Eating quality of “Vitela Tradicional do Montado”-PGI veal and Mertolenga-PDO veal and beef

Meat Science 94 (2013) 63–68

Contents lists available at SciVerse ScienceDirect

Meat Science journal homepage: www.elsevier.com/locate/meatsci

Eating quality of “Vitela Tradicional do Montado”-PGI veal and Mertolenga-PDO veal and beef Ana C.G. Monteiro ⁎, Eduardo Gomes, António S. Barreto, Marina F. Silva, Magda A. Fontes, Rui J.B. Bessa, José P.C. Lemos CIISA, FMV, Universidade Técnica de Lisboa — Avenida da Universidade Técnica, Pólo Universitário Alto da Ajuda, 1300-477 Lisboa, Portugal

a r t i c l e

i n f o

Article history: Received 5 March 2012 Received in revised form 7 December 2012 Accepted 10 December 2012 Keywords: Beef Meat quality Mertolenga Sensory analysis Veal

a b s t r a c t Physicochemical and sensory characteristics were measured in veal and beef from the Portuguese Mertolenga breed having 3 quality labels as follows: Mertolenga-PDO beef and veal which apply to purebred animals and “Vitela Tradicional do Montado”-PGI veal which applies to crossbred animals. Measurements were made in longissimus lumborum muscle aged for 6 days. The temperature 3 h post-mortem (T3), cooking losses and Warner–Bratzler shear force (WBSF) reflected carcass weight (CW) differences between groups. The pigment content was influenced by age, with beef having higher values than veal. WBSF correlated negatively with intramuscular fat in Mertolenga-PDO beef, but not on veal. WBSF correlated positively with cooking losses and negatively with myofibrillar fragmentation index, tenderness, juiciness and overall acceptability. Cooking losses and juiciness were the main contributors for the tenderness differences. Vitela Tradicional do Montado-PGI and Mertolenga-PDO veal had lighter colour and were considered tender. The three meat types were well discriminated based on pHu, a* and C* parameters by canonical discriminant analysis. © 2012 Elsevier Ltd. All rights reserved.

1. Introduction Consumer demand for beef is highly influenced by consumer concerns about beef quality, health issues, nutritional value and safety, environment and animal welfare requirements (Xue, Mainville, You, & Nayga, 2010), as well as by the product origin (Banović, Grunert, Barreira, & Fontes, 2010). Nevertheless, sensory characteristics still remain the main purchasing and repeated purchasing criteria (Calkins & Hodgen, 2007). The promotion of products having characteristics that could be of considerable benefit to the rural economy, in particular to less-favoured or remote areas, has been encouraged. Portuguese beef production is mainly based on intensive systems with crossbred cattle, but the production of autochthonous beef breeds has increased in the last decade (GPP, 2009), which has an important economic and social role in the population/regions where they are undertaken. The Protected Designation of Origin (PDO) and Protected Geographic Identification (PGI) are labels usually under geographical groupings and with specific genotypes and production systems (Council Regulation (EC) n 510/2006 of 20 March 2006, 2006), partly created to meet the increasing consumers demand for quality guarantees, animal welfare and environment protection (Guerreiro, 2001). One such example is Mertolenga breed, the largest herd among Portuguese cattle breeds, which can be marketed as Mertolenga-PDO beef and Mertolenga-PDO veal when animals are

⁎ Corresponding author. Tel.: +351 213 652 884; fax: +351 213 652 889. E-mail address: [email protected] (A.C.G. Monteiro). 0309-1740/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.meatsci.2012.12.011

purebred, or as “Vitela Tradicional do Montado”-PGI veal, when animals are crossbred, mainly with Limousin and Charolais sires. Although these meats have been marketed for years, scientific studies supporting their differentiation are scarce and limited to the composition of its lipid fraction (Alfaia et al., 2006; Costa et al., 2008; Monteiro, Fontes, Bessa, Prates, & Lemos, 2012). Therefore, the aim of this study is to determine the physicochemical characteristics and sensory attributes of Portuguese “Vitela Tradicional do Montado”-PGI veal and Mertolenga-PDO veal and beef, raised under the typical production systems.

2. Materials and methods 2.1. Animals This study was performed on 126 animals of “Vitela Tradicional do Montado-PGI” calves, Mertolenga-PDO calves and Mertolenga-PDO young bulls selected from 23 different herds, divided in two slaughter periods. The first slaughter period occurred in the autumn, whereas the second slaughter period occurred in late spring and summer of the following year. The three meat types have Mertolenga breed in common, Mertolenga-PDO veal and Mertolenga-PDO beef are from purebred animals (Mertolenga-PDO calves and young bulls, respectively) and “Vitela Tradicional do Montado-PGI” veal from crossbred animals, mainly Limousin × Mertolenga and Charolais × Mertolenga (“Vitela Tradicional do Montado-PGI” calves).

64

A.C.G. Monteiro et al. / Meat Science 94 (2013) 63–68

According to the specifications these meat products are defined based on age and carcass weight (CW) as: 1- “Vitela Tradicional do Montado”-PGI veal — calves until 12 months and 180 kg of CW; 2- Mertolenga-PDO veal — calves until 15 months and 220 kg CW; 3- Mertolenga-PDO beef — young bulls from 15 to 30 months and more than 220 kg CW. The animals of this study were distributed into the three meat groups as shown in Table 1. All animals were raised in a traditional semi-extensive production system in the Alentejo region of Portugal based on natural pastures under holm and cork oak (“Montado”). Calves remained with their mothers on natural pastures until weaning (6–9 months of age), after that they were supplemented with concentrate and straw. The young bulls (Mertolenga-PDO beef) were finished on concentrate (Table 2) during 3–5 months. The animals were slaughtered and dressed in an officially approved slaughterhouse, according to standard methods, using a captive bolt stunner, followed by sticking and bleeding, from June to September. Unfortunately, in the first slaughter period due to a failure in the electrically stimulation device, carcasses were not electrically stimulated. Thus, the first slaughter period data on variables like T3 (temperature 3 h post-mortem), pH3 (pH 3 h post-mortem), myofibrillar fragmentation index (MFI), cooking losses, Warner–Bratzler shear force (WBSF) and sensory attributes were not included in the study. In the second slaughter period after slaughter the carcasses were electrically stimulated (70 V, 60 Hz for 2 min, right after bleeding, 3 min of death) according to the device specifications (Dalpino Est 01, Indústrias e Serra Dalpino, Lda., Campinas, Brazil), to avoid cold shortening. Classification of conformation and fatness degree were assessed by trained and experienced evaluators using the SEUROP system. Carcass conformation was based on visual assessment of muscle mass development (S = superior; E = excellent; U = very good; R = good; O = fair; P = poor), and the fatness degree was based on the amount and distribution of fat and a higher number indicates the thickest fat cover (1 = poor, 2 = slight, 3 = average, 4 = high, 5 = very high). Three hours post-mortem, when muscle temperature and pH (T3 and pH3) were measured, the conditions in the cooling chamber were registered: air velocity ranged between 0.1 and 0.6 m/s; temperature ranged between 3 and 5 °C; and humidity ranged between 82 and 92%. 2.2. Analysis of meat quality The longissimus lumborum pH3 and T3 were measured 3 h postmortem at the 1st lumbar vertebrae level with a HI 99163 portable pH-meter (Hanna Instruments Inc., Rhode Island, USA). After 72–96 h samples of longissimus lumborum were collected between the first and third lumbar vertebrae (around 0.7–1.0 kg), vacuum packaged and stored at 0–1 °C until being processed. Six days after slaughter ultimate pH

Table 1 Number, age and carcass weight of the animals used in the two slaughter periods. VTM-PGI calves

Mertolenga-PDO calves

Mertolenga-PDO bulls

First slaughter period (autumn) n 21 Age (months) 9 ± 1.7 Carcass weight (kg) 166 ± 11.0

20 11 ± 2.6 167 ± 10.8

17 24 ± 3.2 296 ± 45.4

Second slaughter period (spring–summer) n 23 Age (months) 10 ± 1.5 Carcass weight (kg) 164 ± 9.5

23 11 ± 2.1 162 ± 12.2

22 18 ± 3.4 251 ± 34.9

Table 2 Chemical composition (%) of concentrate feed according to “Vitela Tradicional do Montado”-PGI veal, Mertolenga-PDO veal and Mertolenga-PDO beef specifications. Characteristics

Calves (age b 12 months)

Young bulls (age > 12 months)

Crude protein Crude fat Crude fibre Starch WSC Ash Calcium Phosphorus

16.0–16.5 2.6–6.0 6.0–9.7 Min 25.0 Min 4.0 Max 9.0 1.0–1.2 0.55–0.62

16.0–18.0 2.5–5.0 6.0–9.8 Min 20.0 Min 4.5 Max 9.0 0.95–1.2 0.50–0.62

WSC = water soluble carbohydrates.

(pHu), colour and dry matter (DM) were measured in the refrigerated samples. Afterwards, the samples were minced, vacuum packaged and stored at −18 °C until laboratory analyses were performed. Two steaks were left intact for WBSF measurement and sensory analysis. The pHu was measured three times in each sample with the same device used to measure pH3. The meat colour measurements were carried out with a Minolta CR 300 colorimeter (Konica Minolta Holdings Inc., Tokyo, Japan) with a C iluminant and a 2° standard observer in the CIELAB space, after 1 h of blooming to allow oxygenation (Larraín, Schaeffer, & Reed, 2008). The dry matter content was determined in muscle samples by microwaves (Smart System 5, CEM Microwaves Technology Ltd., Buckingham, UK), following the device specifications. The total pigment content (g/100 g DM) was determined through the quantification of the cyanometmyoglobin and cyanomethemoglobin as described by Wierbicki, Cahill, Kunkle, Klosterman, and Deatherage (1955). The intramuscular fat content (g/100 g DM) was measured according to the AOAC official method 945.16 (AOAC, 2000) in fresh samples. The total collagen concentration (g/100 g DM) was determined through hidroxiproline quantification according to the AOAC official method 990.26 (AOAC, 2000). The collagen solubility was determined as described by Silva, Patarata, and Martins (1999) however, the dilution steps were altered. After hydrolysis (16 h at 105 °C) and the volume adjusted to 100 ml and 150 ml for soluble and insoluble collagen, respectively, aliquots of 25 ml (soluble collagen) and 10 ml (insoluble collagen) were transferred to graduated tubes and the volume adjusted to 50 ml. The total collagen was the sum of the soluble and insoluble fractions. The collagen solubility was expressed as total collagen percentage. The myofibrillar fragmentation index (MFI) was determined as described by Culler, Parrish, Smith, and Cross (1978). The myofibrillar suspension was used for the protein concentration measurement. A steak 2.5 cm thick was used for cooking losses and WBSF determination. After thawing at 0–4 °C for 24 h, the steaks were weighted, grilled (Modular 65/70 FTES electric griddle, Modular System Ltd., Italy) until achieved 70 °C of internal temperature, and weighted again for cooking losses determination. The internal temperature was controlled with a needle thermocouple probe (Lufft C120, Munich, Germany). After cooling to room temperature each sample provided a minimum of eight strips with a 1 cm2 cross section removed parallel to the muscle fibre orientation. The cores were sheared perpendicular to the longitudinal orientation of the muscle fibres, using a TA-TX Plus Texture Analyser (Stable Micro Systems Ltd., Surrey, UK) equipped with a Warner–Bratzler shear blade. The maximum shear force corresponding to the highest peak of the curve was measured in kg. The steaks for sensory analysis were thawed and cooked following the same procedure of WBSF measurements. Every sample was cut into 2 × 2 × 2 cm3 cores and maintained at 60 °C in heated plaques until tasted. The seven trained panellists assessed a profile composed by tenderness, juiciness, flavour and overall acceptability scoring the samples on a ten-point structured line scale.

A.C.G. Monteiro et al. / Meat Science 94 (2013) 63–68

65

2.3. Statistical analysis

3. Results

The beef type effect was studied by analysis of variance using the PROC MIXED procedure of Statistical Analysis Systems (SAS) software package, version 9.1 (Statistical Analysis System, 2004). The data was checked for normality and homoscedasticity. For some variables significant differences of variances between groups were found and thus were analysed by PROC MIXED model allowing for variance heterogeneity. Variables expected to be deeply affected by the electric stimulation like T3, pH3, MFI, cooking losses, WBSF and sensory traits, were only assessed and used in the statistical analysis for the second slaughter period. For all the remaining variables, the full data set was used. The model used in the full dataset variables included the fixed effects of meat type, period and their interaction. For those variable available only for the second slaughter period, a simplified model with only meat type effects was applied. The use of CW as covariate was tested and retained in the model when the slope was significantly different from zero. Variables analysed using CW as covariate were T3, total pigment and intramuscular free fat content. In order to avoid adjustment to non-biological and non-commercial overall CW mean, the veal data was adjusted to the means of calves CW (165 kg) and the beef data was adjusted to the mean bull CW (271 kg), according to the procedure described by Milliken and Johnson (2002). The relationship between the variables in each meat type was explored with the Pearson's correlation though using only the second slaughter period. Whenever a common correlation pattern was observed within all meat types, means were standardised and the pooled relationship between variables calculated. The canonical discriminant analysis was applied to the data in order to distinguish beef types. Variable selection for discriminant analysis was achieved using the significant variables after PROC MIXED analysis. The stepwise discriminant analysis (PROC STEPDISC) (SAS, 2004) selected the variables with higher discriminant ability. The linear discriminant functions were developed using the PROC DISCRIMINANT (SAS, 2004) to determine the coefficients of the meat quality variables that maximise the differences between the three meat groups.

3.1. Physicochemical characteristics and sensory attributes of meat The T3 and pH3 measurements were only assessed in the second slaughter period. Mertolenga-PDO beef resulted by definition from the oldest and heaviest animals at slaughter (Table 1), which resulted in a greater muscular mass leading to a slower temperature rate decline and hence to a higher T3, despite not being different from Mertolenga-PDO veal. Mertolenga-PDO beef presented 60% of the carcasses classified with O2 and the remaining 40% classified with R2 or R3. All the veal carcasses were classified with LO (male or female calves with more than 6 months of age). There were no significant differences between groups in pH3 of longissimus lumborum muscle, nevertheless Mertolenga-PDO veal and beef presented the highest pHu value (Table 3). The Table 3 shows the meat colour parameters with MertolengaPDO beef having the lowest L* value and the highest a* and C* values. On the contrary, the “Vitela Tradicional do Montado-PGI” veal presented the lowest a* and C* values. The “Vitela Tradicional do Montado-PGI” veal had the lowest and Mertolenga-PDO beef the highest values of pigment content (2.47 g/100 g DM), which was 41% and 57% higher than those values presented by the Mertolenga-PDO veal and “Vitela Tradicional do Montado-PGI” veal, respectively. No differences were found between meat types in dry matter content and intramuscular fat content. “Vitela Tradicional do Montado-PGI” veal had the lowest total collagen content, and the highest collagen solubility. Total collagen content was higher in the first slaughter period than in the second one. The values of myofibrillar fragmentation index were similar for all the analysed meat types, averaging 23.7. Mertolenga-PDO beef had the highest value of cooking losses and WBSF. The cooking losses of Mertolenga-PDO beef were 11% and 12% higher than the values presented by “Vitela Tradicional do Montado-PGI” and Mertolenga-PDO veal types, respectively. The value of Mertolenga-PDO beef WBSF was 25% and 31% higher than Mertolenga-PDO and “Vitela Tradicional do Montado-PGI” veal types, respectively.

Table 3 Physical chemical and sensory characteristics of longissimus lumborum muscle of “Vitela Tradicional do Montado”-PGI veal (VTM-PGI veal), Mertolenga-PDO veal and Mertolenga-PDO beef. VTM-PGI veal

T3 (°C)cd pH3d pHu L* a* b* h* C* Dry matter (%) Pigments (g/100 g DM) c Free fat (g/100 g DM) c Total collagen (g/100 g DM) Collagen solubility (%) MFId Cooking losses (%)d WBSF (kg)d Tendernessd Juicinessd Flavourd Overall acceptabilityd

Mertolenga-PDO veal

Mertolenga-PDO beef

Mean

SEM

Mean

SEM

Mean

SEM

16.02b 6.08 5.46b 35.38a 16.83c 2.12 6.86 17.03c 26.21 1.06c 2.34 2.46b 19.22a 22.28 22.51b 5.12b 5.08 4.12 3.70 4.48

0.624 0.053 0.048 0.376 0.270 0.260 0.762 0.283 0.156 0.141 0.125 0.070 0.559 2.047 0.734 0.435 0.364 0.321 0.230 0.360

18.27a 6.21 5.62a 34.23b 18.40b 2.01 5.97 18.58b 25.97 1.43b 2.36 2.92a 16.80b 25.79 22.34b 5.56b 6.03 4.79 3.75 5.31

0.624 0.053 0.049 0.380 0.273 0.263 0.771 0.287 0.158 0.143 0.127 0.071 0.556 2.047 0.743 0.435 0.355 0.313 0.230 0.351

19.89a 6.06 5.64a 30.50c 20.45a 2.29 5.82 20.70a 26.14 2.45a 2.68 3.04a 16.37b 23.01 25.25a 7.43a 4.96 4.44 3.58 4.59

0.638 0.054 0.050 0.395 0.283 0.273 0.800 0.298 0.164 0.148 0.132 0.074 0.587 2.093 0.751 0.445 0.355 0.313 0.224 0.351

S

*** 0.08 * *** *** ns ns *** ns ** ns *** *** ns * ** 0.08 ns ns ns

Statistical probability: ns, P > 0.05; *, P b 0.05; **, P b 0.01; ***, P b 0.001; means in the same row with different subscripts are significantly different; S = significance; SEM = standard error of the mean. T3 = muscle temperature measured 3 h post-mortem; pH3 = muscle pH measured 3 h post-mortem; pHu = muscle pH measured 6 days post-mortem; DM = dry matter; MFI = myofibrillar fragmentation index; WBSF = Warner–Bratzler shear force. c — variables in which the covariate (CW) was significant; d — variables determined only considering data from the second slaughter period.

66

A.C.G. Monteiro et al. / Meat Science 94 (2013) 63–68

The sensory tenderness value presented by Mertolenga-PDO veal, despite not statistically different from “Vitela Tradicional do Montado-PGI” veal and Mertolenga-PDO beef, tended (P = 0.08) to be higher. Considering juiciness, flavour and overall acceptability, the panellists considered the three meat types similar to each other, and scored them as slightly dry and slightly acceptable. Both veal types were scored as having moderately bland beef flavour, whilst PDO beef had slightly intense beef flavour. 3.2. Correlations between variables The correlation coefficients from age and carcass weight with the other variables studied were very similar. These two variables were highly correlated with a correlation coefficient of 0.86 (P b 0.001), thus their correlations with the other variables will be discussed together. There are several significant correlations between colour parameters and pigment content which correlated negatively with L* (Pb 0.001) and positively with a* (Pb 0.001) and C* (Pb 0.001), as well as with pHu which also correlated negatively with a* (Pb 0.05), b* (Pb 0.001), h* (Pb 0.001) and C* (Pb 0.01). The correlation coefficients among meat physicochemical characteristics and sensory attributes are shown in Table 4. WBSF was positively correlated with age, CW and cooking losses, and inversely correlated with MFI, tenderness, juiciness and overall acceptability. Sensory tenderness was negatively correlated with cooking losses and positively with juiciness, flavour and overall acceptability. Cooking losses explained a WBSF and sensory tenderness variation of 14% and 23%, respectively. Juiciness correlated positively with flavour and overall acceptability. Overall acceptability correlated negatively with cooking losses and positively with flavour. 3.3. Discriminant analysis The canonical discriminant analysis was applied to the meat quality variables in order to discriminate the beef types used in this study (Table 5), resulting in two discriminant functions. The discriminant functions obtained used eight variables, from which pHu, a* and C* parameters had the highest discriminant power in both roots. After cross-validation, results varied between 90.0% (Mertolenga-PDO veal and beef) and 100% of correct classification (“Vitela Tradicional do Montado-PGI” veal). The differences between meat types in quality variables selected by canonical discriminant analysis can be evaluated by the values of squared Mahalanobis distance (D 2). In line with this, the differences were smaller between “Vitela Tradicional do Montado-PGI” veal and Mertolenga-PDO veal (13.6), than between “Vitela Tradicional do Montado-PGI” veal and Mertolenga-PDO beef (21.7) or between Mertolenga-PDO veal and Mertolenga-PDO beef (18.0), as illustrated in Fig. 1. 4. Discussion All meat types presented pH3 (6.0–6.2; Chambaz, Scheeder, Kreuzer, & Dufey, 2003) and pHu values within the range considered normal (5.4 ≤ pHu ≤ 5.7; Young, Zhang, Farouk, & Podmore, 2005) for beef and veal. All the colour results are consistent with the differences in animal age, as well as with other studies that pointed an increase in muscle pigment content with age (Gil et al., 2001; Serra et al., 2008) especially up to 24 months (Renerre, 1986). After adjustment to age the differences between “Vitela Tradicional do Montado-PGI” and MertolengaPDO veal disappear, whilst the differences between the two veal types and Mertolenga-PDO beef remain significant, indicating an age effect on pigment content.

The colour parameters reflected the differences in the pigment content, as L* and a* were highly correlated with the pigment content. Despite the strong correlation of pHu with b* and h* parameters (Pb 0.001), and a moderate negative correlation with a* (Pb 0.05) and C* (Pb 0.01) parameters it seems that the pigment content had a greater influence on meat colour than the pHu, since Mertolenga-PDO veal and beef presented similar pHu value, but different colour parameters. These differences remained even after data standardisation. The results in colour parameters suggest that Mertolenga-PDO beef was darker and had a more intense and saturated colour, which is in agreement with other authors (Lawrie, 1998; Serra et al., 2004), who found similar results in older animals. Considering a previous study of our team relating consumer visually appreciation of beef colour (Banović et al., 2010) with colour instrumental measurements, we realised that despite the colour differences between the meat types, the values presented by all of them were within the colour range that have a great acceptability by consumers (Monteiro, 2012). All meat types had low intramuscular fat content (0.60–0.69 g/100 g muscle), which is an advantage from the consumers' health point of view, but can also compromise the eating quality. Intramuscular free fat as determined in our work did not include phospholipids, which explains in part the low fat levels found. Alasnier, Rémignon, and Gandemer (1996) stated a 0.7–0.9 g/100 g muscle contribution from phospholipids to total muscle fat content, which would double the value presented in our study. These values are lower than those reported by Monteiro (2003) (3.32 g/100 g free fat in DM) in Mertolenga-PDO beef, but the animals from this study were older and heavier, characteristics that are known to increase the intramuscular fat content (Vestergaard et al., 2000. Several authors suggested that intramuscular fat content improves meat tenderness (Hocquette et al., 2010; Savell & Cross, 1988). Unexpectedly, Mertolenga-PDO beef intramuscular free fat content was positively, although weakly, correlated with WBSF which is in contradiction with the aforementioned authors. Nevertheless, in both veal types intramuscular free fat content and WBSF were not correlated. However, as already stated the intramuscular free fat content of the meat types in the present study was very low and within a short range of values, thus it is unlikely that intramuscular fat content would have a significant effect on shear force. We would expect Mertolenga-PDO veal to have higher WBSF value than “Vitela Tradicional do Montado-PGI” veal, as Mertolenga-PDO veal presented lower collagen solubility than the other two meat types and higher collagen content than “Vitela Tradicional do Montado-PGI” veal, and several authors have reported that collagen content and solubility, as well as myofibrillar fragmentation post-mortem affect meat shear force value (Bailey & Light, 1989). Considering our study we concluded that none of the collagen measurements accounted for WBSF or sensory tenderness variations. These results are in line with other studies reporting that the collagen content is neither related with WBSF (Christensen et al., 2011; Hunsley, Vetter, Kline, & Burroughs, 1971) nor with taste panel tenderness (Hunsley et al., 1971), as in muscles with low collagen content, like longissimus, it might provide a limited contribution to background toughness in comparison with myofibrillar toughness (Ngapo et al., 2002). Moreover, the low collagen content may induce a lower technical precision in the measurement of total and soluble collagen amounts (Listrat & Hocquette, 2004). All meat types presented similar MFI and this parameter was negatively correlated with WBSF (Savell & Cross, 1988; Silva et al., 1999). However, MFI only accounted for 10% of the total variation in WBSF. Our results suggest that WBSF was more influenced by cooking losses than by other physicochemical characteristics. Likewise sensory tenderness was more influenced by cooking losses than by the remaining studied variables (Destefanis, Barge, Brugiapaglia, & Tassone, 2000; Serra et al., 2008; Silva et al., 1999). WBSF and sensory tenderness were moderately correlated which was in agreement with the majority of the research published

A.C.G. Monteiro et al. / Meat Science 94 (2013) 63–68

67

Table 4 Pearson's correlation coefficients between sensory attributes and physical and chemical meat quality characteristics of muscle longissimus lumborum from “Vitela Tradicional do Montado”-PGI veal, Mertolenga-PDO veal and Mertolenga-PDO beef.

WBSF Tenderness Juiciness Flavour OA

Age

CW

CL

MFI

WBSF

Tenderness

Juiciness

Flavour

0.41⁎⁎⁎ −0.18 −0.00 −0.06 −0.09

0.44⁎⁎⁎ −0.21 −0.07 −0.16 −0.16

0.48⁎⁎⁎ −0.37⁎⁎ −0.22 −0.02 −0.31⁎

−0.31⁎ −0.03 −0.15 −0.03 −0.09

– −0.38⁎⁎ −0.36⁎⁎ −0.22 −0.40⁎⁎

– – 0.72⁎⁎⁎ 0.42⁎⁎⁎ 0.90⁎⁎⁎

– – – 0.44⁎⁎⁎ 0.89⁎⁎⁎

– – – – 0.50⁎⁎⁎

Statistical probability of correlation coefficients: no superscript, P > 0.05. CW = carcass weight; CL = cooking losses; MFI = myofibrillar fragmentation index; WBSF = Warner–Bratzler shear force; OA = overall acceptability. ⁎ P b 0.05. ⁎⁎ P b 0.01. ⁎⁎⁎ P b 0.001.

(e.g. Chambaz et al., 2003; Destefanis et al., 2000; Silva et al., 1999; Vestergaard et al., 2000). Moreover, WBSF was inversely correlated with overall acceptability (Destefanis et al., 2000), indicating the importance of hardness/tenderness to beef acceptability. The values obtained for WBSF indicate that “Vitela Tradicional do Montado-PGI” veal and Mertolenga-PDO veal are tenderer than Mertolenga-PDO beef (Table 3). The WBSF value presented by the two veal types (averaging 5.36 kg) is below 5.5 the upper limit beneath which beef is considered to have a large acceptability by the Portuguese consumers (Simões & Lemos, 2005). Nevertheless, the value presented by Mertolenga-PDO beef is above that value, indicating a lower acceptability by the Portuguese consumers. Other authors refer 6.12 kg as the threshold separating tender and tough meat (Shackelford, Wheeler, & Koohmaraie, 1997), which would indicate Mertolenga-PDO beef as a tough meat. This appears to be due to an insufficient ageing period, which is still a problem in the Portuguese meat industry. Beef ageing is frequently not enough which is reflected in high WBSF values and consumer dissatisfaction. Studies concerning meat quality in samples of the market place and their relation with consumer expectations have been overlooked. This is particularly important in beef marketed with quality labels, as PDO and PGI, which is more expensive than regular beef and is considered by the consumer has having higher quality (Banović et al., 2010). Panellists considered the three meat types similar in tenderness, juiciness, flavour and overall acceptability. All these sensory attributes were correlated with overall acceptability, as stated by others (Monsón, Sañudo, & Sierra, 2005). However, tenderness and juiciness had a greater correlation coefficient, indicating these two attributes are the main contributors to the overall acceptability, which was expectable since consumers' acceptability is largely based on sensory tenderness (Destefanis et al., 2000) and tenderness and juiciness are

highly correlated. This relation has been reported previously by several authors (e.g. Serra et al., 2008; Silva et al., 1999), with correlation coefficients ranging from 0.58 to 0.88. This relationship may be explained by the fact that the tender the meat is more quickly juices are released by chewing, which gives the sensation of a juicier meat (Savell & Cross, 1988). The correlations observed between the sensory attributes could also be due to the “halo effect”, i.e. when an attribute is enhanced by other characteristics of the product (Gill et al., 2010). According to the canonical discriminant analysis the two veal types are more similar to each other than to the Mertolenga-PDO beef. Our team compared the results of the discriminant analysis from the same samples of these meat types but having the lipid composition as discriminant variable and realised that the distance between the veal types were likewise smaller and that Mertolenga-PDO beef was closer to Mertolenga-PDO veal than to “Vitela Tradicional do Montado-PGI” veal (Monteiro et al., 2012). The first function (root 1) discriminates both veal, having the “Vitela Tradicional do Montado-PGI” veal negative loadings and the Mertolenga-PDO veal positive ones (Fig. 1). The pHu, a* and C* parameters had the highest discriminant power in both roots to discriminate the three meat types. Considering the discriminant analysis performed just on the two veal types, the variables that discriminate both veal were pHu and total collagen in both roots, T3 in the first root and overall acceptability in the second one. Our results clearly indicate that despite the small differences obtained in the physicochemical characteristics and the similarity of the production systems of the animals it was feasible to allocate meat samples into one of the three meat types with good accuracy. 5. Conclusion The study of the three meat categories involving the Mertolenga breed has shown that Mertolenga-PDO beef is darker and redder

Table 5 Results of canonical discriminant analysis: loadings of correlation matrix between predictor variables (standardised canonical coefficients) and discriminant functions (roots 1 and 2), and some statistics for each function.

Variables pHu a* L* C* GI Total collagen WBSF Overall acceptability Statistics Canonical R Eigenvalue Cumulative proportion Probability

Root 1

Root 2

−3.3668 −3.3346 −0.5776 3.4423 −1.8047 1.0654 0.2747 0.0943

5.9908 5.0445 0.4065 −4.4586 −0.8787 1.6950 0.0553 0.2516

0.874 3.968 0.636 b0.0001

0.825 2.275 1.000 b0.0001

5 4 3 2 1 -4

-2

0 -1 0

2

4

6

-2 -3 -4 -5

VTM-PGI veal

Mertolenga-PDO veal

Mertolenga-PDO beef

Fig. 1. Plot of the canonical discriminant analysis of “Vitela Tradicional do Montado”-PGI veal, Mertolenga-PDO veal and Mertolenga-PDO beef according to physical, chemical and sensory characteristics.

68

A.C.G. Monteiro et al. / Meat Science 94 (2013) 63–68

than "Vitela Tradicional do Montado"-PGI and Mertolenga-PDO veal. However, despite being darker the colour is still within the Portuguese consumer range of acceptability. Both veal types are medium tender and considered to have a great acceptability by the Portuguese consumer. Mertolenga-PDO beef is tougher than the two veal types and so less acceptable, indicating an insufficient ageing period. The ageing period must be increased in the Portuguese industry in order to improve beef quality. Cooking losses was the physicochemical characteristic that contributed more to explain the hardness/tenderness differences observed between meat types. The differences observed between the two veal types and Mertolenga-PDO beef can be largely explained by the age and/or carcass weight differences. The three meat types were discriminated based on pHu, a* and C* parameters, whist the discrimination between the two veal types was based on pHu, total collagen, T3 and overall acceptability. Acknowledgements The financial support from the Ministry of Agriculture (AGRO/2004/ 422) and the individual grant from FCT to Ana C.G. Monteiro (SFRH/BD/ 31091/2006) are acknowledged. References Alasnier, C., Rémignon, H., & Gandemer, G. (1996). Lipid characteristics associated with oxidative and glycolytic fibres in rabbit muscles. Meat Science, 43, 213–224. Alfaia, C. M. M., Quaresma, M. A. G., Castro, M. L. F., Martins, S. I. V., Portugal, A. P. V., Fontes, C. M. G. A., Bessa, R. J. B., & Prates, J. A. M. (2006). Fatty acid composition including isomeric profile of conjugated linoleic acid and, cholesterol in Mertolenga-PDO beef. Journal of the Science of Food and Agriculture, 86, 2196–2205. AOAC 945.16 (2000). Oil in cereals adjuncts. Petroleum ether extraction method 945.16. Official methods of analysis (pp. 31) (17th ed.). Gaithersburg, MA, USA: Association of Official Analytical Chemists International (chapter 27). AOAC 990.26 (2000). Hydroxyproline in meat and meat products. Colorimetric method 990.26. Official methods of analysis (pp. 13) (17th ed.). Gaithersburg, MA, USA: Association of Official Analytical Chemists International (chapter 39). Bailey, A. J., & Light, N. D. (1989). Connective tissue in meat and meat products. London e New York: Elsevier Applied Science. Banović, M., Grunert, K., Barreira, M. M., & Fontes, M. A. (2010). Consumers' quality perception of national branded, national store branded, and imported store branded beef. Meat Science, 84, 54–65. Calkins, C. R., & Hodgen, J. M. (2007). A fresh look at meat flavor. Meat Science, 77, 63–80. Chambaz, A., Scheeder, M. R. L., Kreuzer, M., & Dufey, P. A. (2003). Meat quality of Angus, Simmental, Charolais and Limousin steers compared at the same intramuscular fat content. Meat Science, 63, 491–500. Christensen, M., Ertbjerg, P., Failla, S., Sañudo, C., Richardson, R. I., Nute, G. R., Olleta, J. L., Panea, B., Albertí, P., Juárez, M., Hocquette, J. F., & Williams, J. L. (2011). Relationship between collagen characteristics, lipid content and raw and cooked texture of meat from young bulls of fifteen European breeds. Meat Science, 87, 61–65. Costa, P., Roseiro, L. C., Bessa, R. J. B., Padilha, M., Partidário, A., Almeida, J. M., Calkins, C. R., & Santos, C. (2008). Muscle fiber and fatty acid profiles of Mertolenga-PDO meat. Meat Science, 78, 502–512. Culler, R. D., Parrish, F. C., Jr., Smith, G. C., & Cross, H. R. (1978). Relationship of myofibril fragmentation index to certain chemical, physical and sensory characteristics of bovine longissimus muscle. Journal of Food Science, 43, 1177–1180. Destefanis, G., Barge, M. T., Brugiapaglia, A., & Tassone, S. (2000). The use of principal component analysis (PCA) to characterize beef. Meat Science, 56, 255–259. Gabinete de Planeamento e Políticas (2009). Anuário Pecuário 2008/2009. (Lisboa, Portugal). Gil, M., Serra, X., Gispert, M., Oliver, M.Á., Sañudo, C., Panea, B., Olleta, J. L., Campo, M., Oliván, M., Osoro, K., Garcís-Cachán, M. D., Cruz-Sagredo, R., & Piedrafita, J. (2001). The effect of breed-production systems on the myosin heavy chain 1, the biochemical characteristics and the colour variables of Longissimus thoracis from seven Spanish beef cattle breeds. Meat Science, 58, 181–188. Gill, J. L., Matika, O., Williams, J. L., Worton, H., Wiener, P., & Bishop, S. C. (2010). Consistency statistics and genetic parameters for taste panel assessed meat quality traits and their relationship with carcass quality traits in a commercial population of Angus-sired beef cattle. Animal, 4, 1–8. Guerreiro, L. (2001). Marketing PDO (products with geographical denominations of origin) and PGI (products with geographical identities). In L. Frewer, E. Risvik, &

H. Schifferstein (Eds.), Food, people and society. A European perspective of consumers' food choices (pp. 281–296)Heidelberg: Springer Verlag. Hocquette, J. F., Gondret, F., Baéza, E., Médale, F., Jurie, C., & Pethick, D. W. (2010). Intramuscular fat content on meat-producing animals: development, genetic and nutritional control, and identification of putative markers. Animal, 4, 303–319. Hunsley, R. E., Vetter, R. L., Kline, E. A., & Burroughs, W. (1971). Effects of age and sex on quality, tenderness and collagen content of bovine longissimus muscle. Journal of Animal Science, 33, 933–938. Larraín, R. E., Schaeffer, D. M., & Reed, J. D. (2008). Use of digital images to estimate CIE color coordinates of beef. Food Research International, 41, 380–385. Lawrie, R. A. (1998). The eating quality of meat. In R. A. Lawrie (Ed.), Meat Science (pp. 212–257) (6th ed.). Oxford, England: Pergamont Press. Listrat, A., & Hocquette, J. F. (2004). Analytical limits of total and insoluble collagen content measurements and of type I and III collagens analysis by electrophoresis in bovine muscles. Meat Science, 68, 127–136. Milliken, G. A., & Johnson, D. A. (2002). Analysis of covariance. In: Chapman & Hall (Ed), Analysis of messy data (vol. 3). New York, USA: CRC Press Company. Monsón, F., Sañudo, C., & Sierra, I. (2005). Influence of breed and ageing time on the sensory meat quality and consumer acceptability in intensively reared beef. Meat Science, 71, 471–479. Monteiro, A. C. G. (2003). Qualidade da carne de bovinos em sistemas de pastoreio: Efeito da castração e da duração do período de acabamento. Dissertação de Mestrado. Universidade Técnica de Lisboa. Lisboa, Portugal. Monteiro, A. C. G. (2012). Relationship between Portuguese consumer preferences, textural properties, chemical composition and nutritional value of beef. Tese de Doutoramento em Ciências Veterinárias. Faculdade de Medicina Veterinária.: Universidade Técnica de Lisboa. Monteiro, A. C. G., Fontes, M. A., Bessa, R. J. B., Prates, J. A. M., & Lemos, J. P. C. (2012). Intramuscular lipids of Mertolenga-PDO beef, Mertolenga-PDO veal and “Vitela Tradicional do Montado”-PGI veal. Food Chemistry, 182, 1486–1994. Ngapo, T. M., Berge, P., Culioli, J., Dransfield, E., De Smet, S., & Claeys, E. (2002). Perimysial collagen crosslinking and meat tenderness in Belgian blue double muscle cattle. Meat Science, 61, 91–102. Council Regulation (EC) nº 510/2006 of 20 March 2006 on the protection of geographical indications and designations of origin for agricultural products and foodstuffs. Official Journal, L93 (2006), 0012–0031 (31/03/2006). Renerre, M. (1986). Influence da facteurs biologiques et technologiques sur la couleur de la viande. Bulletin Technologique C. R. Z. V. Theix, INRA, 65. (pp. 41–45). Savell, J. W., & Cross, H. R. (1988). The role of fat in the palatability of beef, pork, and lamb. In: NRC (Ed.), Designing foods. Animal product options in the marketplace (pp. 345–355). Washington D.C.: National Academy Press. Serra, X., Gil, M., Gispert, M., Guerrero, L., Oliver, M. A., Sañudo, C., Campo, M. M., Panea, B., Olleta, J. L., Quintanilla, R., & Piedrafita, J. (2004). Characterisation of young bulls of Bruna dels Pirineus cattle breed (selected from old Brown Swiss) in relation to carcass, meat quality and biochemical traits. Meat Science, 66, 425–436. Serra, X., Guerrero, L., Guàrdia, M. D., Gil, M., Sañudo, C., Panea, B., Campo, M. M., Olleta, J. L., Piedrafita, J., Graciá-Cachán, J., & Oliver, M. A. (2008). Eating quality of young bulls from three Spanish beef breed-production systems and its relationships with chemical and instrumental meat quality. Meat Science, 79, 98–104. Shackelford, S. D., Wheeler, T., & Koohmaraie, M. (1997). Tenderness classification of beef: I. Evaluation of beef longissimus shear force at 1 or 2 days post-mortem as a predictor of aged beef tenderness. Journal of Animal Science, 75, 2417–2422. Silva, J. A., Patarata, L., & Martins, C. (1999). Influence of ultimate pH on bovine meat tenderness during ageing. Meat Science, 52, 453–459. Simões, J. A., & Lemos, J. P. C. (2005). Efeito do sexo, da idade e da escolaridade dos consumidores na avaliação da tenrura da carne de bovino. Proceedings of the Congresso de Ciências Veterinárias. 13–15 Outubro 2005, Estação Zootécnica Nacional, Vale de Santarém, Portugal. Statistical Analysis System (2004). SAS/STAT 9.1 user's guide. Cary, NC, USA: SAS Institute Inc. Vestergaard, M., Therkildsen, M., Henckel, P., Jensen, L. R., Andersen, H. R., & Sejrsen, K. (2000). Influence of feeding intensity, grazing and finishing feeding on meat and eating quality of young bulls and the relationship between muscle fibre characteristics, fibre fragmentation and meat tenderness. Meat Science, 54, 187–195. Wierbicki, E., Cahill, V. R., Kunkle, L. E., Klosterman, E. W., & Deatherage, F. E. (1955). Effect of castration on biochemistry and quality of beef. Journal of Agricultural and Food Chemistry, 3, 244–248. Xue, H., Mainville, D., You, W., & Nayga, R. M., Jr. (2010). Consumer preference and willingness to pay for grass-fed beef: Empirical evidence in-store experiments. Food Quality and Preference, 21, 857–866. Young, O. A., Zhang, S. X., Farouk, M. M., & Podmore, C. (2005). Effects of pH adjustment with phosphates on attributes and functionalities of normal and high pH beef. Meat Science, 70, 133–139.