Published December 5, 2014
Meta-analysis of the effect of animal maturity on muscle characteristics
in different muscles, breeds, and sexes of cattle
N. M. Schreurs,*1 F. Garcia,* C. Jurie,† J. Agabriel,* D. Micol,* D. Bauchart,‡
A. Listrat,† and B. Picard†
*Equipe Systèmes de Production; †Croissance et Métabolisme du Muscle; and ‡Nutriments et Métabolismes,
INRA UR1213 Herbivores, Site de Theix, F-63122 Saint-Genès-Champanelle, France
ABSTRACT: The effect of animal maturity on fiber
cross-sectional area, percentage of fiber types, activities of isocitrate dehydrogenase (ICDH) and lactate
dehydrogenase (LDH), total and insoluble collagen
and lipid concentration was investigated in the longissimus thoracis (LT), semitendinosus (ST), and triceps
brachii (TB) muscles. The analysis considered 2,642
muscle samples from bulls, steers, and cows of Aubrac,
Charolais, Limousin, Montbéliard, and Salers breeds.
For the bulls, the fiber cross-sectional area, percentage of slow oxidative fibers, and ICDH activity showed
a quadratic relationship (P < 0.05), and the percentage of fast oxidative-glycolytic and fast glycolytic fibers and LDH activity showed a cubic relationship (P
< 0.05) with increased maturity. A linear relationship
was observed for the collagen and lipid muscle characteristics. The response equation coefficients for different muscles indicate that development of muscle
characteristics is different for each muscle. Compared
with the other muscles, ST muscle had a greater fiber
cross-sectional area, proportion of fast glycolytic fibers,
LDH activity, and collagen content. The LT muscle had
a greater proportion of slow-oxidative fibers and lipid
(P < 0.05). Within the ST muscle, all characteristics except lipid concentration showed different development
between the breeds. Steers showed greater changes in
muscle fiber cross-sectional area, percentage of fast
oxidative-glycolytic and fast glycolytic fibers, and total
lipid in the muscle with increasing maturity compared
with bulls. The mean fiber cross-sectional area and
percentage of fast glycolytic fibers was greater and the
mean lipid concentration was less in bulls compared
with steers (P < 0.05). Data for cows were from more
mature animals. Muscle characteristics in cows did not
show large changes with increasing degree of maturity.
Muscle type accounts for a greater proportion of the
variation in the muscle characteristics than breed and
sex of the animal.
Key words: beef, breed, cattle, maturity, meta-analysis, muscle, sex
©2008 American Society of Animal Science. All rights reserved.
INTRODUCTION
Optimizing beef quality is important for consumer
satisfaction and economic sustainability in the beef industry. Measures of beef quality have been correlated
to differences in muscle characteristics (Renand et al.,
2001). Variation in beef quality between and within animals is partly attributed to genetic and environmental
production factors that affect the muscle characteristics. To manipulate beef quality for economic advantage it is necessary to understand how factors such as
muscle type, breed, and sex influence the muscle characteristics in the growing animal.
Experiments have identified that muscle characteristics such as contractile fiber cross-sectional area,
1
Corresponding author: [email protected]
Received January 18, 2008.
Accepted July 1, 2008.
J. Anim. Sci. 2008. 86:2872–2887
doi:10.2527/jas.2008-0882
metabolic enzyme activity, collagen content and solubility, and lipid content change as cattle mature (Jurie et al., 1995a; Wegner et al., 2000). Differences in
muscle characteristics also occur between muscle types
(Jurie et al., 1995b; Von Seggern et al., 2005) and sexes
(Picard et al., 1995). Few differences in muscle characteristics are believed to exist between cattle breeds
raised under similar production circumstances (Jurie
et al., 2005).
Experiments to investigate muscle characteristics in
growing animals have considered only a limited number of factors and normally use small numbers of animals. A meta-analysis summarizes many experiments
and considers many factor levels, increasing the likelihood that inferences reflect the population. McPhee et
al. (2006) used a meta-analytic approach to assess factors affecting carcass characteristics in steers. A similar approach would be valuable for studying muscle
characteristics.
2872
Meta-analysis of muscle characteristics
The objective of this study was to perform a metaanalysis to 1) establish the effect of degree of maturity with different muscle types, breeds, or sexes on 11
muscle characteristics in French cattle and 2) rank the
factors for the extent they influence changes in muscle
characteristics with maturity. This will aid the construction of a dynamic model of muscle development.
MATERIALS AND METHODS
Animal Care and Use Committee approval was not
necessary for this study as the data used in this metaanalysis were obtained from previously performed experiments.
Study Sample
During a period from 1986 to 2006, several experiments were conducted by personnel of the Herbivore
Research Unit of the Institut National de la Recherche
Agronomique (INRA) in France to investigate muscle
characteristics in French cattle. A database was created to compile the data from these experiments and to
allow the relationship between degree of maturity and
muscle characteristics to be explored in a meta-analysis.
Access to the original results from the experiments was
available so the database contained measurements of
muscle characteristics from individual muscle samples
of individual animals. As such, this meta-analysis used
the measurements from individual animal muscles of
each experiment rather than treatment means, which
are commonly used in meta-analytic studies when only
published results are available.
Database Collection and Coding
for Meta-Analysis
For the meta-analysis, the quantitative and qualitative details extracted from the experimental results included the muscle from which the sample was obtained;
the sex, breed, and identity number of the animal from
which the muscle sample was obtained; the BW of the
animal when muscle samples were obtained; name of
the experiment in which the sampled animal was included; and measurements for the muscle characteristics. Codes were used to distinguish the data according
the animal’s breed, sex, and muscle type as needed.
Unique codes were given to each experiment.
Muscle Characteristics and Factors
Considered in Meta-Analysis
Eleven muscle characteristics that were considered
the most likely to explain differences in beef quality
were evaluated in the meta-analysis (Table 1). These
characteristics, which were the dependent variables
in the meta-analyses, were the mean muscle fiber
cross-sectional area (CSA; µm2), percentage of slow
oxidative fibers (%SO), fast oxidative-glycolytic fibers
2873
(%FOG), and fast glycolytic fibers (%FG), activity of
isocitrate dehydrogenase (ICDH; µmol⋅min−1⋅g−1), and
lactate dehydrogenase (LDH; µmol⋅min−1⋅g−1) in fresh
muscle, concentration of insoluble (CINS; µg of OHproline⋅mg−1) and total (TCOL; µg of OH-proline⋅mg−1)
collagen in dry muscle, and concentration of phospholipid (PL; mg⋅g−1), triglyceride (TG; mg⋅g−1), and total
lipid (TLIP; mg⋅g−1) in fresh muscle.
Degree of Maturity. The degree of maturity
(DoM) of an animal when muscle samples were obtained was included as the independent variable for
examining the effects of muscle type, breed, and sex
on the muscle characteristics. Degree of maturity was
defined as the proportion of mature BW attained when
the muscle sample was taken (i.e., BW at sampling
divided by mature BW; Fitzhugh and Taylor, 1971).
The mature BW used to calculate the DoM for different breeds and sexes are given in Table 2. A DoM >1
is possible when a sampled animal has a BW greater
than the standard mature BW for its breed and sex (as
defined in Table 2). Degree of maturity was treated as
a continuous, quantitative variable to allow analysis of
the range of DoM values in the database.
Muscle Type. The analysis focused on the characteristics of 3 muscles: the longissimus thoracis (LT),
semitendinosus (ST), and triceps brachii (TB). These
muscles were the predominant muscles of interest in
the experiments in the database and represent 3 distinct regions of the carcass; namely, the loin, round,
and chuck. Meat cuts from these 3 muscles differ in
quality (in particular, tenderness) allowing differences
in muscle characteristics between muscles to be fully
investigated.
Sexes and Breeds. The database (Table 1) and
the meta-analysis considered 3 sexes and 5 breeds. The
sexes were categorized as bulls, cows, and steers. The
breeds included in the database were Aubrac, Charolais, Limousin, Montbéliard, and Salers. Two experiments used steers from crossbreeding of Charolais and
Salers cattle.
Data Investigation and Statistical Analyses
Preliminary Data Investigation. A test for
homogeneity of experiment means was performed. For
all muscle characteristics, there was a difference (P <
0.001) between the means in each experiment. This
identified the need for a random-effects model with
the experiment as a random variable to account for
between-experiment variation. The Levene’s test was
used to assess the variance between fixed-effect groups.
When the test indicated unequal variance, it was accommodated for by incorporating the Satterthwaite’s
approximation into the statistical analyses (Spilke et
al., 2005). Outlying values were identified from a PROC
UNIVARIATE analysis of the data. Graphical investigation indicated that these outlying values were not
measurement errors and it was decided to leave them
in the analyses, as they were possibly representative
LI
MO
CH
MO
CH
MO
MO
LI, SA
LI
SA
CH
CH
CH
CH
CH
CH
SA
LI
SA
CH × SA
CH × SA
CH
CH
CH
CH
AU, CH, LI, SA
AU, CH, LI, SA
AU, CH, LI, SA
AU, CH, LI, SA
AU, CH, LI, SA
AU, CH, LI, SA
CH
CH
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Total
F
C, M
C
C
M
M
M
M
M
F
M
M
M
M
C
C
M
M
M
C
C
C
C
C
C
M
M
M
F
F
F
C, F
C, F
Sex2
LT, ST, TB
LT, ST, TB
LT, ST, TB
LT, ST, TB
LT, ST, TB
LT, ST
ST
LT, ST
LT, ST
TB
LT
LT
LT
LT, ST
LT, ST
LT
LT, ST
LT, ST
LT, ST
LT
LT
ST
ST
ST
LT, ST
LT, ST, TB
LT, ST, TB
LT, ST, TB
LT, ST, TB
LT, ST, TB
LT, ST, TB
TB
LT, TB
Muscle3
0.84–1.13
0.16–0.65
0.40–0.73
0.37–0.68
0.41–0.66
0.09–0.25
0.06–0.62
0.47–0.65
0.04–0.57
0.50–1.19
0.45–0.77
0.42–0.74
0.47–0.77
0.48–0.73
0.56–0.76
0.57–0.77
0.48–0.61
0.43–0.71
0.17–0.42
0.60–0.67
0.52–0.67
0.57–0.74
0.62–0.79
0.65–0.77
0.45–0.59
0.55–0.95
0.55–1.07
0.57–0.86
0.88–1.30
0.86–1.29
0.86–1.15
0.57–0.94
0.50–0.86
Degree of
maturity
38
96
91
59
87
95
69
26
1,635
1,951
90
184
108
160
21
173
24
14
59
219
22
%SO
38
96
90
59
87
95
69
26
24
28
90
184
143
160
104
173
24
14
59
219
22
65
34
19
29
CSA
1,634
38
96
90
59
87
95
69
26
90
184
108
160
21
173
24
14
59
219
22
%FOG
1,635
38
96
91
59
87
95
69
26
90
184
108
160
21
173
24
14
59
219
22
%FG
48
162
158
126
68
219
23
65
34
20
58
46
29
90
295
24
12
18
24
24
30
38
96
91
59
87
95
69
26
54
2,188
68
219
23
65
34
20
58
46
29
90
295
24
12
18
24
24
30
38
96
91
59
87
95
69
26
54
2,309
LDH
167
162
160
126
ICDH
1,073
38
95
86
50
85
91
59
25
1,092
38
96
90
56
85
93
65
25
24
23
65
38
20
58
23
29
90
23
65
38
20
58
23
29
90
24
34
69
71
TCOL
34
69
71
CINS
985
68
96
91
59
87
95
69
12
18
24
65
38
20
42
46
29
126
PL
1,034
68
96
91
59
87
95
69
7
24
12
18
24
18
65
38
20
42
46
29
126
TG
1,053
68
96
91
59
87
95
69
26
24
12
18
24
18
65
38
20
42
46
29
126
TLIP
2
AU = Aubrac, CH = Charolais, LI = Limousin, MO = Montbéliard, and SA = Salers.
C = castrated (steer), F = cow, and M = intact male.
3
LT = longissimus thoracis, ST = semitendinosus, and TB = triceps brachii.
4
The number of observations for the muscle characteristic of interest is indicated. Designation for dependent variables: CSA = average muscle fiber cross-sectional area (µm2); %SO = percentage of slow oxidative fibers; %FOG = percentage of fast oxidative-glycolytic fibers; %FG = percentage of fast glycolytic fibers; ICDH = activity of isocitrate dehydrogenase (µmol⋅min−1⋅g−1 of fresh
muscle); LDH = activity of lactate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); CINS = insoluble collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); TCOL = total collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); PL = phospholipid concentration (mg⋅g−1 of fresh muscle); TG = triglyceride concentration (mg⋅g−1 of fresh muscle); TLIP = total lipid concentration
(mg⋅g−1 of fresh muscle).
1
Breed1
Experiment
Number of observations4
Table 1. Summary of the database, which comprised 33 experiments that investigated the muscle characteristics from French cattle breeds
2874
Schreurs et al.
2875
Meta-analysis of muscle characteristics
of the population. Data investigation also identified a
difference in the DoM between bulls and cows. Most
bulls were at a DoM <1, whereas most cows were at a
DoM >0.85. This is a characteristic of French beef production systems in which cows are slaughtered for beef
production at a much later age than bulls but usually
before 10 yr of age to retain carcass value. Consequently, when analyzing the effect of muscle type and breed,
the analyses were carried out on the data of bulls and
cows separately and the effect of sex considered bulls
versus steers (effect of castration).
Statistical Analysis of Muscle Characteristics with Degree of Maturity in Different
Muscles, Breeds, and Sexes. Three sets of analy-
ses were carried out. The first set compared the effect
of each muscle (LT, ST, and TB) on the muscle characteristics. The analysis was carried out on data from
bulls and then repeated with data from cows. A further
2 sets of analyses used data from the ST muscle only to
assess the effect of breed (in bulls and cows separately)
and castration (steers vs. bulls).
Muscle characteristics were analyzed using mixed
models (PROC MIXED, SAS Inst. Inc., Cary, NC) using
REML for estimating the variance components. In this
manner, the fixed-effect terms of the statistical model
were muscle type (set 1), breed (set 2), or castration
(set 3) with DoM as the covariate. The initial model
tested for cubic, quadratic, and linear effects with DoM
and their possible interaction with muscle type (set 1),
breed (set 2), or castration (set 3). The higher order
terms and interactions that were not significant were
sequentially removed from the model and analysis repeated. Analyses incorporated the experiment from
which the data originated as the random effect. The
statistical model initially included the random intercept, linear, quadratic, and cubic effects as appropriate for the model utilized and a possible covariance
between these (option UN). The covariance parameter
was regarded as significantly different from zero at
a probability of P < 0.07, which is more lenient than
the usual 0.05 level because accurate estimations of
variance and covariance require a substantial number of observations (St-Pierre, 2001). If the covariance
parameter was not correlated, the UN option was removed and analyses repeated. Nonsignificant random
effects were removed sequentially from the statistical
model and analyses repeated as required. To account
for the multidimensional space that is created when
results are from a mixed model regression, the observations in the figures have been adjusted (Y values on
the regression line plus residuals) as recommended by
St-Pierre (2001).
RESULTS
Database Composition
The final database contained measurements of
muscle characteristics from 2,642 individual muscle
samples of 33 experiments. More measurements were
available for fiber and enzyme characteristics than
for collagen and lipid characteristics (Table 1). The
ST muscle accounted for 1,134 samples, LT for 1,026
samples, and TB for 482 samples. The number of samples from bulls, steers, and cows were 1,437, 726, and
479, respectively. The Charolais breed contributed 987
samples; Limousin, 850; Montbéliard, 326; Salers, 323;
Aubrac, 126; and Charolais × Salers, 30.
Effect of Degree of Maturity on Muscle
Characteristics
Cross-Sectional Fiber Area. The CSA increased
quadratically with DoM in bulls (P < 0.001; Figure 1)
and linear changes were observed in cows (Figure 1).
Mean CSA was greater in ST muscle than in LT and
TB muscles in bulls (P < 0.05; Table 3) and cows (P <
0.05; Table 4). The muscles of the bulls had different
linear and quadratic coefficients when regressing CSA
against DoM (P < 0.001; Table 3). The greatest increase
in CSA was seen in the ST muscle of bulls (Figure 1).
The linear coefficients indicate that the CSA continued to increase in the LT and TB muscle but decreased
slightly in the ST muscle of cows (Table 4; Figure 1).
Mean CSA was greater in Charolais compared with
Aubrac and Montbéliard bulls (P < 0.05; Table 5) and
greater in Limousin cows compared with other cow
breeds (P < 0.05; Table 6). The bull breeds differed in
their linear coefficients (P < 0.01) but not quadratic coefficients (Table 5). The linear coefficient was greatest
with Aubrac and least in Montbéliard bulls. Compared
with steers, the bulls had a greater linear coefficient (P
< 0.01) and a greater mean CSA (P < 0.05; Table 7).
Percentage of Slow Oxidative, Fast Oxidative-Glycolytic, and Fast Glycolytic Muscle
Fibers. The %SO showed a quadratic and %FOG and
%FG a cubic relationship with DoM in bulls but fiber
percentages did not change in cows (Figure 2).
In both bulls and cows the mean %SO was greatest
in the LT muscle and least in the ST with TB intermediate (P < 0.05; Tables 3 and 4). For the relationship of
%SO with DoM, the linear and quadratic coefficients
were different between the muscle types of bulls (P <
0.05; Table 3), and the linear coefficient was different
between muscle types of cows (P < 0.05; Table 4). Mean
Table 2. Mature BW (kg) used to calculate degree of
maturity for different breeds and sexes of cattle in the
meta-analysis
Breed
Cow
Aubrac
Charolais
Limousin
Salers
Montbéliard
Charolais × Salers
650
750
700
650
Steer
1,150
900
1,000
Bull
950
1,150
1,050
1,000
950
2876
Schreurs et al.
Figure 1. Quadratic relationship for bulls (A) and linear relationship for cows (B) when mean muscle fiber
cross-sectional area (CSA) of the longissimus thoracis (LT, □), semitendinosus (ST, ∆), or triceps brachii (TB, ♦)
muscle is regressed with the degree of maturity of the animal.
%SO was greater in Montbéliard bulls and Aubrac and
Salers cows compared with the other breeds (P < 0.05;
Tables 5 and 6). There was a linear relationship of
%SO with DoM when considering just ST muscle, and
linear coefficients were different between bull breeds
(P < 0.05; Table 5). The %SO did not differ between
steers and bulls (Table 7).
The ST and TB muscles had a greater mean %FOG
compared with the LT muscle in both bulls and cows (P
< 0.05; Tables 3 and 4). For the relationship of %FOG
with DoM in bulls, there was a difference between the
muscles for the linear (P < 0.001), quadratic (P < 0.05),
and cubic (P < 0.05) coefficients (Table 3). Mean %FOG
was greatest in Aubrac and Salers bulls and least in
Montbéliard bulls (P < 0.05; Table 5). In cows, the Limousin had a decreased mean %FOG compared with the
other cow breeds (P < 0.05; Table 6). When comparing
breeds using just the ST muscle, the %FOG gave a quadratic relationship with DoM. The linear coefficient for
this relationship was negative and different between
the bull breeds (P < 0.01; Table 6). Mean %FOG was
not different between bulls and steers however, regressing %FOG against DoM gave different linear (P
< 0.001) and quadratic (P < 0.01) coefficients for bulls
and steers (Table 7).
The ST muscle had a greater mean %FG compared
with the LT and TB in bulls and cows (P < 0.05; Tables
3 and 4). When regressing %FG with DoM, the linear coefficient differed between muscle types of bulls
(P < 0.001; Table 3). There was no difference in the
mean %FG between bull breeds but Limousin cows
had a greater mean %FG compared with the other cow
breeds (P < 0.05; Table 6). When comparing bull breeds
using just the ST muscle, the %FG gave a quadratic
relationship with DoM. The linear coefficient for this
relationship was different between the bull breeds (P
< 0.001; Table 5). The mean %FG was greater in bulls
than steers. The quadratic relationship of FOG% in the
ST muscle with DoM gave different linear (P < 0.001)
and quadratic (P < 0.001) coefficients for bulls and
steers (Table 7).
Activities of Isocitrate Dehydrogenase and
Lactate Dehydrogenase. In bulls, ICDH activ-
ity showed a quadratic (Figure 3) and LDH activity a
cubic (Figure 4) relationship with DoM but there was
no change for these enzyme activities with increasing
DoM in cows.
The TB muscle had the greatest and the ST muscle
the least mean ICDH activity in both bulls and cows (P
< 0.05; Tables 3 and 4). For the relationship of ICDH
activity with DoM in bulls, there was a difference between the muscles for the linear and quadratic coefficients (P < 0.001; Table 3). Mean ICDH activity was
greater in Montbéliard bulls compared with Limousin
bulls (P < 0.05; Table 5) with other bull breeds intermediate. Charolais cows had less mean ICDH activity
compared with the other cow breeds (P < 0.05; Table 6).
The linear and quadratic coefficients for the relationship of ICDH activity with DoM were different between
the bull breeds (P < 0.001; Table 5) but the same between bulls and steers. Mean ICDH activity was greater in bulls compared with steers (P < 0.05; Table 7).
The ST muscle had a greater mean LDH activity compared with the LT and TB muscles in both bulls and
cows (P < 0.05; Tables 3 and 4). There was a difference
between the muscle types in bulls for the linear coefficient of the relationship of LDH activity with DoM (P
< 0.05; Table 3). Mean LDH activity was not different
between the bull breeds. In the cows, mean LDH activity was greatest in the muscle of Limousin and least in
the muscle of Charolais and Salers (P < 0.05; Table 6).
When regressing LDH activity with DoM, the linear
277–519
187–571
72–157
n1
%SO
***
***
***
*
*
NS
NS
4.2
10.7
29.2
67.6
8.2
−18.5
−32.3
−6.8
19.4
***
NS
***
*
***
NS
NS
881.1
56.9
1,771.8
5,679.9
8,497.8
−1,746.2
−2,998.0
−1,949.4
5,778.0
2,904 ± 124b 29.1 ± 1.1a
3,894 ± 119a 12.8 ± 1.1c
2,701 ± 146b 25.9 ± 1.9b
CSA
−92.5
−92.5
−92.5
44.7
44.7
44.7
54.8
−66.0
−161.6
48.4
65.7
52.8
45.9
48.0
42.9
***
***
***
***
NS
*
NS
51.1 ± 2.4b
62.2 ± 2.4a
50.4 ± 2.5b
%FG
−52.3
131.2
327.3
−21.3
−86.7
−212.9
35.0
42.4
68.1
***
***
***
***
*
*
*
19.0 ± 2.0b
25.1 ± 1.9a
27.4 ± 4.2a
%FOG
2.0
7.8
6.9
−1.1
−10.2
−9.9
1.8
4.5
5.6
*
***
***
***
***
NS
NS
1.8 ± 0.1b
1.6 ± 0.1c
2.5 ± 0.1a
ICDH
881.8
881.8
881.8
−2,399.6
−2,399.6
−2,399.6
1,666.3
2,162.6
1,924.7
749.7
507.0
573.2
***
***
***
***
NS
*
NS
1,050 ± 54b
1,078 ± 54a
1,014 ± 58b
LDH
Enzyme activity2
1.31
−0.93
−0.05
1.6
4.9
3.6
NS
***
***
NS
NS
NS
NS
2.5 ± 0.2c
4.3 ± 0.2a
3.8 ± 0.2b
CINS
−1.63
−6.34
−4.23
4.1
9.5
7.7
*
***
***
NS
NS
NS
NS
3.1 ± 0.1c
5.3 ± 0.1a
4.9 ± 0.2b
TCOL
Collagen2
−1.5
−1.5
−1.5
7.5
7.9
9.2
***
***
NS
NS
NS
NS
NS
6.5 ± 0.1c
6.9 ± 0.1b
8.2 ± 0.1a
PL
12.8
−2.5
3.9
2.2
4.9
4.7
*
NS
**
NS
NS
NS
NS
10.9 ± 1.1a
3.2 ± 1.2c
7.3 ± 1.3b
TG
Lipids2
14.1
0.9
7.1
10.2
11.6
13.7
**
NS
*
NS
NS
NS
NS
19.7 ± 0.9a
12.2 ± 1.1b
18.5 ± 1.2a
TLIP
1
Within a column, means without a common superscript letter differ (P < 0.05).
The number of samples used for analysis is given as a range as some muscle characteristics have more observations than others.
2
CSA = average muscle fiber cross-sectional area (µm2); %SO = percentage of slow oxidative fibers; %FOG = percentage of fast oxidative-glycolytic fibers; %FG = percentage of fast glycolytic fibers; ICDH = activity of isocitrate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); LDH = activity of lactate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); CINS = insoluble collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); TCOL = total collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); PL = phospholipid concentration (mg⋅g−1 of fresh muscle); TG = triglyceride
concentration (mg⋅g−1 of fresh muscle); and TLIP = total lipid concentration (mg⋅g−1 of fresh muscle).
3
Parameter estimates are for the mixed model equation: Yij = α1LTij + α2STij + α3TBij + γ1i + (β1LTij + β2STij + β3TBij + γ2i)DoMij + (δ1LTij + δ2STij + δ3TBij + γ3i)DoMij2 + (θ1LTij + θ2STij + θ3TBij
+ γ4i)DoMij3 + εij, where i = 1, ……, 33 experiments; j = 1, ……, ni values of the continuous variable, DoM; Yij = the expected outcome of the muscle characteristic Y, observed at DoM j in the experiment i; LTij, STij, TBij = dummy variables that take on the value of 1 or 0 to indicate the muscle being considered for an observation at DoM j and in experiment i; DoMij = the value j of the
continuous variable, DoM, in experiment i; α1–3, β1–3, δ1–3, θ1–3 = coefficients for the intercept, linear, quadratic, and cubic fixed effects, respectively, when regressing Y on DoM for the muscles
LT, ST, and TB; γ1i, γ2i, γ3i, γ4i = the random intercept, linear, quadratic and cubic effect of experiment; and εij = the unexplained residual error.
NS = not significant; *P < 0.05; **P < 0.01; ***P < 0.001.
a–c
Least squares means
Longissimus thoracis
Semitendinosus
Triceps brachii
Significance of fixed effects
DoM
Muscle
DoM × muscle
DoM2
DoM2 × muscle
DoM3
DoM3 × muscle
Fixed effect estimates3
Intercept
Longissimus thoracis (α1)
Semitendinosus (α2)
Triceps brachii (α3)
Linear coefficient
Longissimus thoracis (β1)
Semitendinosus (β2)
Triceps brachii (β3)
Quadratic coefficient
Longissimus thoracis (δ1)
Semitendinosus (δ2)
Triceps brachii (δ3)
Cubic coefficient
Longissimus thoracis (θ1)
Semitendinosus (θ2)
Triceps brachii (θ3)
Item
Muscle fibers2
Table 3. Least squares means (±SEM) and parameter estimates for the relationship of muscle characteristics with degree of maturity (DoM) for the longissimus thoracis, semitendinosus, or triceps brachii muscles of bulls
Meta-analysis of muscle characteristics
2877
79–135
80–156
106–171
n1
30.7 ± 0.6a
9.7 ± 0.5c
28.5 ± 0.5b
NS
***
*
45.6
5.4
28.5
−15.0
4.4
−0.01
*
***
*
1,388.3
5,118.5
970.2
1,667.6
−627.3
2,441.2
%SO
3,039 ± 205c
4,498 ± 200a
3,386 ± 192b
CSA
−0.9
−0.9
−0.9
17.4
25.7
28.7
NS
***
NS
16.5 ± 3.4c
24.8 ± 3.4b
27.8 ± 3.4a
%FOG
3.5
3.5
3.5
49.8
62.9
40.5
NS
***
NS
53.2 ± 3.8b
66.4 ± 3.8a
44.0 ± 3.8c
%FG
0.15
0.15
0.15
1.7
1.1
2.2
NS
***
NS
1.8 ± 0.2b
1.3 ± 0.2c
2.4 ± 0.2a
ICDH
−45.3
−45.3
−45.3
1,048.5
1,140.1
999.5
NS
***
NS
1,004 ± 75b
1,095 ± 76a
955 ± 75c
LDH
Enzyme activity2
−0.75
−0.75
−0.75
3.1
4.4
4.0
*
***
NS
2.3 ± 0.1c
3.6 ± 0.1a
3.3 ± 0.1b
CINS
0.38
−0.31
−2.40
2.2
4.5
6.3
NS
***
*
2.6 ± 0.1c
4.2 ± 0.1a
3.9 ± 0.1b
TCOL
Collagen2
0.6
0.6
0.6
5.5
5.8
6.8
NS
***
NS
6.1 ± 0.1c
6.5 ± 0.1b
7.5 ± 0.1a
PL
1.1
1.1
1.1
20.0
6.7
15.5
NS
***
NS
21.0 ± 1.0a
7.8 ± 1.0c
16.5 ± 1.0b
TG
Lipids2
3.4
3.4
3.4
27.3
13.6
23.6
NS
***
NS
30.7 ± 1.0a
17.0 ± 1.0c
27.0 ± 1.0b
TLIP
1
Within a column, means without a common superscript letter differ (P < 0.05).
The number of samples used for analysis is given as a range as some muscle characteristics have more observations than others.
2
CSA = average muscle fiber cross-sectional area (µm2); %SO = percentage of slow oxidative fibers; %FOG = percentage of fast oxidative-glycolytic fibers; %FG = percentage of fast glycolytic fibers; ICDH = activity of isocitrate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); LDH = activity of lactate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); CINS = insoluble collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); TCOL = total collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); PL = phospholipid concentration (mg⋅g−1 of fresh muscle); TG = triglyceride
concentration (mg⋅g−1 of fresh muscle); and TLIP = total lipid concentration (mg⋅g−1 of fresh muscle).
3
Parameter estimates are for the linear mixed model equation: Yij = α1LTij + α2STij + α3TBij + γ1i + (β1LTij + β2STij + β3TBij + γ2i)DoMij + εij. Where: i = 1, ……, 33 experiments and j = 1, ……, ni
values of the continuous variable, DoM; Yij = the expected outcome of the dependent variable Y observed at DoM j in the experiment i; LTij, STij, TBij = dummy variables that take on the value
of 1 or 0 to indicate the muscle being considered for an observation at DoM j and in experiment i; DoMij = the value j of the continuous variable, DoM, in experiment i; α1–3 and β1–3 = coefficients
for the intercept and linear fixed effects, respectively, when regressing Y on DoM for the muscles LT, ST and TB; γ1i and γ2i = the random intercept and linear effect of experiment; εij = the unexplained residual error.
NS = not significant; * = P < 0.05; ** = P < 0.01; *** = P < 0.001.
a–c
Least squares means
Longissimus thoracis
Semitendinosus
Triceps brachii
Significance of fixed effects
DoM
Muscle
DoM × muscle
Fixed effect estimates3
Intercept
Longissimus thoracis (α1)
Semitendinosus (α2)
Triceps brachii (α3)
Linear coefficient
Longissimus thoracis (β1)
Semitendinosus (β2)
Triceps brachii (β3)
Item
Muscle fibers2
Table 4. Least squares means (±SEM) and parameter estimates for the relationship of muscle characteristics with degree of maturity (DoM) for the longissimus thoracis, semitendinosus, or triceps brachii muscles of cows
2878
Schreurs et al.
19–21
50–80
35–366
26–58
83–108
n1
**
NS
*
NS
NS
NS
NS
−1.2
4.3
9.0
12.8
11.6
19.1
13.4
2.5
12.7
−1.7
***
***
**
**
NS
NS
NS
−3,603.1
185.0
−67.6
922.9
−392.3
14,730.6
10,530.5
10,141.5
6,143.6
10,631.0
−3,982.4
−3,982.4
−3,982.4
−3,982.4
−3,982.4
7.3 ± 3.6b
10.2 ± 2.1b
10.1 ± 1.3b
18.4 ± 1.8a
10.9 ± 2.0b
%SO
1,829 ± 735c
3,792 ± 419a
3,370 ± 264ab
2,623 ± 358bc
3,258 ± 298ab
CSA
46.1
46.1
46.1
46.1
46.1
−87.7
−64.4
−61.8
−47.5
−67.4
65.4
42.7
44.2
32.8
50.9
***
*
**
***
NS
NS
NS
38.3 ± 5.3a
25.9 ± 3.3bc
28.6 ± 2.3bc
23.4 ± 3.3c
32.8 ± 3.2ab
%FOG
−42.7
−42.7
−42.7
−42.7
−42.7
63.0
48.3
57.5
35.2
60.7
37.9
52.9
46.3
54.0
42.1
***
NS
***
***
NS
NS
NS
54.9 ± 6.1
63.4 ± 4.2
60.9 ± 3.2
58.8 ± 4.6
58.1 ± 4.0
%FG
1.7
7.8
6.2
1.7
−1.5
−2.1
−11.5
−6.9
−3.4
2.0
1.9
5.3
2.8
3.4
0.7
***
NS
***
***
***
NS
NS
1.3 ± 0.9ab
1.9 ± 0.4ab
1.2 ± 0.1b
2.2 ± 0.4a
1.3 ± 0.2ab
ICDH
2,703.8
838.2
−1,304.4
−4,179.1
−279.0
−4,177.1
−1,488.8
1,658.2
3,152.9
−13.8
2,661.9
1,650.0
661.3
494.2
1,249.2
***
NS
***
***
**
NS
NS
1,405 ± 263
1,168 ± 129
1,097 ± 72
856 ± 210
1,166 ± 79
LDH
Enzyme activity2
−0.32
−7.36
−1.09
−3.30
−4.09
7.6
7.4
−1.60
2.18
−1.00
4.9
10.6
5.6
***
***
**
NS
NS
NS
NS
5.1
2.5
4.6
*
***
***
NS
NS
NS
NS
−2.3
−2.3
−2.3
−2.3
8.4
8.7
8.5
8.4
***
NS
NS
NS
NS
NS
NS
6.7 ± 0.2
4.8 ± 0.1a 5.5 ± 0.1a
PL
7.0 ± 0.2
6.8 ± 0.1
6.7 ± 0.2
TCOL
4.1 ± 0.3b 4.7 ± 0.4b
3.9 ± 0.2b 5.8 ± 0.2a
3.9 ± 0.1b 4.9 ± 0.2b
CINS
Collagen2
−3.0
−3.0
−3.0
−3.0
6.3
6.9
6.9
6.8
NS
NS
NS
NS
NS
NS
NS
4.1 ± 0.9
4.7 ± 0.9
4.7 ± 0.7
4.6 ± 0.9
TG
Lipids2
0.2
0.2
0.2
0.2
12.4
13.3
14.0
13.4
NS
NS
NS
NS
NS
NS
NS
12.5 ± 0.9
13.4 ± 0.9
14.1 ± 0.7
13.5 ± 0.9
TLIP
1
Within a column, means without a common superscript letter differ (P < 0.05).
The number of samples used for analysis is given as a range as some muscle characteristics have more observations than others.
2
CSA = average muscle fiber cross-sectional area (µm2); %SO = percentage of slow oxidative fibers; %FOG = percentage of fast oxidative-glycolytic fibers; %FG = percentage of fast glycolytic fibers; ICDH = activity of isocitrate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); LDH = activity of lactate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); CINS = insoluble collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); TCOL = total collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); PL = phospholipid concentration (mg⋅g−1 of fresh muscle); TG = triglyceride
concentration (mg⋅g−1 of fresh muscle); and TLIP = total lipid concentration (mg⋅g−1 of fresh muscle).
3
Parameter estimates for the linear mixed model equation: Yij = α1AUij + α2CHij + α3LIij + α4MOij + α5SAij + γ1i + (β1AUij + β2CHij + β3LIij + β4MOij + β5SAij + γ2i)DoMij + (δ1AUij + δ2CHij + δ3LIij +
δ4MOij + δ5SAij + γ3i)DoMij2 + εij, where i = 1, ……, 33 experiments and j = 1, ……, ni values of the continuous variable, DoM; Yij = the expected outcome of the muscle characteristic Y, observed at
DoM j in the experiment i; AUij, CHij, LIij, MOij, SAij = dummy variables that have a value of 1 or 0 to indicate the breed being considered for an observation at DoM j and in experiment i; DoMij
= the value j of the continuous variable, DoM, in experiment i; α1–5, β1–5, δ1–5 = coefficients for the intercept, linear, quadratic, and cubic effects, respectively, when regressing Y on DoM for the
different breeds; γ1i, γ2i, γ3i = the random intercept, linear, quadratic, and cubic effect of experiment; and εij = the unexplained residual error.
NS = not significant; *P < 0.05; **P < 0.01; ***P < 0.001.
a–c
Least squares means
Aubrac
Charolais
Limousin
Montbéliard
Salers
Significance of fixed effects
DoM
Breed
DoM × breed
DoM2
DoM2 × breed
DoM3
DoM3 × breed
Fixed effect estimates3
Intercept
Aubrac (α1)
Charolais (α2)
Limousin (α3)
Montbéliard (α4)
Salers (α5)
Linear coefficient
Aubrac (β1)
Charolais (β2)
Limousin (β3)
Montbéliard (β4)
Salers (β5)
Quadratic coefficient
Aubrac (δ1)
Charolais (δ2)
Limousin (δ3)
Montbéliard (δ4)
Salers (δ5)
Item
Muscle fibers2
Table 5. Least squares means (±SEM) and parameter estimates for the relationship of muscle characteristics with degree of maturity (DoM) for the semitendinosus muscle from different breeds of bulls
Meta-analysis of muscle characteristics
2879
18–21
22
19–108
21
n1
11.3 ± 1.1ab
9.5 ± 1.1bc
9.0 ± 0.5c
12.5 ± 1.1a
NS
*
NS
12.3
10.5
10.0
13.5
−1.0
−1.0
−1.0
−1.0
NS
***
NS
3,194.0
3,346.8
4,503.7
3,007.5
786.2
786.2
786.2
786.2
%SO
3,976 ± 330b
4,129 ± 324b
5,286 ± 150a
3,790 ± 340b
CSA
−11.8
−11.8
−11.8
−11.8
38.6
37.5
33.3
37.2
NS
**
NS
26.8 ± 1.6a
25.7 ± 1.6a
21.5 ± 0.7b
25.5 ± 1.7a
%FOG
12.8
12.8
12.8
12.8
49.1
52.0
56.8
49.2
NS
***
NS
61.9 ± 1.8b
64.8 ± 1.8b
69.6 ± 0.8a
62.0 ± 1.9b
%FG
LDH
0.14
0.14
0.14
0.14
0.8
0.7
0.8
0.8
NS
*
NS
80.8
80.8
80.8
80.8
899.3
843.3
1,017.3
798.0
NS
***
NS
0.95 ± 0.06a
978 ± 76b
b
0.78 ± 0.06
922 ± 73bc
0.97 ± 0.03a 1,096 ± 75a
0.90 ± 0.06a
876 ± 74c
ICDH
Enzyme activity2
1.96
−1.91
−5.11
−2.32
1.4
5.8
8.2
6.3
**
**
**
3.5 ± 0.1b
3.9 ± 0.1a
2.9 ± 0.2c
4.0 ± 0.1a
CINS
1.29
−1.70
−3.09
−2.93
2.8
6.1
6.8
7.6
*
*
*
4.1 ± 0.1b
4.3 ± 0.1ab
3.7 ± 0.2c
4.6 ± 0.1a
TCOL
Collagen2
−0.08
−0.08
−0.08
−0.08
6.5
6.5
6.7
6.6
NS
NS
NS
6.4 ± 0.1
6.4 ± 0.1
6.6 ± 0.1
6.5 ± 0.1
PL
−1.81
−1.81
−1.81
−1.81
11.0
10.5
7.7
9.2
NS
*
NS
9.1 ± 1.0a
8.6 ± 0.9a
5.8 ± 1.0b
7.3 ± 1.0ab
TG
Lipids2
0.03
0.03
0.03
0.03
19.0
17.6
14.4
16.8
NS
*
NS
19.1 ± 1.0a
17.7 ± 1.0a
14.4 ± 1.1b
16.9 ± 1.0ab
TLIP
1
Within a column, means without a common superscript letter differ (P < 0.05).
The number of samples used for analysis is given as a range as some muscle characteristics have more observations than others.
2
CSA = average muscle fiber cross-sectional area (µm2); %SO = percentage of slow oxidative fibers; %FOG = percentage of fast oxidative-glycolytic fibers; %FG = percentage of fast glycolytic fibers; ICDH = activity of isocitrate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); LDH = activity of lactate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); CINS = insoluble collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); TCOL = total collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); PL = phospholipid concentration (mg⋅g−1 of fresh muscle); TG = triglyceride
concentration (mg⋅g−1 of fresh muscle); and TLIP = total lipid concentration (mg⋅g−1 of fresh muscle).
3
Parameter estimates for the linear mixed model equation: Yij = α1AUij + α2CHij + α3LIij + α4SAij + γ1i + (β1AUij + β2CHij + β3LIij + β4SAij + γ2i)DoMij + εij, where i = 1, ……, 36 experiments and j
= 1, ……, ni values of the continuous variable age; Yij = the expected outcome of the dependent variable Y observed at age j in the experiment i; AUij, CHij, LIij, SAi = dummy variables that take
on the value of 1 or 0 to indicate the breed being considered for an observation at age j and in experiment i; DoMij = the value j of the continuous variable, DoM, in experiment i; α1–5 and β1–5 =
the average intercepts and regressing coefficients of Y on age for the different breeds; γ1i and γ2i = the intercept and slope for the random effect of experiment; and εij = the unexplained residual
error.
NS = not significant; *P < 0.05; **P < 0.01; ***P < 0.001.
a–c
Least squares means
Aubrac
Charolais
Limousin
Salers
Significance of fixed effects
DoM
Breed
DoM × breed
Fixed effect estimates3
Intercept
Aubrac (α1)
Charolais (α2)
Limousin (α3)
Salers (α4)
Linear coefficient
Aubrac (β1)
Charolais (β2)
Limousin (β3)
Salers (β4)
Item
Muscle fibers2
Table 6. Least squares means (±SEM) and parameter estimates for the relationship of muscle characteristics with degree of maturity (DoM) for the semitendinosus muscle from different breeds of cows
2880
Schreurs et al.
156–297
187–571
n1
%SO
9.8
10.2
5.0
5.0
962.6
243.3
6,015.1
8,483.2
−2,147.9
−2,147.9
**
NS
NS
NS
NS
NS
NS
***
NS
**
**
NS
NS
NS
3,174 ± 217b 12.1 ± 1.6
3,583 ± 159a 12.5 ± 1.2
CSA
−63.5
−63.5
231.7
134.7
−150.2
−90.7
48.8
42.5
***
NS
***
***
**
**
NS
29.2 ± 2.5
25.4 ± 1.8
%FOG
−261.1
−43.9
208.4
54.1
26.3
48.3
***
NS
***
***
***
NS
NS
55.1 ± 3.0b
61.8 ± 2.2a
%FG
9.4
9.4
−11.3
−11.3
4.5
4.8
**
***
NS
***
NS
NS
NS
1.5 ± 0.1b
1.8 ± 0.1a
ICDH
1,524.4
1,524.4
−3,258.1
−3,258.1
2,269.1
2,269.1
639.8
560.2
***
***
NS
***
NS
***
NS
1,111 ± 54a
1,031 ± 53b
LDH
Enzyme activity2
−1.00
−1.00
3.9
4.3
NS
NS
NS
NS
NS
NS
NS
3.8 ± 0.2
4.2 ± 0.2
CINS
−2.79
−2.79
6.9
6.8
NS
NS
NS
NS
NS
NS
NS
5.2 ± 0.2
5.1 ± 0.2
TCOL
Collagen2
1.7
−1.5
6.1
7.9
NS
*
**
NS
NS
NS
NS
7.2 ± 0.3
7.0 ± 0.3
PL
16.5
−2.5
−2.5
6.4
*
*
**
NS
NS
NS
NS
8.3 ± 0.9a
4.8 ± 1.0b
TG
Lipids2
22.1
1.3
3.4
12.8
***
*
**
NS
NS
NS
NS
17.8 ± 1.3a
13.6 ± 1.4b
TLIP
1
Within a column, means without a common superscript letter differ (P < 0.05).
The number of samples used for analysis is given as a range as some muscle characteristics have more observations than others.
2
CSA = average muscle fiber cross-sectional area (µm2); %SO = percentage of slow oxidative fibers; %FOG = percentage of fast oxidative-glycolytic fibers; %FG = percentage of fast glycolytic fibers; ICDH = activity of isocitrate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); LDH = activity of lactate dehydrogenase (µmol⋅min−1⋅g−1 of fresh muscle); CINS = insoluble collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); TCOL = total collagen concentration (µg of OH-proline⋅mg−1 of dry muscle); PL = phospholipid concentration (mg⋅g−1 of fresh muscle); TG = triglyceride
concentration (mg⋅g−1 of fresh muscle); and TLIP = total lipid concentration (mg⋅g−1 of fresh muscle).
3
Parameter estimates are for the linear mixed model equation: Yij = α1Steersij + α2Bullsij + γ1i + (β1Steersij + β2Bullsij + γ2i)DoMij + (δ1Steersij + δ2Bullsij + γ3i)DoMij2 + (θ1Steersij + θ2Bullsij +
γ4i)DoMij3 + εij, where i = 1, ……, 33 experiments and j = 1, ……, ni values of the continuous variable, DoM; Yij = the expected outcome of the muscle characteristic Y, observed at DoM j in the
experiment i; Steersij, Bullsij = dummy variables that have a value of 1 or 0 to indicate the castration status being considered for an observation at DoM j and in experiment i; DoMij = the value
j of the continuous variable, DoM, in experiment i; α1–2, β1–2, δ1–2, θ1–2 = coefficients for the intercept, linear, quadratic, and cubic effects, respectively, when regressing Y on DoM for the different
castration status; γ1i, γ2i, γ3i, γ4i = the random intercept, linear, quadratic, and cubic effects of experiment; and εij = the unexplained residual error.
NS = not significant; *P < 0.05; **P < 0.01; ***P < 0.001.
a,b
Least squares means
Steers
Bulls
Significance of fixed effects
DoM
Castration
DoM × castration
DoM2
DoM2 × castration
DoM3
DoM3 × castration
Fixed effect estimates3
Intercept
Steers (α1)
Bulls (α2)
Linear coefficient
Steers (β1)
Bulls (β2)
Quadratic coefficient
Steers (δ1)
Bulls (δ2)
Cubic coefficient
Steers (θ1)
Bulls (θ2)
Item
Muscle fibers2
Table 7. Least squares means (±SEM) and parameter estimates for the relationship of muscle characteristics with degree of maturity (DoM) for the semitendinosus muscle from steers and bulls
Meta-analysis of muscle characteristics
2881
2882
Schreurs et al.
Figure 2. Percentage of slow oxidative (%SO, □), fast oxidative-glycolytic (%FOG, ♦), and fast glycolytic fibers
(%FG, ∆) in the semitendinosus muscle of bulls (A) and cows (B) when regressed with the degree of maturity of
the animal.
(P < 0.001) and quadratic (P < 0.01) coefficients were
different between the bull breeds (Table 5). Bulls had
less mean LDH activity than steers (P < 0.05; Table 7)
but linear, quadratic, and cubic coefficients for the relationship with DoM were the same between sexes.
Insoluble and Total Collagen Concentration. The CINS and TCOL concentration in the mus-
cle of both bulls and cows showed linear changes with
increasing DoM (Tables 3 and 4). In both bulls and
cows, the mean CINS concentration in dry muscle was
greatest in the ST muscle and least in the LT muscle
(P < 0.05; Tables 3 and 4). In the bulls, the linear coef-
ficient was different between the muscles (P < 0.001;
Table 3). The linear coefficient was positive for LT
muscle, negative for ST muscle, and close to zero for
TB muscle. Mean CINS concentration was greater in
the muscle of Salers bulls and greater in Salers and
Charolais cows compared with other breeds (Tables 5
and 6). The linear coefficient when regressing CINS
concentration against DoM differed for different bull (P
< 0.001) and cow breeds (P < 0.01). The Charolais bulls
and Aubrac cows had a positive linear coefficient but
it was negative for other bull and cow breeds (Table 5
and 6). Mean CINS concentration and the relationship
Figure 3. Quadratic relationship for bulls (A) and linear relationship for cows (B) when isocitrate dehydrogenase (ICDH) activity of fresh longissimus thoracis (LT, □), semitendinosus (ST, ∆), or triceps brachii (TB, ♦) muscle
is regressed with the degree of maturity of the animal.
2883
Meta-analysis of muscle characteristics
Figure 4. Cubic relationship for bulls (A) and linear relationship for cows (B) when lactate dehydrogenase
(LDH) activity of fresh longissimus thoracis (LT, □), semitendinosus (ST, ∆), or triceps brachii (TB, ♦) muscle is
regressed with the degree of maturity of the animal.
of CINS concentration with DoM did not differ between
bulls and steers (Table 7).
Mean TCOL concentration was greatest in ST muscle and least in LT muscle in both bulls and cows (P
< 0.05; Tables 3 and 4). For the relationship of TCOL
concentration in the muscle with DoM in bulls, the linear coefficient was negative and the value differed for
each muscle type (P < 0.05; Table 3). In cows, the linear
coefficient was positive for LT muscle and negative for
ST and TB muscles (P < 0.05; Table 4). Mean TCOL
concentration was greater in Charolais and Salers
bulls and cows compared with Aubrac and Limousin
(P < 0.05; Tables 5 and 6). When TCOL concentration
was regressed with DoM the linear coefficient differed
between the bull (P < 0.01; Table 5) and cow breeds
(P < 0.05; Table 6). The linear coefficient was negative
for all except the Aubrac cows. Mean TCOL and linear
coefficients for the relationship of TCOL concentration
with DoM did not differ between bulls and steers.
Phospholipid, Triglyceride, and Total Lipid Concentration. Concentrations of PL, TG, and
TLIP in the muscle developed linearly with DoM in
bulls (Table 3; Figure 5) but did not change in cows
(Table 4; Figure 5).
Mean PL concentration was greatest in TB muscle
and least in LT muscle of both bulls and cows (P < 0.05;
Tables 3 and 4). For the relationship of PL concentration with DoM in bulls, the linear coefficient did not
differ between the muscles or breeds. In bulls the linear coefficient for the relationship of PL with DoM was
negative but in steers it was positive (P < 0.01; Table
7).
Mean TG concentration was greatest in the LT muscle and least in ST muscle for both bulls and cows (P
< 0.05; Tables 3 and 4). The linear coefficients were
different between the muscles for the relationship of
TG with DoM in bulls (P < 0.01; Table 3). The linear
coefficient was negative in the ST muscle and greater
in the LT muscle than the TB muscle. Mean TG concentration and linear coefficients for the relationship
with DoM did not differ between bull breeds. In cows,
Aubrac and Charolais breeds had a greater mean TG
concentration compared with Limousin (P < 0.05; Table 6). Mean TG was greater in steers than in bulls.
The linear coefficient for the relationship of TG with
DoM for steers was large and positive compared with
the negative coefficient with bulls (P < 0.01; Table 7).
Mean TLIP concentration was greatest in LT muscle and least in ST muscle in both bulls and cows (P <
0.05; Tables 3 and 4). Regression of TLIP concentration with DoM for bulls gave positive linear coefficients
for all muscles, indicating increasing TLIP concentration with DoM. The increase was greatest for LT and
lowest for ST muscle (P < 0.05; Figure 5). Mean TLIP
concentration and its development with DoM did not
differ between bull breeds. Mean TLIP was greater for
the Aubrac and Charolais cows compared with Limousin cows (P < 0.05; Table 6). Steers had a greater
mean TLIP compared with bulls. There was a greater
increase in muscle TLIP concentration in steers than
bulls (P < 0.01; Table 7).
DISCUSSION
Methodological Aspects
The objective of this study was to quantitatively review the development of 11 muscle characteristics with
increasing animal maturity in different muscle types,
breeds, and sexes of French cattle. Degree of maturity was used so that the muscle characteristics for the
different animal types could be compared at a similar
2884
Schreurs et al.
Figure 5. Linear relationship for bulls (A) and cows (B) when total lipid concentration (TLIP) in the fresh longissimus thoracis (LT, □), semitendinosus (ST, ∆), or triceps brachii (TB, ♦) muscle is regressed with the degree of
maturity of the animal.
physiological state. The use of original measurements,
as in this meta-analysis, is considered advantageous
because the data are not limited to that of published
means. This removes the possibility of publication
bias, which occurs when only selective results from a
study are published or when subjective judgments are
made when selecting publications for a meta-analysis
(Finney, 1995).
By defining the experiment from which data were
obtained as a random effect, the regression analysis
adjusts for the within- and between-experiment variation allowing the true variation due to the effects of
animal maturity, muscle type, breed, and sex to be considered across all experiments. Therefore, a broader inference space is utilized as the focus moves from the
experimental level to a larger set of levels constituting the population. This increases the likelihood that
inferences reflect the population and future observations (St-Pierre, 2001). The ability to detect the true
effects of the factors (maturity, muscle, breed, and
sex) is aided further by the measures from individual
samples being determined in the same laboratory using the same procedures. Thus, no variation is added
from different laboratory techniques. More data were
available for the fiber and enzyme characteristics than
for the collagen and lipid characteristics. This implies
that statistical power is greater for inference with the
fiber and enzyme results compared with the collagen
and lipid results.
Development of Muscle Characteristics
in Different Muscles of Cattle
The greatest changes in the muscle characteristics
with increasing DoM were seen in the immature bulls.
This indicates that the majority of change in muscle
characteristics occurs between birth and maturity. The
lack of change in the muscle characteristics of cows
reflects the steady state in muscle composition that is
reached in mature cattle. Jurie et al. (2006), using mature cows aged 4 to 9 yr, also found no changes in muscle characteristics with age. Thus, at a DoM >0.8, there
appears to be less change in muscle characteristics and
a plateau is reached. These changes in muscle characteristics correspond to the growth pattern in beef cattle
where BW changes most rapidly in young animals with
growth rate slowing toward maturity. Interactions of
the muscle type, breed, or castration with DoM, DoM2,
or DoM3 indicate that the development of muscle characteristics differs between the various muscles, breeds,
or sexes being considered. The regression analysis can
be used to evaluate how the muscle characteristics are
changing with DoM. A linear relationship indicates a
constant increase or decrease with increasing DoM. A
quadratic or cubic relationship indicates that at greater
DoM, the increase or decrease of muscle characteristics
is not proportional to that at less DoM.
Development of muscle characteristics with DoM is
unique for each muscle, and mature cattle have established noticeable differences in their characteristics
between muscles. Typically, postnatal growth is associated with an increased muscle fiber size because the
number of muscle fibers is fixed at birth (Wegner et al.,
2000). Hypertrophy of the muscle fibers explains the
increase in CSA with increasing DoM in young bulls.
The small increase in CSA in the cows represents reduced protein deposition in the muscles of older animals reaching their adult BW (Owens et al., 1993).
In accordance with the means in this study, previous
research has indicated that in bulls, the LT and TB
muscles have a smaller CSA than ST muscle (Jurie et
al., 2005). The CSA in LT and ST muscles increases
Meta-analysis of muscle characteristics
at less DoM but then declines at greater DoM as indicated by the positive linear coefficients and negative
quadratic coefficients. The greater linear coefficient for
the ST indicates that the increase in CSA is greater
for ST muscle compared with LT. The TB muscle is opposite in its CSA development with a decline in CSA at
early DoM followed by an increase at later DoM. This
confirms the muscle-specific allometry observed by
Brandstetter et al. (1998).
Jurie et al. (2006) showed that the ST muscle of cows
was more glycolytic and the TB muscle was more oxidative, which is consistent with the bull and cow results
in this meta-analysis. The literature indicates that glycolytic fiber percentage and glycolytic enzyme activity
increase up to approximately 12 to 16 mo of age (Jurie
et al., 1995a) after which they decline with a change
toward an increasingly oxidative fiber composition and
activity (Jurie et al., 2005). Puberty occurs in cattle at
a DoM of approximately 0.5 (Smith et al., 1976). The
DoM near puberty represents a period when an inflection point is reached in the quadratic and cubic relationships established for the fiber type percentages in
this meta-analysis. In parallel to the decreasing proportion of oxidative fibers and increasing proportion of
glycolytic muscle fibers before puberty, the activity of
the oxidative enzyme ICDH decreased and the activity
of LDH (a glycolytic enzyme) increased. The dynamic
nature of the fiber composition and metabolic activity
with DoM reflects energy requirements of the muscle
during development and occurs in response to substrate supply and to chemical, hormonal, and neural
signals (Hocquette et al., 1998; Oddy et al., 2001). The
changes observed in this meta-analysis reflect the use
of glycolytic metabolism to supply energy during periods of rapid growth before puberty (Brandstetter et al.,
1998). Changes in the developmental rate of metabolic
enzyme activity and fiber percentage occur at the same
DoM for all muscles but the magnitude of the change
differs between muscles because of their different anatomical locations and different requirements for nutrients and energy in response to physiological function
(Lefaucheur and Gerrard, 2000; Oddy et al., 2001). The
greatest changes in metabolic enzyme activity were
seen in the ST muscle, which may reflect the greater
energy requirement of this muscle due to its role in animal movement.
Previous research indicates that the ST muscle has
the greatest collagen content and the LT had the least
collagen content (Jurie et al., 2005, 2006; Von Seggern
et al., 2005), which was also found with the results
from bulls and cows in this meta-analysis. Aging of
intramuscular collagen has been noted to increase its
thermal stability and reduce its solubility with cooking
(Nishimura et al., 1996). This meta-analysis found that
in young bulls, CINS increased in the LT muscle and
decreased in ST, with little change in the TB muscle
with increasing animal maturity. This suggests that
only the LT muscle is susceptible to reductions in collagen solubility as the animal grows. Nishimura et al.
2885
(1996) studied collagen concentration in 7-mo-old bovine fetuses through to 36-mo-old steers and indicated
that the majority of change in collagen concentration
and solubility in the muscle occurred before 6 mo of
age; after that time, the total and soluble collagen decreased only fractionally. Given this, the major changes
in collagen characteristics are not likely to be observed
in this study as collagen data could not be obtained for
animals with low DoM.
In accordance with the means in our study, the literature shows that LT muscle has a greater lipid concentration compared with ST muscle, and TB muscle
is intermediate (Jurie et al., 2005; Von Seggern et al.,
2005). The increased TLIP in the muscles of bulls reflects increased fat deposition as protein deposition declines toward maturity (Robelin, 1986; Owens et al.,
1993; Hocquette et al., 1998). The plateau in lipid concentration of cows can be attributed to the reduced rate
of fat deposition observed in older animals (Owens et
al., 1993). Furthermore, cull cows for beef production
in France are slaughtered at a similar body condition
and this probably equalized the muscle lipid concentration across the cows. Phospholipids are structural
lipids found in muscle cell membranes and the small
decline in PL with DoM in bulls is likely to be a result
of other muscle cell components increasing at a greater
rate. Triglycerides are the major component of total
lipid; hence, the changes in TG tended to mimic the
changes in TLIP.
Influence of Breed on Muscle
Characteristics
Few differences in muscle characteristics were observed between Aubrac, Charolais, Limousin, and Salers bulls at 15 to 24 mo of age by Jurie et al. (2005). In
this meta-analysis, a greater number of animals and an
additional dairy breed were considered and the muscle
characteristics were found to differ between breeds.
Breeds differ in their BW at maturity and the length
of time to reach maturity. Dairy breeds such as Holstein
and Montbéliard tend to be early maturing, whereas
beef breeds such as Charolais and Limousin are later
maturing. The rustic breeds, Aubrac and Salers, are
used primarily for beef production although they have
less mature BW and are earlier maturing than the beef
breeds. Use of DoM allows for comparison between
breeds with differing mature BW and rate of maturity.
Differences in muscle characteristics between breeds
are likely to be a consequence of metabolic and physiological differences in the breeds that have evolved in
response to production purposes. The smaller linear
coefficient indicates that there was a smaller increase
in CSA with DoM in Montbéliard bulls compared with
other breeds. Furthermore, the dairy and rustic breeds
tended to have a smaller mean CSA and greater oxidative fiber type and enzymatic activity. Glycolytic metabolism is used to provide energy for muscle growth,
and fiber percentages are altered at different rates be-
2886
Schreurs et al.
tween the breeds to meet energy requirements (Hocquette et al., 1998; Oddy et al., 2001). The fiber proportions and metabolic enzyme activities of the dairy
and rustic breeds are likely to be due to these breeds
having less propensity toward protein deposition and
lean muscle growth compared with other breeds, and,
therefore, requiring less muscle fiber hypertrophy and
less demand for glycolytic metabolism.
The breeds differed in rates of change in CINS and
TCOL, but this is likely because of changes in collagen
that were not proportional to breed-dependent changes
in other muscle components resulting in dilution or
concentration of collagen in the muscle with increasing
maturity (Gerrard et al., 1987). Salers bulls and cows
had the greatest CINS and TCOL compared with other
breeds. Earlier maturing breeds tend to deposit more
collagen with a greater insoluble proportion (Campo
et al., 2000), which may account partly for the greater
CINS in Salers in this study.
The concentration of lipid in the muscle and its development with DoM was not different between the
bull breeds. This indicates that maturity level is a
strong driver of lipid development in different cattle
breeds. Although the mean TG and TLIP were different between cow breeds, the maximum difference was
only 0.3% of the muscle mass, which may not lead to
noticeable differences in meat quality.
Influence of Sex on Muscle Characteristics
Different muscle characteristics between bulls and
steers are likely a result of differences in gonadal hormones, notably testosterone, in animals after puberty.
Testosterone is a driver for rapid BW gain and metabolism that promotes lean muscle growth as opposed to
fat deposition (Seideman et al., 1982). This probably
promoted the greater mean CSA and greater increase
in CSA with DoM in bulls than steers. Testosterone has a role in controlling changes in muscle fibers
(Seideman and Crouse, 1986), which may have been
responsible for the greater mean %FG and the smaller
changes in the respective cubic and quadratic relationship of %FOG and %FG with DoM that were observed
in bulls compared with steers. Despite the greater proportion of glycolytic fibers, LDH activity was less and
oxidative activity greater in bulls. Oxidative metabolic
activity utilizes fatty acids as an energy source (Ashmore, 1974) and is likely to be the link to the reduced
lipid concentration and deposition in bulls compared
with steers. Testosterone has a stimulating effect on
collagen synthesis (Gerrard et al., 1987) although this
was not evident in our study, which considered many
breeds.
Hypotheses for Model Development
This study indicates that muscle characteristics are
not likely to follow a linear development with increasing maturity. Nonlinear analysis and modeling would
provide better predictive equations that are more use-
ful for decision-making. The muscle characteristics
of each muscle develop following the same general
pattern with increasing maturity; however, rates at
which the muscle characteristics change differ among
muscles. When regressing the muscle characteristics
against DoM, the coefficient values and means differed
between muscles, indicating that muscle type accounts
for the largest proportion of the variation in muscle
characteristics. For modeling, this suggests that the
same nonlinear equation can be used for all muscles
but each muscle will have to be modeled separately.
Within a muscle, the breeds and sexes also differed in
their coefficients and mean values; therefore, breed and
sex have a role in determining muscle characteristics
and meat quality. Amplitude and rates of change in the
muscle characteristics with different sexes and breeds
could be modeled by fitting different parameter values
for the different breeds and sexes within a muscle. To
minimize the number of parameters that need to be
fitted, the meta-analysis indicates that breeds could
be grouped. The dairy and rustic breeds differ in their
muscle characteristics, in particular fiber composition
and enzyme activity, compared with the beef breeds
so it might be possible to group into beef and nonbeef
breeds. Similarly, cow and steer data can be grouped so
that parameters are fitted for male and nonmale animals because changes in muscle characteristics with
different sexes appear to be strongly influenced by
testosterone production. To improve the functionality
for decision-making, it is envisioned that the effect of
DoM on muscle characteristics will be combined with a
growth model.
LITERATURE CITED
Ashmore, C. R. 1974. Phenotypic expression of muscle fiber types
and some implications to meat quality. J. Anim. Sci. 38:1158–
1164.
Brandstetter, A. M., B. Picard, and Y. Geay. 1998. Muscle fiber characteristics in four muscles of growing male cattle I. Postnatal
differentiation. Livest. Prod. Sci. 53:15–23.
Campo, M. M., P. Santolaria, C. Sañudo, J. Lepetit, J. L. Olleta, B.
Panea, and P. Alberti. 2000. Assessment of breed type and ageing time effects on beef meat quality using two different texture
devices. Meat Sci. 55:371–378.
Finney, D. J. 1995. A statistician looks at met-analysis. J. Clin. Epidemiol. 48:87–103.
Fitzhugh, H. A. Jr., and St. C. S. Taylor. 1971. Genetic analysis of
degree of maturity. J. Anim. Sci. 33:717–725.
Gerrard, D. E., S. J. Jones, E. D. Aberle, R. P. Lemenager, M. A.
Diekman, and M. D. Judge. 1987. Collagen stability, testosterone secretion and meat tenderness in growing bulls and steers.
J. Anim. Sci. 65:1236–1242.
Hocquette, J.-F., I. Ortigues-Marty, D. Pethick, P. Herpin, and X.
Fernandez. 1998. Nutritional and hormonal regulation of energy metabolism in skeletal muscles of meat-producing animals.
Livest. Prod. Sci. 56:115–143.
Jurie, C., J.-F. Martin, A. Listrat, R. Jailler, J. Culioli, and B. Picard.
2005. Effects of age and breed of beef bulls on growth parameters, carcass and muscle characteristics. Anim. Sci. 80:257–
263.
Jurie, C., J.-F. Martin, A. Listrat, R. Jailler, J. Culioli, and B. Picard.
2006. Carcass and muscle characteristics of beef cull cows between 4 and 9 years of age. Anim. Sci. 82:415–421.
Meta-analysis of muscle characteristics
Jurie, C., J. Robelin, B. Picard, and Y. Geay. 1995a. Post-natal changes in the biological characteristics of semitendinosus muscle in
male Limousin cattle. Meat Sci. 41:125–135.
Jurie, C., J. Robelin, B. Picard, G. Renand, and Y. Geay. 1995b. Inter-animal variation in the biological characteristics of muscle
tissue in male Limousin cattle. Meat Sci. 39:415–425.
Lefaucheur, L., and D. Gerrard. 2000. Muscle fiber plasticity in farm
mammals. J. Anim. Sci. 77(E Suppl.):1–19.
McPhee, M. J., J. W. Oltjen, T. R. Famula, and R. D. Sainz. 2006. Meta-analysis of factors affecting carcass characteristics of feedlot
steers. J. Anim. Sci. 84:3143–3154.
Nishimura, T., K. Ojima, A. Liu, A. Hattori, and K. Takahashi. 1996.
Structural changes in the intramuscular connective tissue during development of bovine semitendinosus muscle. Tissue Cell
28:527–536.
Oddy, V. H., G. S. Harper, P. L. Greenwood, and M. B. McDonagh.
2001. Nutritional and developmental effects on the intrinsic
properties of muscles as they relate to the eating quality of beef.
Aust. J. Exp. Agric. 41:921–942.
Owens, F. N., P. Dubeski, and C. F. Hanson. 1993. Factors that alter the growth and development of ruminants. J. Anim. Sci.
71:3138–3150.
Picard, B., J. Robelin, and Y. Geay. 1995. Influence of castration and
postnatal energy restriction on the contractile and metabolic
characteristics of bovine muscle. Ann. Zootech. 44:347–357.
Renand, G., B. Picard, C. Touraille, P. Berge, and J. Lepetit. 2001.
Relationships between muscle characteristics and meat quality
traits of young Charolais bulls. Meat Sci. 59:49–60.
2887
Robelin, J. 1986. Growth of adipose tissues in cattle; partitioning
between depots, chemical composition and cellularity: A review.
Livest. Prod. Sci. 14:349–364.
Seideman, S. C., H. R. Cross, R. R. Oltjen, and B. D. Schanbacher.
1982. Utilization of the intact male for red meat production: A
review. J. Anim. Sci. 55:826–840.
Seideman, S. C., and J. D. Crouse. 1986. The effects of sex condition,
genotype and diet on bovine muscle fiber characteristics. Meat
Sci. 17:55–72.
Smith, G. M., H. A. Fitzhugh Jr., L. V. Cundiff, T. C. Cartwright, and
K. E. Gregory. 1976. Heterosis for maturing patterns in Hereford, Angus and Shorthorn cattle. J. Anim. Sci. 43:380–388.
Spilke, J., H. P. Piepho, and X. Hu. 2005. Analysis of unbalanced
data by mixed linear models using the MIXED procedure of the
SAS system. J. Agron. Crop Sci. 191:47–54.
St-Pierre, N. R. 2001. Invited review: Integrating quantitative findings from multiple studies using mixed model methodology. J.
Dairy Sci. 84:741–755.
Von Seggern, D. D., C. R. Calkins, D. D. Johnson, J. E. Brickler,
and B. L. Gwartney. 2005. Muscle profiling: Characterizing the
muscles of the beef chuck and round. Meat Sci. 71:39–51.
Wegner, J., E. Albrecht, I. Fiedler, F. Teuscher, H.-J. Papstein, and
K. Ender. 2000. Growth- and breed-related changes of muscle
fiber characteristics in cattle. J. Anim. Sci. 78:1485–1496.