Do sarcomere length, collagen content, pH, intramuscular fat and desmin degradation explain variation in the tenderness of three ovine muscles?

Meat Science 113 (2016) 51–58

Contents lists available at ScienceDirect

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

Do sarcomere length, collagen content, pH, intramuscular fat and desmin degradation explain variation in the tenderness of three ovine muscles? Colin.P. Starkey a,b,⁎, Geert.H. Geesink b, Damian Collins c, V. Hutton Oddy d, David L. Hopkins e a

Cooperative Research Centre for Sheep Industry Innovation, Armidale, New South Wales, Australia University of New England, Armidale, New South Wales, Australia NSW Department of Primary Industries, Elizabeth MacArthur Agricultural Institute, Woodbridge Rd, Menangle, New South Wales, Australia d NSW Department of Primary Industries, Beef Industry Centre, UNE, Armidale, New South Wales, Australia e NSW Department of Primary Industries, Centre for Red Meat and Sheep Development, Cowra, New South Wales, Australia b c

a r t i c l e

i n f o

Article history: Received 18 August 2015 Received in revised form 5 November 2015 Accepted 17 November 2015 Available online 18 November 2015 Keywords: Lamb Tenderness Desmin Collagen Sarcomere length

a b s t r a c t The longissimus (n = 118) (LL), semimembranosus (n = 104) (SM) and biceps femoris (n = 134) (BF) muscles were collected from lamb and sheep carcases and aged for 5 days (LL and SM) and 14 days (BF) to study the impact of muscle characteristics on tenderness as assessed by shear force (SF) and sensory evaluation. The impact of gender, animal age, collagen content, sarcomere length (SL), desmin degradation, ultimate pH and intramuscular fat (IMF) on tenderness was examined. The main factors which influenced SF of the LL were IMF, SL and desmin degradation, but for sensory tenderness, IMF, ultimate pH and gender were the main factors. The SF and sensory tenderness of the SM was best predicted by the degree of desmin degradation. For the BF soluble collagen and animal age both influenced SF. Different factors affect tenderness across muscles and not one prediction model applied across all muscles equally well. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction The eating quality of meat is determined by the tenderness, juiciness and flavour (Thompson, 2002). Of these traits, tenderness is affected by both production and processing factors (Young, Hopkins, & Pethick, 2005). Since there is variation in the tenderness of lamb at the retail level (Safari, Channon, Hopkins, Hall, & van de Ven, 2002), it is important to understand what interactions occur within and across different muscles, and how these interactions influence the variation in tenderness. There are three major factors which can impact on the tenderness of meat; these include collagen content and solubility (background toughness), muscle shortening (toughening) and ageing (tenderisation) (Hopkins & Geesink, 2009). The latter two factors (muscle shortening and ageing) take place during post-mortem storage (Hopkins & Thompson, 2001). The impact of sarcomere length has been researched extensively (Hopkins & Thompson, 2001; Rhee, Wheeler, Shackelford, & Koohmaraie, 2004; Smulders, Marsh, Swartz, Russell, & Hoenecke, 1990; Wheeler & Koohmaraie, 1994). As such Rhee et al. (2004) found that overall sarcomere length was significantly correlated to tenderness. The impact of ageing (proteolysis) on tenderness is thought to be due to weakening of the myofibrillar structure as result of degradation of muscle proteins, like titin, nebulin and desmin (Hopkins & Thompson, 2002). ⁎ Corresponding author at: Cooperative Research Centre for Sheep Industry Innovation, Armidale, New South Wales, Australia. E-mail address: [email protected] (C.P. Starkey).

http://dx.doi.org/10.1016/j.meatsci.2015.11.013 0309-1740/© 2015 Elsevier Ltd. All rights reserved.

A number of methodologies have been used to study proteolysis including the study of specific proteins (Hopkins & Thompson, 2002). One such protein of interest is desmin, because it is a calpain substrate (Geesink, Kuchay, Chishti, & Koohmaraie, 2006; Huff-Lonergan & Lonergan, 2005), and because it is important to the function and integrity of muscle cells. In a study conducted by Starkey, Geesink, Oddy, and Hopkins (2015) desmin degradation was the most important factor for explaining variation in shear force over different ageing periods in lamb longissimus. Connective tissue (i.e. collagen) content and solubility impact on tenderness and this is referred to as background toughness (Veiseth, Shackelford, Wheeler, & Koohmaraie, 2004). The solubility of collagen is affected by a number of factors such as animal age (Young & Braggins, 1993), sex and muscle type (Wheeler, Shackelford, & Koohmaraie, 2000), and this leads to a variation in tenderness. The extent of this effect varies however, according to the muscle under study. For the longissimus Warner et al. (2010) suggested that total collagen content is of limited value when predicting tenderness, whereas the level of soluble collagen would be expected to affect tenderness. For lamb (ovine) there are no published studies which have examined the effect of indices of muscle structure and degradation across a number of muscles. The study of Starkey et al. (2015) only focussed on the tenderness of the longissimus muscle. For other species such as pigs, Wheeler et al. (2000) reported that sarcomere length, total collagen and proteolysis (quantified as desmin degradation) when combined could explain more than 50% of the variation in sensory tenderness across 5 muscles.

52

C.P. Starkey et al. / Meat Science 113 (2016) 51–58

This was for muscle aged for 1 day and the relationship varied according to muscle. Thus, the objective of the experiment detailed in this paper was to examine the effect of sarcomere length, collagen content, pH, intramuscular fat and indicators of proteolysis on the variation in lamb meat tenderness (objectively and subjectively measured) for three different lamb muscles: longissimus thoracis et lumborum (loin), semimembranosus (topside) and biceps femoris (silverside). 2. Materials and methods 2.1. Animal background and slaughter Two hundred and thirty one mixed sex lambs and 28 male hoggets (first permanent incisor) from the Sheep CRC Information Nucleus flock (van der Werf, Kinghorn, & Banks, 2010) were sampled at slaughter in 2011. The animals were bred at the UNE Kirby farm (Armidale NSW Australia) and slaughtered from April to August in 4 separate kill groups at a processing plant in Tamworth, NSW. The lambs were offspring of 10 different sire breeds which included Merino, maternal and terminal sire types. The hoggets (castrated males) were the progeny of 16 different sires, all Merino, joined to Merino dams, and were all slaughtered as part of the third lamb slaughter. The slaughter process consisted of electrical stunning, followed by exsanguination aided by a pre-dressing medium voltage electrical stimulation system. Carcases were trimmed to AUS-MEAT specifications (Anonymous, 2005) and weighed to determine hot carcase weights (HCW), and chilled for 24 h at 2 °C. The initial pH and temperature recordings were taken as soon as the first of the carcases entered the chiller. This was conducted on the left hand side of the m. longissimus lumborum at the caudal end over the lumbar sacral junction, as described by Hopkins et al. (2011). Three subsequent measures were recorded over the next 3 h with approximately 50 min between measurements. These measurements were used to calculate the Temperature@pH 6 and pH@Temperature18 values as per van de Ven, Pearce, and Hopkins (2014). The ultimate pH was measured approximately 24 h post-mortem, on the same section of muscle. Muscle pH was recorded using a glass combination pH probe (potassium chloride) (Ionode intermediate junction pH electrode TPS, Pty Ltd., Brisbane) attached to a data recording metre (TPS WP-40). The temperature data was recorded using a steel probe attached to the same metre. The pH metre was calibrated before each set of recordings by using buffers at pH 4 and 6.88 at room temperature (approximately 20 °C) and at 24 h at chiller temperature (approximately 5 °C). 2.2. Sampling The hind quarter (leg and aitch bone) was separated from the saddle and rack and the forequarter (shoulder), as described by Pannier, Gardner et al. (2014). The left and right topsides were removed from the hind legs. Then the gracillus and adductor muscles were removed to leave the semimembranosus (SM) which was vacuum packed. The semimembranosus from the left legs were used for subsequent sensory testing, these being a subset of those reported on by Pannier, Gardner et al. (2014). The semimembranosus from the right leg was used for subsequent shear force testing. Both samples were aged for 5 days at 2 °C then frozen at − 22 °C. Samples were taken after 5 days of ageing for determination of collagen content (20 g), sarcomere length measurement and measurement of desmin degradation (5 g combined) then frozen at − 22 °C. The biceps femoris muscle was removed from the right side and prepared into a 65 g shear force block which was aged for 14 days at 2 °C. Samples were also collected for determination of collagen content (20 g), sarcomere length measurement and measurement of desmin degradation (5 g combined) after 14 days of ageing and then frozen at −22 °C. The right side longissimus lumborum was boned out from the 12/13th rib to the lumbar sacral junction (Pannier, Gardner et al., 2014). The

subcutaneous fat and epimysium were removed from the longissimus, which was then divided into 3 sections. The caudal end was used for shear force testing (65 g block) after 5 days of ageing at 2 °C. The cranial end of the longissimus was used for intramuscular fat (IMF) (20 g) analysis, collagen content determination (20 g), sarcomere length measurement and measurement of desmin degradation (combined 5 g). A portion for sensory testing was prepared from the left side longissimus from the 5th rib to the lumbar sacral junction and was a subset from the study conducted by Pannier, Gardner et al. (2014). After the muscle samples were aged for the allotted 5 days, they were frozen and stored at −22 °C until testing. The same 3 muscles (semimembranosus, longissimus and biceps femoris) were removed from the hogget carcases as described. The samples collected from the carcases were the same as from the lamb samples, with; longissimus (shear force, sensory tenderness, sarcomere length, IMF, collagen and desmin), semimembranosus (shear force, sensory tenderness, sarcomere length, collagen and desmin) and biceps femoris (shear force, sarcomere length, collagen and desmin) samples collected. After determination of the shear force, one hundred and eighteen samples (103 lamb, 15 hogget) from the longissimus, 104 (90 lamb, 14 hogget) from the semimembranosus and 134 samples (119 lamb, 15 hogget) from the biceps femoris were selected for detailed analysis. Samples were selected to allow for an even distribution from all three muscles from the four slaughters across the range in shear force to ensure there was sufficient variation to develop models. 2.3. Shear force determination There were two different methods used by 3 different laboratories to assess shear force. The first method used for the determination of shear force (SF) was described by Starkey et al. (2015). In this method samples were cooked in a 70 °C water bath for a period of 30 min and stored overnight at 3–4 °C before shear force determinations. This was used for the biceps femoris samples only. The second method was described by Hopkins, Toohey, Warner, Kerr, and van de Ven (2010) and in this method longissimus and semimembranosus samples were cooked in a 71 °C water bath for 35 min and stored overnight at 3–4 °C. The blocks were approximately 65 mm long, 43 mm wide and 23 mm high and weighed approximately 65 g. All shear force measurements were conducted on a Lloyd Instruments LRX Materials Testing Machine fitted with a 500 N load cell (Lloyd Instruments Ltd., Hampshire UK). The machine setups were slightly different, with the laboratory which tested the biceps femoris using the method described by Starkey et al. (2015) where a straight edged blade moved upward at 100 mm/min. The other two laboratories used the method described by Hopkins et al. (2010) for measurement of the longissimus and semimembranosus samples. All laboratories used the following procedure with; six subsamples were tested with a rectangular cross section of 1 cm2. The fibre direction ran parallel to the length of the sample and at right angles to the shearing surface. The amount of force required to cut through the fibres was measured as peak force in Newtons. The average of 6 subsamples was recorded. 2.4. Measurement of sarcomere length, collagen content and desmin degradation The method for the determination of sarcomere length (SL) was similar to that previously described by Cross, West, and Dutson (1981), with the details given by Starkey et al. (2015). The method to determine total and soluble collagen was derived from AOAC method 990.26 (AOAC, 2000) as previously described by Starkey et al. (2015). Desmin degradation was determined by using SDS-PAGE and Western blotting methods described by Geesink, Mareko, Morton, and Bickerstaffe (2001) with full detail provided by Starkey et al. (2015).

C.P. Starkey et al. / Meat Science 113 (2016) 51–58

2.5. Intramuscular fat measurement The method used to determine the intramuscular fat (IMF) level was similar to that described by Perry, Shorthose, Ferguson & Thompson (2001). The only variation in the method was that the percentage of the IMF was calculated on a wet weight basis using near infrared spectroscopy as described by Pannier, Pethick et al. (2014). 2.6. Sensory testing Sampling for sensory testing was conducted as per the Meat Standards Australia (MSA) sensory evaluation system as described by Pannier, Gardner et al. (2014) with all samples tested by 10 untrained consumers. Briefly, 5 slices (15 mm thick) were halved to form 10 portions and these were grilled on a Silex griller until cooked to a medium degree of doneness with an internal temperature of 65 °C. The portions were then rested for 2 min before being evaluated by panellists. These samples were assessed as part of the study by Pannier, Gardner et al. (2014), using the same methodology for assessing sensory samples. Each session had 60 untrained consumers testing 36 samples and being a subset of the Pannier, Gardner et al. (2014) study, the samples were assessed in 62 sessions obtaining 10 different consumer responses per sample and providing an evaluation of the measured traits out of a score of 100. These traits include; tenderness, overall liking, juiciness, flavour and odour. The portions were then graded using a star system with the following categories; unsatisfactory (1–2 stars), good everyday (3 star), better than everyday (4 star) and premium (5 star). An average was taken of the 10 responses for statistical analysis. The panellists were given a starter portion (not part of the experiment) before receiving the experimental portions which included 3 from the longissimus and 3 from the semimembranosus muscles. Allocations of the samples across consumer sessions were made using a latin square design as described by Thompson et al. (2005). 2.7. Statistical analysis Mixed models were fitted using ASReml-R (Butler, 2009) within the R software environment (R Core Team, 2014) to develop prediction models for shear force and sensory traits. A log transformation was applied to desmin to adjust for skewness. A forward stepwise selection approach was used to determine a prediction model. At each step, the covariate with the lowest P-value from available covariates was added to the model if its P-value was less than 0.05, otherwise the process was stopped and the selected covariates were taken as the final model. Standard regression diagnostics were supplemented by added variable plots (as implemented in the car package in R) to help assess fit. The significance of fixed terms was determined using Wald-type F tests with Kenward-Rogers adjustments (Kenward and Roger (1997)) as implemented in ASReml. Significance of random terms was determined using a residual maximum likelihood ratio test (REMLRT), calculating the P-value following Stram and Lee (1994). A pseudo-R2 was calculated at each stage of the stepwise selection using the marginal approach of Nakagawa & Schielzeth (2013). For the overall analysis of shear force, a stepwise selection was performed with data for all muscles combined. The covariate terms that were fitted were sarcomere length, total collagen, soluble collagen, insoluble collagen, log(desmin), pH@Temp18, Temp@pH 6, and the fixed terms were muscle (longissimus, semimembranosus, biceps femoris), sex (wethers, ewes) and animal age (lambs, hoggets). The random terms were slaughter day, SFLab, sire, dam and individual animal (Model 1). In this case, SFLab is the laboratory used to test shear force. A second phase of stepwise selection was conducted to assess two way interactions of sex, age, muscle with the covariates, using the final model from the previous stepwise selection as the base model.

53

A stepwise selection was also performed for each of the three muscles separately for the analysis of shear force. The covariates were sarcomere length, total collagen, soluble collagen, insoluble collagen, log(desmin), pH@Temp18, Temp@pH 6, and the fixed terms were sex and animal age. Random terms were slaughter day, SFLab, sire, dam, and individual animal (Model 2). For the longissimus muscle, the covariates ultimate pH and IMF were also included as potential covariates only for the longissimus. The sensory data analysis was similar to the SF analysis. The covariates were sarcomere length, total collagen, soluble collagen, insoluble collagen, log(desmin), pH@Temp18, Temp@pH 6, and the fixed terms were muscle, sex, and animal age. The random terms were panellist + slaughter day, sire, dam + individual animal (Model 3). A stepwise selection was also performed for sensory traits for each of the muscles separately. The covariates were sarcomere length, total collagen, soluble collagen, insoluble collagen, log(desmin), pH@Temp18, Temp@pH 6, and the fixed terms were sex and animal age. The random terms were panellist + slaughter day, sire, dam and individual animal (Model 4). For the longissimus muscle, the covariates ultimate pH and IMF were also included as potential covariates. The final model investigated was the correlation between sensory tenderness and shear force using the model (Model 5) below: Sensory tenderness = SF + muscle + sex + slaughter day + SFLab + sire + dam The random terms fitted are in italics. 3. Results 3.1. Muscle measurements As shown in Table 1, there was considerable variation in tenderness (shear force and sensory tenderness) and related traits within and between muscles. 3.2. The relationship between shear force, sensory tenderness and meat quality traits within muscles When analysed separately within muscles (Model 1), it was found for the longissimus that shear force was significantly affected by IMF (P = 0.003), sarcomere length (P = 0.001) and log desmin (P = 0.015) with a combined pseudo R2 of 14.0% (Table 2, note all coefficients were negative). These results mean, for example, that an increase in IMF

Table 1 Unadjusted means, standard deviations and ranges for meat quality measures according to muscle (combined lamb and hogget data).

SF (N)

Sensory tenderness Total collagen (mg/g)

Soluble collagen (mg/g)

Desmin (Degraded/Intact)

Sarcomere length (μm)

IMF (%) Temp@pH 6 (°C) pH@Temp18 Ultimate pH

Muscle

Number

Mean ± SD

Range

LL SM BF LL SM LL SM BF LL SM BF LL SM BF LL SM BF LL LL LL LL

118 104 134 108 96 118 104 134 118 104 134 118 104 134 118 104 134 118 231 231 231

29.8 ± 11.6 48.1 ± 16.3 39.6 ± 9.7 72.6 ± 10.2 45.5 ± 12.4 13.0 ± 2.4 14.8 ± 3.2 18.3 ± 5.5 2.6 ± 0.7 2.1 ± 0.5 5.3 ± 2.3 2.3 ± 2.3 2.5 ± 1.1 3.5 ± 3.0 1.8 ± 0.2 1.6 ± 0.1 1.6 ± 0.1 4.1 ± 0.8 12.6 ± 6.9 6.18 ± 0.2 5.69 ± 0.1

13.4–70.2 19.8–87.2 22.4–61.7 39.6–91.9 23.1–74.1 6.92–20.9 9.06–23.4 9.50–40.6 0.61–5.31 1.02–3.42 1.18–9.83 0.08–14.7 0.77–7.08 1.07–19.9 1.40–2.16 1.43–2.07 1.34–2.11 2.64–6.24 0.90–34.1 5.68–6.59 5.50–6.81

54

C.P. Starkey et al. / Meat Science 113 (2016) 51–58

Table 2 Muscle traits which significantly affected shear force within individual muscles (regression coefficients, standard errors and probability level). Coefficient

std error

P-value

M. Longissimus Intercept IMF Sarcomere length Log(Desmin)

72.24 −3.86 −15.22 −2.92

11.74 1.20 5.37 1.12

0.001 0.003 0.001 0.015

M. Semimembranosus Intercept Log(Desmin)

57.18 −11.38

10.62 2.33

0.008 0.001

49.87 −1.07 −7.18

3.48 0.41 3.15

0.001 0.010 0.026

M. Biceps femoris Intercept Soluble collagen Animal age (days)

3.3. The relationship between shear force, sensory tenderness and meat quality traits across muscles The factors that were related to shear force across all muscles were sarcomere length, log desmin and soluble collagen (all P b 0.001). Each was negatively correlated with shear force (Table 4). Other factors, such as; slaughter day (P = 0.021), laboratory where shear force was tested (P b 0.001) and sire (P b 0.001) all had a significant effect on shear force across all muscles with a pseudo R2 of 30.0%. There was an interaction between muscle type and log desmin (P = 0.004; Fig. 1), such that the degradation of desmin in the semimembranosus (P b 0.001) and longissimus (P b 0.01) was related to shear force, but this Table 3 Muscle traits which significantly affected sensory tenderness within individual muscles (regression coefficients, standard errors and probability level).

M. Semimembranosus Intercept Log(Desmin)

All muscles

Coefficient

std error

P-value

Intercept Muscle SM Muscle BF Sarcomere length log(Desmin) Soluble collagen Sex (Male) Muscle SM:log(Desmin)a Muscle BF:log(Desmin)

54.45 23.44 14.60 −13.61 −2.89 −1.70 3.31 −6.99 1.51

10.06 2.30 12.05 4.01 1.13 0.39 1.38 2.42 1.91

0.002 0.001 0.292 0.001 0.011 0.001 0.018 0.004 0.430

a

from 3 to 4% would lead to a 3.9 N decrease in shear force. Other factors which contributed to the variation in shear force were the laboratory where shear force was tested (P b 0.001) and sire (P b 0.001). Low correlations were observed between the rate of pH and temperature decline (Temp@pH 6 and pH@Temp18) and shear force and neither had an effect (P N 0.05) on shear force. For the semimembranosus only log desmin (P = 0.001) was correlated with shear force (Table 2) with a pseudo R2 of 7.5%. Other significant factors were slaughter day (P b 0.001) and laboratory where shear force was tested (P b 0.001). For the biceps femoris the significant factors were soluble collagen (P = 0.01) and animal age (P = 0.026) (older animals were tougher) with a pseudo R2 of 18.4%. Sire (P = 0.008) and slaughter day (P = 0.031) were also significantly related to shear force in the biceps femoris. When analysed separately within muscles (Model 3), sensory tenderness of the longissimus was related to IMF (P = 0.012), ultimate pH (P = 0.042), and sex (P = 0.013) as shown in Table 3 with a pseudo R2 of 15.9%. The sensory tenderness increased with an increase in IMF and ultimate pH, but was lower for males. Sire (P b 0.002) and panellist (P = 0.013) impacted on the variance in sensory tenderness. For the sensory tenderness of the semimembranosus, the final model only contained log desmin (P b 0.001) such that as the degradation of desmin increased, sensory tenderness increased with a pseudo R2 of 15.5%. There was a significant difference (P b 0.006) between panellists.

M. Longissimus Intercept IMF Ultimate pH Sex (Male)

Table 4 Muscle traits that significantly explain the variation in shear force across all muscles (regression coefficients, standard errors and probability level).

Coefficient

std error

P-value

−25.79 4.09 15.23 −5.64

31.53 1.10 5.29 2.08

0.416 0.012 0.042 0.013

36.8 10.79

2.62 2.53

0.001 0.001

Indicates an interaction term.

was not the case for the biceps femoris with a low regression coefficient (P = 0.43), as shown in Table 4. Of all the muscles, the semimembranosus had a relatively strong correlation between shear force and log desmin (Fig. 1). Males on average produced higher shear force values than females (P = 0.018). For the variation in sensory tenderness, as assessed by the panellists (Model 4), it was identified that across both muscles, muscle type (P = 0.001) and log desmin (P = 0.001) were significant. The interactions between muscle type and log desmin (P = 0.002) were significant with a pseudo R2 of 63.3%. The sensory tenderness was lower in the semimembranosus (28.7 ± 1.42) than in the longissimus. As log desmin increased the sensory tenderness (2.85 ± 1.15) increased. The other factors that impacted on sensory tenderness were sire (P b 0.001) and panellist (P = 0.009). Sensory tenderness and shear force were strongly negatively correlated (Fig. 2). There was a strong muscle effect with the semimembranosus, having a lower mean sensory tenderness compared to the longissimus (−18.8 ± 1.67). Sex had minimal impact on the model because there was no difference between the estimates for males and females. Both laboratory (P b 0.001) and sire (P b 0.001) made a significant contribution to the variance of shear force. 4. Discussion 4.1. The relationship between shear force and meat quality traits within and across muscles For the longissimus, the significant factors that explained the variance in shear force were IMF, sarcomere length and desmin degradation. The fact that increasing levels of IMF led to a reduction in shear force, in the current study is consistent with the results from other studies (Mortimer et al., 2014; Warner et al., 2010). Our results show for every 1% increase in IMF, shear force decreased by 3.9 N (Table 2). It has been suggested that an increase in IMF leads to a decrease in shear force due potentially to a decrease in the density of muscle (Karlsson, Klont, & Fernandez, 1999). Based on this finding, it is interesting that in the current study only a relatively low amount of the variation in shear force could be explained by IMF, when sarcomere length and the degradation of desmin were also included in the model (Pseudo R2 of 14%). The impact of sarcomere length on shear force has been studied extensively (e.g. Hopkins et al., 2011). As sarcomere length increases the shear force decreases as shown by Rhee et al. (2004) for bovine longissimus aged for 14 days. In the current study, there was no attempt to increase the range in sarcomere length, with a natural variation in sarcomere length (1.3–2.1 μm) and therefore, the impact on shear force would be lessened as sarcomere length was only significant in the longissimus. The rate of pH and temperature decline for the longissimus suggests that some degree of shortening occurred, with relatively few carcasses passing through the optimal pH/temperature window (Thompson, 2002), but this was not reflected in the correlations between pH/temperature decline and sarcomere length (r b 0.1). However, the rate of pH and temperature decline had no effect on

C.P. Starkey et al. / Meat Science 113 (2016) 51–58

55

Fig. 1. Scatter plots of the relationship between shear force and log desmin within and across all muscles. Shear force (N) = 54.5–2.9 ± 1.13 (Log desmin), note the differences of the scattering and bunching of data points between the muscles.

shear force, contrary to other reports (Hopkins et al., 2011), but agreeing with the findings of Hopkins, van de Ven, and Holman (2015). Clearly, the current study has included a more extensive list of predictors and it could be that these over shadow rate of pH decline effects. For the longissimus, degradation of desmin was the third factor found to impact on shear force. This is consistent with previous reports (Rhee et al., 2004; Starkey et al., 2015). Like Starkey et al. (2015), a study conducted by Silva, Patarata, and Martins (1999) found that indices of proteolysis (myofibrillar fragmentation index and protein solubility) were related to variation in tenderness. Based on other studies, if the degradation of proteins such as titin and nebulin had also been measured then given their abundance, rate of degradation (Ho, Stromer, Rouse, & Robson, 1997) and role in skeletal muscle (Hopkins & Thompson, 2002), it is anticipated that further variation in shear force would have been explained (Ho, Stromer, & Robson, 1996). Unfortunately, unlike the study of Starkey et al. (2015) particle size was not measured in the current study, and if it had been this may have helped to explain more of the variation in shear force given it is likely to reflect the extent of the degradation of many different muscle proteins rather than just one targeted protein. It is of interest that the amount of degraded desmin was also negatively related to shear force in the semimembranosus, with the slope of the regression coefficient for log desmin of − 11.38 (P b 0.001, Fig. 1, Table 2). Rhee et al. (2004) also found a strong and significant (P b 0.001) negative correlation with shear force and the amount of degraded desmin in 14 day aged bovine m. semimembranosus. These results suggest that proteolysis is a more important determinant of tenderness for the semimembranosus than for the other muscles examined in this study.

In contrast (to the longissimus) for the biceps femoris, two different factors explained the variance in shear force; soluble collagen and animal age. Both of these factors were negatively correlated to shear force of the biceps femoris. There is a dearth of data on this muscle in lamb, however, based on the results of Rhee et al. (2004) in beef, the stronger influence of collagen on shear force in the biceps femoris compared to the longissimus was expected (although these authors measured total collagen content). It is of interest that animal age had a positive effect on shear force independent of collagen solubility given that it has been shown that as animals age the solubility of collagen is lowered (Young & Braggins, 1993). The “age” effect suggests that some other aspect of increasing maturity was impacting on muscle structure. One possibility is that desmin degradation did not identify the full extent of post-mortem proteolysis occurring within the muscle and this could have possibly been explained by measuring the calpastatin and calpain activity (Cruzen, Paulino, Lonergan, & Huff-Lonergan, 2014) or that the longer ageing of the biceps femoris reduced the significance of desmin degradation. Hopkins, Stanley, Toohey et al. (2007) found that as the animal's age increased from 8 to 22 months the ovine semimembranosus became tougher, so a similar effect in the biceps femoris was not unexpected, but again it is unclear why in the current study there was an age effect per se for the shear force of the biceps femoris, but not in the longissimus and semimembranosus. Modelling of data for the three muscles (Table 4) indicated that the same factors (covariates) were as important as when data were modelled within muscles (Table 2). As for individual muscles, an increase in sarcomere length overall lead to a reduction in shear force. For example, the analysis indicates that for every 1 μm increase in sarcomere length shear force would decrease by 15.2 Newtons, with 1 μm

56

C.P. Starkey et al. / Meat Science 113 (2016) 51–58

Fig. 2. Sensory tenderness vs shear force by muscle. Predicted line (solid) with approximate 95% CI (dashed) overlaid.

being the difference between extremely tough (1.2 μm) and extremely tender (2.2 μm) muscles, a scale effect similar to that reported by Hopkins, Stanley, Toohey et al. (2007). Proteolysis, as reflected by the degradation of desmin, had an overall effect on the variation in shear force, this driven by the significance of this trait for both the longissimus and semimembranosus. As previously reported, the lower the amount of desmin degradation, the tougher the muscles (e.g. Koohmaraie, Whipple, Kretchmar, Crouse, & Mersmann, 1991). Overall collagen content impacted on the variation in shear force, and the relationship between collagen and tenderness has previously been researched by authors such as: Seideman, Koohmaraie, and Crouse (1987); Rhee et al. (2004) and Starkey et al. (2015). Seideman et al. (1987) found a weak (r = − 0.01) non-significant (P N 0.05) relationship between soluble collagen and shear force values on bovine m. longissimus aged for 7 days. A study conducted by Starkey et al. (2015) found similar results as Seideman et al. (1987), where there was no relationship between shear force values and soluble collagen over different ageing periods in ovine m. longissimus. The biceps femoris is a locomotive muscle which has a higher content of collagen (Lepetit, 2007) so the fact collagen was found to have a significant effect on shear force was unsurprising. The sex of the animals had a significant effect on shear force, with castrated males being tougher by 3.3 N than females. As discussed by Hopkins and Mortimer (2014), the effect of sex on tenderness has not always been clear. For example, early work conducted by Corbett, Furnival, Southcott, Park, and Shorthose (1973) on young lambs (approximately 5 months of age) involving the longissimus and semimembranosus muscles found no gender effects on shear force. In contrast, Hopkins, Stanley, Martin, Toohey, and Gilmour (2007) found that the ovine longissimus of male lambs were significantly (P b 0.05) tougher compared to female lambs when measured at 8 months of age. This trend was also found by Warner et al. (2010), who found that females produced slightly lower shear force values than wethers (castrated males). A possible cause for the sexual differences observed

within these aforementioned experiments could be associated with the sexual maturity of the lambs. In a paper by Foster (1981) it was suggested that there are seasonal differences between sexual maturity of males and females. These differences in sexual maturity can be up to 10 weeks apart with males maturing at around 20 weeks of age and female lambs maturing around 30 weeks of age. 4.2. The relationship between sensory tenderness and meat quality measures within and across muscles The main factors that explained sensory tenderness in the longissimus (Model 2) were IMF, ultimate pH and the sex of the animals. IMF and ultimate pH were both positively correlated with sensory tenderness as found in many other studies (e.g. Hopkins, Hegarty, Walker, & Pethick, 2006; Pannier, Gardner, et al., 2014; Pannier, Pethick, et al., 2014) and the IMF effect was consistent with the effect on shear force. Consistent with the shear force results, the males produced tougher longissimus when tested for sensory tenderness. When sex was removed from the model, ultimate pH and IMF were still significant as explanatory predictors of sensory tenderness of the longissimus, with no new predictors becoming significant. This result suggests that the “sex” effect is independent of other measurable traits (desmin, collagen and sarcomere length) and IMF (Pannier, Pethick, et al., 2014) and this could warrant further investigation. None of the suspected explanatory factors (desmin, collagen and sarcomere length) had any significance in explaining sensory tenderness in the longissimus. This was an unexpected result with other authors Wheeler et al. (2000), and Rhee et al. (2004) finding at least one of these explanatory factors to be significant. There was only one factor that explained a significant part of the variation in semimembranosus sensory tenderness and that was the degradation of desmin. Thus, as the extent of desmin degradation increased, sensory tenderness of the semimembranosus increased and, this effect was consistent with the decrease in shear force. Desmin degradation

C.P. Starkey et al. / Meat Science 113 (2016) 51–58

was positively correlated and significantly related with sensory tenderness in results reported by Wheeler et al. (2000) in porcine m. semimembranosus 1 day aged samples, as observed within this experiment. The model for sensory tenderness (Model 4) which was constructed across muscles contained two significant factors, muscle type and log desmin (P b 0.001). As such, the semimembranosus was tougher than the longissimus as found previously (e.g. Hopkins, Stanley, Toohey et al., 2007). Desmin degradation was positively correlated to sensory tenderness, such that sensory tenderness increased as the extent of desmin degradation increased. Previous research from others such as: Rhee et al. (2004) and Wheeler et al. (2000) found mixed results. While Rhee et al. (2004) found a significant relationship between desmin degradation and sensory tenderness (P b 0.05) for 14 day aged bovine samples, Wheeler et al. (2000) observed a negative correlation, that was not significant in 1 day aged porcine muscle samples. 4.3. Relationship between shear force and sensory tenderness The relationship between shear force and sensory tenderness (Model 5) for both the semimembranosus and the longissimus were both strong and negatively correlated (r = − 0.45 ± 0.05). Shear force and sensory tenderness were significantly different (P b 0.05) for both muscles when modelled individually. Hopkins et al. (2006) found the relationship between shear force and sensory tenderness to have a regression coefficient (−0.76 ± 0.08) that was higher than what was observed within the current study in ovine m. longissimus aged for 5 days. The differences observed between the current study and Hopkins et al. (2006) would suggest that predictions of sensory tenderness from shear force would differ between studies. When comparing the data to the findings of Hopkins et al. (2006) it can be observed that the data within the current study had a higher and larger range of shear force values for the longissimus and this would impact on the final model. Similarly, Pannier, Gardner et al. (2014) had a wider range in the dataset they used than that of Hopkins et al. (2006) and the regressions were similar to the results presented in the current experiment. One possible reason for the similarity of the regression is that this experiment used a subset of data presented by the aforementioned author. In an experiment conducted by Hopkins, Lamb, Kerr, and van de Ven (2013) on the assessment of different tenderness measuring devices it was found that the shear force from any of the measuring devices could explain some of the variation in consumer sensory values of tenderness. It is of interest that the slope of the relationship between sensory tenderness and shear force in the current study is similar in the longissimus and the semimembranosus, but the intercept is different. This confirms that the latter muscle is intrinsically tougher than the former muscle (see e.g. Hopkins, Stanley, Martin, et al., 2007). Thus, a significant relationship between the two traits (sensory tenderness and shear force) will not always allow one to be used as a proxy for the other. 5. Conclusion This study shows that to predict the shear force of ovine meat, an overall model could be developed, but this would require time consuming measures like desmin degradation, sarcomere length and collagen solubility and it would have to include muscle type. Whilst measures of sarcomere length, collagen and desmin account for some of the explainable variation in shear force/tenderness, other aspects such as IMF, sex and animal age also contribute, but are muscle dependent. For sensory tenderness a relatively large amount of the variation could be explained by considering muscle type and desmin degradation, but again this provides limited scope for application in industry. The implications of this study suggest that simple models of factors such as sarcomere length, collagen content, desmin degradation and IMF are unlikely to be useful across a range of muscles.

57

Conflict of interest The authors declare no conflict of interest. Acknowledgements We would like to acknowledge the Sheep CRC for funding this project. We gratefully acknowledge the technical assistance of Ms. Xuemei Han and Mr. Mitchell Hope (UNE) and Mr. Matt Kerr and Ms. Tracy Lamb (NSW DPI). I also wish to acknowledge Dr. Liselotte Pannier of Murdoch University for granting me access to the sensory data set and providing feedback on this paper and Dr. Remy van de Ven (NSW DPI) for his calculation of pH decline parameters. References Anonymous (2005). Handbook of Australia meat (7th ed.). Sydney: AUS MEAT. AOAC (2000). OFFICIAL METHOD 990.26, 985.14, 992.23, 991.36 and 923.03, Official Methods of Analysis (17th ed.). Gaithersburg, MD: Association of Official Analytical Chemists (2000). Butler, D. (2009). Asreml: Asreml fits the linear mixed model. R package version, 3. Corbett, J., Furnival, E., Southcott, W., Park, R., & Shorthose, W. (1973). Induced cryptorchidism in lambs: Effects on growth rate, carcass and meat characteristics. Animal Production, 16, 157–163. Cross, H., West, R., & Dutson, T. (1981). Comparison of methods for measuring sarcomere length in beef semitendinosus muscle. Meat Science, 5, 261–266. Cruzen, S.M., Paulino, P.V., Lonergan, S.M., & Huff-Lonergan, E. (2014). Postmortem proteolysis in three muscles from growing and mature beef cattle. Meat Science, 96, 854–861. Foster, D.L. (1981). Mechanism for delay of first ovulation in lambs born in the wrong season (fall). Biology of Reproduction, 25, 85–92. Geesink, G., Kuchay, S., Chishti, A., & Koohmaraie, M. (2006). μ-Calpain is essential for postmortem proteolysis of muscle proteins. Journal of Animal Science, 84, 2834–2840. Geesink, G., Mareko, M., Morton, J., & Bickerstaffe, R. (2001). Effects of stress and high voltage electrical stimulation on tenderness of lamb m. longissimus. Meat Science, 57, 265–271. Ho, C. -Y., Stromer, M.H., Rouse, G., & Robson, R.M. (1997). Effects of electrical stimulation and postmortem storage on changes in titin, nebulin, desmin, troponin-T, and muscle ultrastructure in Bos indicus crossbred cattle. Journal of Animal Science, 75, 366–376. Ho, C. -Y., Stromer, M.H., & Robson, R.M. (1996). Effect of electrical stimulation on postmortem titin, nebulin, desmin, and troponin-T degradation and ultrastructural changes in bovine longissimus muscle. Journal of Animal Science, 74, 1563–1575. Hopkins, D., L., Stanley, D., Toohey, E., Gardner, G., Pethick, D., & van de Ven, R. (2007n). Sire and growth path effects on sheep meat production. 2. Meat and eating quality. Australian Journal of Experimental Agriculture, 47, 1219–1228. Hopkins, D.L., Toohey, E., Warner, R., Kerr, M., & van de Ven, R. (2010). Measuring the shear force of lamb meat cooked from frozen samples: Comparison of two laboratories. Animal Production Science, 50, 382–385. Hopkins, D.L., & Geesink, G. (2009). Protein degradation post mortem and tenderisation. Applied muscle biology and meat science (pp. 149–173). USA: CRC Press, Taylor & Francis Group. Hopkins, D.L., & Mortimer, S.I. (2014). Effect of genotype, gender and age on sheep meat quality and a case study illustrating integration of knowledge. Meat Science, 98, 544–555. Hopkins, D.L., & Thompson, J.M. (2001). The relationship between tenderness, proteolysis, muscle contraction and dissociation of actomyosin. Meat Science, 57, 1–12. Hopkins, D.L., & Thompson, J.M. (2002). The degradation of myofibrillar proteins in beef and lamb meat using denaturing electrophoresis — An overview. Journal of Muscle Foods, 13, 81–102. Hopkins, D.L., Hegarty, R., Walker, P., & Pethick, D. (2006). Relationship between animal age, intramuscular fat, cooking loss, pH, shear force and eating quality of aged meat from sheep. Animal Production Science, 46, 879–884. Hopkins, D.L., Lamb, T.A., Kerr, M.J., & van de Ven, R. (2013). The interrelationship between sensory tenderness and shear force measured by the G2 tenderometer and a Lloyd texture analyser fitted with a warner-bratzler head. Meat Science, 93, 838–842. Hopkins, D.L., Stanley, D.F., Martin, L.C., Toohey, E.S., & Gilmour, A.R. (2007). Genotype and age effects on sheep meat production. 3. Meat quality. Australian Journal of Experimental Agriculture, 47, 1155–1164. Hopkins, D.L., Toohey, E., Lamb, T., Kerr, M., de Ven, R., & Refshauge, G. (2011). Explaining the variation in the shear force of lamb meat using sarcomere length, the rate of rigor onset and pH. Meat Science, 88, 794–796. Hopkins, D.L., van de Ven, R.J., & Holman, B.W.B. (2015). Modelling lamb carcase pH and temperature decline parameters: Relationship to shear force and abattoir variation. Meat Science, 100, 85–90. Huff-Lonergan, E., & Lonergan, S. M. (2005). Mechanisms of water-holding capacity of meat: The role of postmortem biochemical and structural changes. Meat Science, 71, 194–204. Karlsson, A.H., Klont, R.E., & Fernandez, X. (1999). Skeletal muscle fibres as factors for pork quality. Livestock Production Science, 60, 255–269. Kenward, M.G., & Roger, J.H. (1997). Small sample inference for fixed effect estimators from restricted maximum likelihood. Biometrics, 53, 983–997.

58

C.P. Starkey et al. / Meat Science 113 (2016) 51–58

Koohmaraie, M., Whipple, G., Kretchmar, D., Crouse, J., & Mersmann, H. (1991). Postmortem proteolysis in longissimus muscle from beef, lamb and pork carcasses. Journal of Animal Science, 69, 617–624. Lepetit, J. (2007). A theoretical approach of the relationships between collagen content, collagen cross-links and meat tenderness. Meat Science, 76, 147–159. Mortimer, S.I., van der Werf, J.H.J., Jacob, R.H., Hopkins, D.L., Pannier, L., Pearce, K.L., ... Pethick, D.W. (2014). Genetic parameters for meat quality traits of Australian lamb meat. Meat Science, 96, 1016–1024. Pannier, L., Gardner, G., Pearce, K., McDonagh, M., Ball, A., Jacob, R., & Pethick, D. (2014). Associations of sire estimated breeding values and objective meat quality measurements with sensory scores in Australian lamb. Meat Science, 96, 1076–1087. Pannier, L., Pethick, D., Geesink, G., Ball, A., Jacob, R., & Gardner, G. (2014). Intramuscular fat in the longissimus muscle is reduced in lambs from sires selected for leanness. Meat Science, 96, 1068–1075. Perry, D., Shorthose, W., Ferguson, D., & Thompson, J. (2001). Methods used in the CRC program for the determination of carcass yield and beef quality. Australian Journal of Experimental Agriculture, 41, 953–1040. R Development Core Team (2014). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing (from URL http://www.Rproject.org/). Rhee, M., Wheeler, T., Shackelford, S., & Koohmaraie, M. (2004). Variation in palatability and biochemical traits within and among eleven beef muscles. Journal of Animal Science, 82, 534. Safari, E., Channon, H.A., Hopkins, D.L., Hall, D.G., & van de Ven, R. (2002). A national audit of retail lamb loin quality in Australia. Meat Science, 61, 267–273. Seideman, S., Koohmaraie, M., & Crouse, J. (1987). Factors associated with tenderness in young beef. Meat Science, 20, 281–291. Silva, J.A., Patarata, L., & Martins, C. (1999). The influence of ultimate pH on bovine meat tenderness during ageing. Meat Science, 52, 453–459. Smulders, F., Marsh, B., Swartz, D., Russell, R., & Hoenecke, M. (1990). Beef tenderness and sarcomere length. Meat Science, 28, 349–363.

Starkey, C., Geesink, G., Oddy, H., & Hopkins, D.L. (2015). Explaining the variation in lamb longissimus shear force (tenderness) across and within ageing periods using protein degradation, sarcomere length and collagen characteristics. Meat Science, 105, 32–37. Stram, D.O., & Lee, J.W. (1994). Variance components testing in the longitudinal mixed effects model. Biometrics, 50, 1171–1177. Thompson, J. (2002). Managing meat tenderness. Meat Science, 62, 295–308. Thompson, J., Gee, A., Hopkins, D., L., Pethick, D., Baud, S., & O'Halloran, W. (2005). Development of a sensory protocol for testing palatability of sheep meats. Animal Production Science, 45, 469–476. Veiseth, E., Shackelford, S., Wheeler, T., & Koohmaraie, M. (2004). Factors regulating lamb longissimus tenderness are affected by age at slaughter. Meat Science, 68, 635–640. van de Ven, R.J., Pearce, K.L., & Hopkins, D.L. (2014). Post-mortem modelling of pH and temperature in related lamb carcases. Meat Science, 96, 1034–1039. Warner, R., Jacob, R., Hocking Edwards, J., McDonagh, M., Pearce, K., Geesink, G., ... Pethick, D. (2010,). Quality of lamb meat from the information nucleus flock. Animal Production Science, 50, 1123–1134. van der Werf, J., Kinghorn, B., & Banks, R. (2010). Design and role of an information nucleus in sheep breeding programs. Animal Production Science, 50, 998–1003. Wheeler, T., & Koohmaraie, M. (1994). Prerigor and postrigor changes in tenderness of ovine longissimus muscle. Journal of Animal Science, 72, 1232. Wheeler, T., Shackelford, S., & Koohmaraie, M. (2000). Variation in proteolysis, sarcomere length, collagen content, and tenderness among major pork muscles. Journal of Animal Science, 78, 958–965. Young, O.A., & Braggins, T.J. (1993). Tenderness of ovine semimembranosus: Is collagen concentration or solubility the critical factor? Meat Science, 35, 213–222. Young, O.A., Hopkins, D.L., & Pethick, D.W. (2005). Critical control points for meat quality in the Australian sheep meat supply chain. Australian Journal of Experimental Agriculture, 45, 593–601.