior stability and textural properties to those from other pig breeds. The variation in those qualities could be explained by the differences in muscle composition and structure. Based on the univariate analysis, higher emulsion stability and textural properties of LW were influenced by higher amounts of myofibrillar and sarcoplasmic proteins, smaller fibre diameter and lower myofibril fragmentation at the time of use. This finding could provide better understanding in the properties and stability of cooked pork meat emulsion which was determined by composition and muscle structure that associated with pig breed. Acknowledgements This work was supported by a Grant from the Thailand Research Fund (TRF) with Office of Small and Medium Enterprises Promotion (OSMEP) through the Royal Golden Jubilee Ph.D. Program to Supaluk Sorapukdee (2.F.PS/50/D.2). The authors would like to express their sincere thanks to Graduate School of Prince of Songkla Uni-

versity and BIOTEC for the financial support, and BHI and BSM for pork used in this study.

References Aberle, E. D., Forrest, J. C., Gerrard, D. E., Mills, E. W., Hedrick, H. B., Judge, M. D., et al. (2001). Principles of meat science (4th ed.). Iowa: Kendall/Hunt Publishing Company. Allais, I. (2010). Emulsification. In F. Toldrá (Ed.), Handbook of meat processing (pp. 145–295). Ames, Iowa: Blackwell Publishing. An, H., Peters, M. Y., & Seymour, T. A. (1996). Roles of endogenous enzymes in surimi gelation. Trends in Food Science & Technology, 7, 321–327. AOAC. (2000). Official methods of analysis (17th ed.). Gaithersberg, MD: Association of Official Analytical Chemists. Bourne, M. C. (1978). Texture profile analysis. Food Technology, 32, 62–65. Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Analytical Biochemistry, 72, 248–254. Channon, H. A., Kerr, M. G., & Walker, P. J. (2004). Effect of Duroc content, sex and ageing period on meat and eating quality attributes of pork loin. Meat Science, 66, 881–888. Colmenero, F. J., Ayo, M. J., & Carballo, J. (2005). Physicochemical properties of low sodium frankfurter with added walnut: Effect of transglutaminase combined with caseinate, KCl and dietary fibre as salt replacers. Meat Science, 69, 781–788. Davey, C. L., & Gilbert, K. V. (1974). Temperature-dependent cooking toughness in beef. Journal of the Science of Food and Agriculture, 25, 931–938. Esbensen, K. H., Guyot, D., Westad, F., & Houmøller, L. P. (2002). Multivariate data analysis-in practice: An Introduction to multivariate data analysis and experimental design (5th ed.). Oslo, Norway: Camo. Fernandez, X., Monin, G., Talmant, A., Mourot, J., & Lebret, B. (1999). Influence of intramuscular fat content on the quality of pig meat—1. Composition of the lipid fraction and sensory characteristics of M. longissimus lumborum. Meat Science, 53, 59–65. Fernández, X., Monin, G., Talmant, A., Mourot, J., & Lebret, B. (1999). Influence of intramuscular fat content on the quality of pig meat—2. Consumer acceptability of M. longissimus lumborum. Meat Science, 53, 67–72. Fernández-Martín, F., López-López, I., Cofrades, S., & Colmenero, F. J. (2009). Influence of adding Sea Spaghetti seaweed and replacing the animal fat with olive oil or a konjac gel on pork meat batter gelation. Potential protein/alginate association. Meat Science, 83, 209–217. Foegeding, E. A., Clark, J. B., & Xiong, Y. L. (1991). Gelation of myofibrillar proteins. In N. Parris & R. Barford (Eds.), Interactions of food proteins (pp. 257–267). Washington, DC: American Chemical Society. Galluzzo, S. J., & Regenstein, J. M. (1978). Emulsion capacity and timed emulsification of chicken breast muscle myosin. Journal of Food Science, 43, 1757–1760. Hashimoto, K., Watabe, S., Kono, M., & Shiro, K. (1979). Muscle protein composition of sardine and mackerel. Bulletin of the Japanese Society of Scientific Fisheries, 45, 1435–1441. Hegarty, G. R., Bratzler, L. J., & Pearson, A. M. (1963). The relationship of some intracellular protein characteristics to beef muscle tenderness. Journal of Food Science, 28, 525–530. Hernández-Hernández, E., & Guerrero-Legarreta, I. (2010). Gelation and emulsion: Applications. In I. Guerrero-Legarreta (Ed.). Handbook of poultry science and technology (Vol. 2, pp. 25–33). New Jersey: John Wiley & Sons, Inc. Hopkins, D. L., Littlefield, P. J., & Thompson, J. M. (2000). A research note on factors affecting the determination of myofibrillar fragmentation. Meat Science, 56, 19–22. Ishioroshi, M., Samejima, K., & Yasui, T. (1983). Heat-induced gelation of myosin filaments at a low salt concentration. Agricultural and Biological Chemistry, 47, 2809–2816. Jones, K. W. (1984). Protein lipid interactions in processed meats. In Proceedings of the 37th annual reciprocal meat conference (pp. 52–57). Kang, G.-H., Park, G.-B., Joo, S.-T., Lee, M., & Lee, S.-K. (2010). Effects of muscle fiber types on gel property of surimi-like materials from chicken, pork and beef. Journal of Muscle Foods, 21, 570–584. Katayama, K., Chin, K. B., Yoshihara, S., & Muguruma, M. (2006). Microbial transglutaminase improves the property of meat protein and sausage texture manufacture with low-quality pork loin. Asian–Australian Journal of Animal Science, 19, 102–108. Ker, Y. C., & Toledo, R. T. (1992). Influence of shear treatments on consistency and gelling properties of whey protein isolate suspensions. Journal of Food Science, 57, 82–85. Lavelle, C. L., & Foegeding, E. A. (1993). Gelation of turkey breast and thigh myofibrils: Effects of pH, salt and temperature. Journal of Food Science, 58, 727–730. Liu, R., Zhao, S., Xiong, S., Qiu, C., & Xie, B. (2008). Rheological properties of fish actomyosin and pork actomyosin solutions. Journal of Food Engineering, 85, 173–179. Martens, M., & Martens, H. (1986). Statistical procedure in food research. London, UK: Elsevier Applied Science. Martens, H., & Martens, M. (2000). Modified Jack-knife estimation of parameter uncertainty in bilinear modelling by partial least squares regression (PLSR). Food Quality and Preference, 11, 5–16.

S. Sorapukdee et al. / Food Chemistry 138 (2013) 1892–1901 McGloughlin, P., Allen, P., Tarrant, P. V., Joseph, R. L., Lynch, P. B., & Hanrahan, T. J. (1988). Growth and carcass quality of crossbred pigs sired by Duroc, Landrace and Large White boars. Livestock Production Science, 18, 275–288. Oliver, M. A., Gispert, M., & Diestre, A. (1993). The effects of breed and halothane sensitivity on pig meat quality. Meat Science, 35, 105–118. Oliver, M. A., Gou, P., Gispert, M., Diestre, A., Arnau, J., & Noguera, J. L. (1994). Comparison of five types of pig crosses. II. Fresh meat quality and sensory characteristics of dry cured ham. Livestock Production Science, 40, 179–185. Palka, K., & Daun, H. (1999). Changes in texture, cooking losses, and myofibrillar structure of bovine M. semitendinosus during heating. Meat Science, 51, 237–243. Pearson, A. M., & Gillett, T. A. (1996). Processed meats (3rd ed.). New York: Chapman and Hall. Plastow, G. S., Carrión, D., Gil, M., García-Regueiro, J. A., Font i Furnols, M., Gispert, M., et al. (2005). Quality of pork genes and meat production. Meat Science, 70, 409–421. Rosell, C. M., & Toldrá, F. (1998). Comparison of muscle proteolytic and lipolytic enzyme levels in raw hams from Iberian and White pigs. Journal of the Science of Food and Agriculture, 76, 117–122. Smith, D. M. (1988). Factors influencing texture formation in comminuted meats. Reciprocal Meat Conference Proceedings, 41, 48–53. Smith, D. M. (2001). Functional properties of muscle proteins in processed poultry products. In A. Sams (Ed.), Poultry meat processing (pp. 181–194). Boca Raton, FL: CRC Press. Steel, R. G. D., & Torrie, J. H. (1980). Principle and procedure of statistics (2nd ed.). New York: Macgraw-Hill. Taylor, R. G., Geesink, G. H., Thompson, V. F., Koohmaraie, M., & Goll, D. E. (1995). Is Z-disk degradation responsible for post mortem tenderization? Journal of Animal Science, 73, 1351–1367.

1901

Toldrá, F., & Flores, M. (2000). The use of muscle enzymes as predictors of pork meat quality. Food Chemistry, 69, 387–395. Tornberg, E. (2005). Effects of heat on meat proteins – Implications on structure and quality of meat products. Meat Science, 70, 493–508. Tsai, R., Cassens, R. G., & Briskey, E. J. (1972). The emulsifying properties of purified muscle proteins. Journal of Food Science, 37, 286–288. Ugalde-Benítez, V. (2012). Meat emulsion. In Y. H. Hui (Ed.), Handbook of meat and meat processing (pp. 447–456). Boca Raton, FL: CRC Press. Visessanguan, W., & An, H. (2000). Effects of proteolysis and mechanism of gel weakening in heat-induced gelation of fish myosin. Journal of Agricultural and Food Chemistry, 48, 1024–1032. Xiong, Y. L., Blanchard, S. P., Ooizumi, T., & Ma, Y. (2010). Hydroxyl radical and ferryl-generating systems promote gel network formation of myofibrillar protein. Journal of Food Science, 75, 215–221. Xiong, Y. (1997). Structure-function relationships of muscle proteins. In S. Damodaran & A. Paraf (Eds.), Food proteins and their applications (pp. 341–392). New York: CRC Press. Ziegler, G. R., & Acton, J. C. (1984). Mechanisms of gel formation by proteins of muscle tissue. Food Technology, 38, 77–82. Zogbi, A. P., & Benejam, W. O. (2010). Gelation and emulsion: Principles. In I. Guerrero-Legarreta (Ed.), Handbook of poultry science and technology, volume 2, secondary processing (pp. 15–21). New Jersey: John Wiley and Sons. Zorba, Ö., & Kurt, Sß. (2006). Optimization of emulsion characteristics of beef, chicken and turkey meat mixtures in model system using mixture design. Meat Science, 73, 611–618. Zorba, Ö., Kurt, Sß., & Gençcelep, H. (1995). Effect of thermal process degrees on protein solubility, different emulsion and electrophoretic characteristics of protein fractions in fresh and frozen beef. Erzurum, Turkey: Atatürk University. Unpublished doctoral dissertation.