Subscriber access provided by Karolinska Institutet, University Library
Article
Sensory and Flavor Chemistry Characteristics of Australian Beef; the Influence of Intramuscular Fat, Feed and Breed. Damian Conrad Frank, Alex J Ball, Joanne M Hughes, Udayasika Piyasiri, Janet Stark, Peter Watkins, and Robyn Dorothy Warner J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.6b00160 • Publication Date (Web): 27 Apr 2016 Downloaded from http://pubs.acs.org on May 6, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 39
Journal of Agricultural and Food Chemistry
Sensory and Flavor Chemistry Characteristics of Australian Beef; the Influence of Intramuscular Fat, Feed and Breed.
Damian Frank*1, Alex Ball2, Joanne Hughes3, Raju Krishnamurthy1, Udayasika Piyasiri1, Janet Stark3, Peter Watkins4 and Robyn Warner4,5 1
Commonwealth Scientific Industrial Research Organisation (CSIRO), 11 Julius Ave, North
Ryde, NSW, 2113, Australia. 2
Meat & Livestock Australia (MLA), Level 1, 40 Mount Street, North Sydney, NSW, 2060,
Australia. 3
Commonwealth Scientific Industrial Research Organisation (CSIRO), 39 Kessels Rd,
Coopers Plains, Qld. 4108, Australia 4
Commonwealth Scientific Industrial Research Organisation (CSIRO), 671 Sneydes Rd.,
Werribee, Vic. 3030, Australia.
5
Current details: Faculty of Veterinary and Agricultural Science, The University of Melbourne, Royal Parade, Parkville, Vic 3010
*Corresponding author: Damian Frank, CSIRO, 11 Julius Ave, North Ryde, NSW 2113. Tel: +61 2 9490 8584 Fax: +61 2 9490 8499 E-mail:
[email protected]
ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 2 of 39
Intramuscular fat and beef flavor
1
Abstract
2
The sensory attributes and flavor chemistry of grilled beef striploins (M. Longissimus
3
lumborum, n=42) varying widely in marbling from commercial production types typical for
4
Southern Australia, were extensively characterized. Striploins from Angus grass-fed
5
yearlings (5.2% - 9.9% intramuscular fat), Angus grain-finished steers (10.2% - 14.9%) and
6
Wagyu grass-fed heifers (7.8% - 17.5%) were evaluated. Inherent differences between
7
samples from grass and grain fed Angus cattle were minimal when the intramuscular fat
8
content was above ~ 5%. Wagyu samples had more intense flavor, higher tenderness and
9
juiciness compared to Angus grass fed samples. Grilled beef flavor, dairy fat and
10
sweetness increased with the marbling level and sourness and astringency decreased.
11
Tenderness and juiciness increased with marbling level and were correlated with Warner-
12
Bratzler peak force measurements. Trained panel sensory differences in flavor
13
corresponded with increases in aroma volatiles and changes non-volatile flavor
14
compounds. Unsaturated fatty acids with potential health benefits (vaccenic, rumenic acids)
15
increased with the level of marbling.
16 17 18 Keywords: Beef, flavor, Wagyu, Angus, marbling, pasture, Warner-Bratzler, olfactometry,
2 ACS Paragon Plus Environment
Page 3 of 39
Journal of Agricultural and Food Chemistry
Intramuscular fat and beef flavor 19
INTRODUCTION
20
Grilled beef flavor arises through a combination of thermally generated aroma volatiles and
21
non-volatile taste components delivered in a matrix of muscle fiber, connective tissue
22
(collagen), warmed-meat juices and partly dissolved fat. The amount of fat within the
23
muscle — the intramuscular fat (IMF) — plays a critical role in the beef eating experience.
24
While a positive relationship between IMF and palatability (tenderness and juiciness) is
25
well-established1-5, its impact on beef flavor is less certain6, although recent studies indicate
26
a positive association.4, 7
27
The amount of IMF within beef muscle is typically assessed as visual marbling on the
28
surface of the meat. In Australia, the Meat Standards Australia (MSA) marbling score
29
(MSA-MB) system is used to score the level of IMF, using a fine visual scale ranging from
30
100 (no visible fat) to a maximum of 1190, in increments of 10 units. The fatty acid
31
composition of IMF is known to be affected by feed, which may in turn affect meat flavor.
32
Previous research has demonstrated distinct grass-fed (pasture) or grain-fed (feedlot,
33
concentrate) flavors in beef.8, 9 Grass-fed beef is the dominant production system used in
34
Australia, although a substantial proportion of cattle are finished on a high energy grain diet.
35
Extensive research has been devoted to understanding the genetics of IMF deposition and
36
marbling.10,
37
eating quality, other breeds such as Angus can also attain high marbling levels, especially
38
on a high nutritional plane.10 In addition, a better understanding of breed-related sensory
39
differences with different levels of IMF would be useful.
40
Sensory “halo-effects” are known to play a confounding role in assessing meat flavor;
41
untrained or naïve consumers tend to rate flavor high, when other attributes such as
42
tenderness and juiciness are also high.4 A primary aim of this research was to objectively
43
evaluate marbling effects on beef flavor using a trained panel to minimize confounding
11
While the Wagyu breed is synonymous with high marbling and excellent
3 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 4 of 39
Intramuscular fat and beef flavor 44
sensory interactions as well as to identify objective chemical markers that may underpin
45
sensory differences. The impact of animal diet (pasture vs. grain) and breed (Wagyu vs.
46
Angus) on beef flavor, after taking into account differences in the fat content, was also an
47
important question.
48
MATERIALS AND METHODS
49
Chemicals
50
Solvents were chromatography grade and purchased from Merck-Millipore (Bayswater,
51
Australia). The GLC-20 fatty acid methyl ester standard, C7-C30 saturated alkane linear
52
retention index mix, the glucose oxidase assay kit, sodium L-lactate-3-13C, methyl
53
tricosanoate, 1,1,3,3-tetraethoxypropane (>96%) and methyl chloroformate reagents were
54
obtained from Sigma-Aldrich (Castle Hill, Australia). Volatile standard reference compounds
55
of greater than 95% purity were also supplied by Sigma-Aldrich except, 2-methylpropanal,
56
3-methylbutanal,
57
dimethylpyrazine, 2-ethyl-3,5-dimethylpyrazine, trimethylpyrazine, furfural, 2-phenylethanal,
58
2-nonanone, which were supplied by Givaudan (ex-Quest), Baulkham Hills, NSW, Australia.
59
2-ethylhexanol, decanal, benzaldehyde and 4-methylphenol were purchased from Fluka
60
(Darmstadt, Germany). Norvaline, individual amino acids and organic acid standards were
61
greater than 95% purity (Sigma-Aldrich).
2,3-pentanedione,
2-methylpyrazine,
2,6-dimethylpyrazine,
2,3-
62 63
Collection of Beef Samples
64
Animals from three typical “production types” were identified from commercial farms for use
65
in
66
(“WagyuGrass”), Robbins Island, Northwest Tasmania (Hammond Farms) (2), 100% full-
67
blood Angus steers finished on a mixed ration including wheat and potato waste for 150
the
study;
(1)
100%
full-blood Wagyu
(Japanese
Black)
grass-fed
heifers
4 ACS Paragon Plus Environment
Page 5 of 39
Journal of Agricultural and Food Chemistry
Intramuscular fat and beef flavor 68
days (“AngusGrain”), Tasmanian Feedlot Pty. Ltd., (Perth, Tasmania) and (3) 100% full-
69
blood Angus grass-fed yearlings (“AngusGrass”), Muirhead Enterprises, (Cape Grim,
70
Tasmania). The cattle were slaughtered in December 2012 at the Greenham Tasmania
71
Pty. Ltd. abattoir (Smithton, Tasmania). After overnight chilling, carcasses were graded by
72
meat inspectors and assigned MSA marbling scores (MSA-MB). Pasture-fed Wagyu is a
73
relatively unique product compared to traditional grain-finished Wagyu. The latter was not
74
available in Tasmania at the time of sample collection. Meat was purchased at commercial
75
wholesale prices. Replicate carcasses for each production type were selected according to
76
nominal marbling bands — low (n=5), medium (n=4) and high (n=5), within each breed/feed
77
combination, giving a total of n = 42 carcasses. These carcasses were labelled and tracked
78
into the boning room and striploins (M. Longissimus lumborum) were boned from the right
79
side of each carcass. Subcutaneous fat was removed and striploins were wet-aged in
80
vacuum for 28 days in a chiller (1 ± 1 oC), before freezing at –20 oC. Frozen striploins were
81
fabricated into standardized steaks (25 x 25 x 75 mm) using a band saw; steaks were
82
vacuum packed and stored at -20 oC until use.12
83
Carcass and Meat Physicochemical Measurements
84
Carcass data were collected as part of routine processing and MSA grading.12-14 These
85
included MSA-MB, hot carcass weight (HCWT), eye muscle area (EMA), dentition, and
86
ossification score, a measure of physiological maturity13. Ultimate pH (upH) and meat color,
87
lightness (L*), redness (a*) and yellowness (b*) were measured approximately 24-hours
88
post-mortem in the chiller according to published protocols.15 Total collagen (TC) and heat
89
soluble collagen (HSC) content in the muscle was determined by measuring the
90
hydroxyproline content in lyophilized muscle (~2 g) expressed as a percentage of wet
91
weight (total) or of the total (heat soluble) fraction.16 HSC samples were defatted with
92
chloroform/methanol and hydrolysate and standards were neutralized with 0.6M NaOH 5 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 6 of 39
Intramuscular fat and beef flavor 93
solution prior to the assay. The IMF (% w/w) content in the raw meat samples was
94
estimated using the method described in Thornton.17
95 96
Thiobarbituric acid reactive substances (TBARS) were measured in duplicate raw meat
97
samples (~2 g) according to published methods.15 The residual glycogen content of the
98
frozen muscle subsamples was measured using a rapid assay modification using H2SO4
99
addition.18 Samples were homogenized (1:10 w/v) in 30 mM HCl for 2 x 15 sec bursts,
100
centrifuged (3,000 rpm, 4°C, 10 minutes). Samples were analyzed for total glucosyl units by
101
incubating 50 µL (37°C, 90 minutes) with the addition of 500 µL of hydrolyzing enzyme
102
amyloglucosidase (Sigma-Aldrich, 1:200 dilution in 40 mM acetate buffer, pH 4.8). Total
103
glucosyl units (mg/g) (considered to be glycogen content) was determined by absorbance
104
at 540 nm in duplicate using a glucose assay kit.
105 106
Warner-Bratzler shear (WBS) force provides an objective measure of meat tenderness.19
107
After overnight defrosting at 4˚C, WBS was determined according to established
108
procedures20. Samples were weighed and cooked in plastic bags in a water bath (70˚C, 60
109
min) and cooled prior to measurement. The amount of water lost during cooking— WBS-
110
cook loss — was calculated by mass balance, expressed as a percent of initial weight (%
111
w/w).
112 113
Beef Grilling Protocol
114
Frozen steaks were thawed overnight on plastic trays at 4 °C and grilled at 220 °C on a
115
commercial clamshell grill (Silex, Marrickville, Australia), according to published protocols 12,
116
13
117
type, the lid was closed and samples were grilled to a final internal temperature of 57 oC
. A thermocouple probe was inserted into the middle of the first of five steaks of the same
6 ACS Paragon Plus Environment
Page 7 of 39
Journal of Agricultural and Food Chemistry
Intramuscular fat and beef flavor 118
(defined as “medium doneness”).21 “Grilling time” was recorded using a stopwatch for each
119
set of five steaks (in seconds). Moisture lost at various steps in grilling and resting was
120
recorded for replicate batches of steaks. The moisture lost during grilling — “grill cook loss”
121
(% w/w), was determined by weight difference using a calibrated balance before and
122
immediately after grilling. “Grill rest loss” (% w/w) was determined after resting grilled meat
123
for 3 minutes under loosely placed aluminum foil and measuring the mass of liquid left in
124
the foil . Grilling data for low (n=10) and high (n=10) IMF samples for each production type
125
(total n=60) were obtained. After resting, steaks were cut into small pieces (~10 g) and
126
immediately placed into a standard wine glass (labelled with unique 3-digit code) and
127
covered with a watch glass, before serving to panelists in individual sensory booths.
128
Sensory Descriptive Analysis
129
Human ethics approval was obtained (CSIRO LR15-2012-C) for the sensory testing.
130
Experienced assessors (nine females and one male, 51± 6 years) participated in five two-
131
hour training sessions conducted over a two week period to generate and define the
132
sensory vocabulary. Published beef lexicons and attributes were considered in the
133
development of the final sensory vocabulary.8, 22, 23 Reference standards were used to help
134
illustrate some attributes (supporting information, Table S1). Assessors were equally
135
exposed to samples representing the experimental design variables. Impressions of meat
136
tenderness and juiciness were given after 3 and 10 chews. Remaining undissolved
137
connective tissue and total number of chews were rated at the point just before swallow.
138
Attributes (except number of chews to swallow) were rated using a 100 mm line scale on a
139
computer screen using the Compusense® five sensory software (Release 4.6, Compusense
140
Inc., Guelph, ON, Canada). Performance was monitored and regular feedback was given
141
until panelists had a clear understanding of all attributes. Sensory descriptive profiling was
142
performed on all samples (n=42) in triplicate over a two week period, hence a total of 30 7 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 8 of 39
Intramuscular fat and beef flavor 143
sensory assessments for each attribute and beef sample. Randomization of presentation
144
order was determined using CycDesigN Software (VSN International, Hemel Hempstead,
145
United Kingdom). A total of 32 attributes were generated by the trained panel to measure
146
grilled beef sensory properties, mostly in agreement with those reported by others.3,
147
Panelists removed the cover of the wineglass and first assessed nine odor attributes
148
orthonasally in the headspace; odor impact, grilled beef, livery, metallic, bloody, caramel,
149
barnyard, hay/grainy and fishy. Panelist then placed one piece of grilled meat in their
150
mouth and after two chews assessed nine flavor attributes retronasally; flavor impact,
151
grilled beef, livery, metallic, bloody, dairy-fat, grassy, hay/grainy and fishy and three taste
152
attributes; sour/acidic, sweet and salty. After swallowing, five aftertaste attributes were
153
rated; acidic, metallic, astringency, oily mouth-coating and lingering. The second piece of
154
meat was used for rating texture attributes; tenderness and juiciness after three and ten
155
chews, number of chews to swallow and amount of connective tissue before swallowing.
156
Fatty Acid Methyl Ester Analysis
157
Subsamples (~2 g) of raw ground meat (from ~30 g sample) were homogenized in
158
chloroform:methanol (2:1) and left at room temperature for 2 hr. Saline (0.73% NaCl) was
159
added and samples were centrifuged (1000 rpm, 5 min, 25 °C). The organic layer was
160
removed and reduced in volume under vacuum for ~16 hr. One mL of tetrahydrofuran, 5 %
161
H2SO4 in methanol, internal standard (IS) — 2 mg/mL methyl tricosanoate in heptane —
162
were respectively added, the mixture vortexed and then heated at 70 °C for 2 hr. After
163
cooling, heptane (2 mL) and saturated NaCl solution (1 mL) was added and, after mixing,
164
the fatty acid methyl esters (FAMEs) were extracted with heptane (2 x 2 mL). The combined
165
organic extract was washed with NaHCO3 solution (5%, 1 mL). The FAMEs (1 µL, split
166
1:50) were separated using a Supelco SP-2560 capillary column (100 m, 0.25 mm, 0.2 µm)
167
in an Agilent 6890 gas chromatograph. The GC oven was isothermally heated at 180 °C
4, 7, 8
8 ACS Paragon Plus Environment
Page 9 of 39
Journal of Agricultural and Food Chemistry
Intramuscular fat and beef flavor 168
with helium as carrier (flow rate = 1.2 mL/min) with the injector heated at 250 °C. An FID
169
(250 °C) was used for detection (flow rates for H2, air and N2 were 45, 450 and 45 mL/min,
170
respectively). Identification was made using a standard FAME mix and standard anhydrous
171
milkfat (prepared in house). Reference samples were also analyzed using mass
172
spectrometry (Agilent 5793 mass selective detector) to facilitate identification.
173
replicates were used for each sample and mean values (n=42) used to calculate total
174
amount of each FAME in the total extracted fat (mg/g). Total amount of each lipid in the IMF
175
was calculated for each sample and expressed as amount of FAME (mg) per 100 g serving
176
(raw).
177
Volatile Extraction and Analysis by Gas Chromatography Mass Spectrometry
178
The collection method for volatiles was designed to mimic dynamic ‘in mouth’ volatile
179
release.24 Separate individual replicate freshly grilled steaks were prepared from low (n=6)
180
and high (n=6) IMF levels for each of the three production types (n=36 samples in total).
181
After resting, middle sections of each steaks of the same type were removed and pooled
182
and (60 g) was suspended in Milli-Q water (1:2 ratio, ~37oC) and homogenized to a fine
183
slurry. The meat suspension, with the addition of an internal standard (4-methyl-1-pentanol,
184
40 ng/g) were concentrated onto Tenax-TA traps (60/80 mesh size, 100 mg) for 30 minutes
185
at 37 oC and analyzed by gas chromatography-mass spectrometry (GC-MS) using an
186
Agilent (ex-Varian GC-MS 4000 ion trap system) according to published protocols.24 To
187
facilitate identification, selected samples were also analyzed by methanol chemical
188
ionization (CI) to obtain the mass of the [M+H]+ parent ion, where applicable. Mass spectral
189
matches were conducted with the NIST-Mass Spectral Search database (Version 2.0,
190
2002). Reference standards (St) were used to confirm the identity of most compounds.
191
Integrated area data were normalized to the IS and semi-quantitative data (mg/kg) were
192
estimated.
Two
9 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 10 of 39
Intramuscular fat and beef flavor 193
Gas Chromatography-Olfactometry
194
Grilled beef volatile Tenax extracts described in the previous section were also
195
simultaneously evaluated by gas chromatography-olfactometry (GC-O) with time intensity
196
(TI) sensory analysis as described previously.24 Six trained assessors evaluated the effluent
197
of each of the six sample types individually; giving a total of 36 sniffs. Odor intensity data
198
were acquired at 1 second intervals by a computer mouse and a 10-point scale using the TI
199
function in the software package SensoMaker® (Version 1.7).25 TI responses had both
200
maximum intensity (height) and duration (width). Integrated area under the curve (AUC)
201
data for each defined odor peak was calculated. The statistical average of AUC values for
202
each odor peak was estimated to obtain an average representative aromagram.
203
Derivatization of Free Amino Acids and Analysis by GC-MS
204
Quantification of free amino acids (FAAs) (except arginine), carnosine and other non-
205
volatile compounds (organic acids and fatty acids) was achieved by methyl chloroformate
206
derivatization and subsequent GC-MS analysis according to published protocols.26, 27 Raw
207
and corresponding grilled low and high IMF samples from each production type were
208
prepared. The purpose of analyzing raw and grilled samples was to measure potential
209
changes in non-volatiles within the surface layer of the meat which may affect the flavor
210
intensity. Slices (~ 4 cm wide) were excised from the middle of two separate steaks and the
211
top surface (~5 mm depth) were reduced into small pieces. A total of 3 animals x 2
212
replicates x 2 marbling levels (low and high) = 12 samples were prepared for each
213
production type (n=36 raw, n=36 grilled). The small pieces of raw or freshly grilled meat (2
214
g) were immediately suspended in ice-cold methanol solution (70%), homogenized,
215
centrifuged and the supernatant was filtered before derivatization. Relative response factors
216
were determined for quantitative ions (m/z) for each analyte and concentrations of FAAs
10 ACS Paragon Plus Environment
Page 11 of 39
Journal of Agricultural and Food Chemistry
Intramuscular fat and beef flavor 217
(mg/100g) were estimated against the internal standard norvaline (100 µg/mL, m/z 130).
218
Lactic acid was quantified against L-lactate-3-13C internal standard isotopomer (1000
219
µg/mL, m/z 46). Chloroformate derivatives (1 µL) were injected at 250 oC (splitless) into the
220
GC-MS (QP-2010-Plus, Shimadzu) and separated on a Sol-Gel Wax column (SGE,
221
Australia, 30 m, 0.25 id, 0.25 µm film) using temperature programming; initial temperature
222
45 oC (held 2 minutes) and then heated at 9 °C/min to 180 °C (held 5min), 40°C/min to
223
220°C (held 5 min). Reference compounds were used to confirm compounds, which were
224
quantified using characteristic ion fragments (m/z). Data for raw and grilled samples from
225
each production type and corresponding IMF data were used in the statistical analysis.
226 227
Statistical Analyses
228
Statistical analyses were performed using GenStat® 16th Edition (VSN International Ltd,
229
Hemel Hempstead, United Kingdom). For statistical purposes, samples were either
230
classified as three “production types” (AngusGrass, AngusGrain or WagyuGrass) or as nine
231
distinct “sample types”, e.g. AngusGrass low marbling (AGL), AngusGrass medium
232
marbling (AGM), AngusGrass high marbling (AGH), and similarly, AngusGrain (AGRNL,
233
AGRNM, AGRNH) and WagyuGrass (WGL, WGM, WGH). Sensory differences were
234
assessed by multivariate analysis of variance (MANOVA) comparing the nine distinct
235
samples, using ‘sample type’ and ‘panelist’ as a fixed effects; different marbling levels were
236
not taken into consideration. A separate MANOVA analysis was conducted using the three
237
‘production types’ and the IMF (for each individual sample) as a covariate term, to correct
238
for differences in marbling level. To ascertain feed effects, AngusGrass and AngusGrain
239
production types were compared by MANOVA using ‘feed’ and ‘panelist’ as fixed factors
240
and IMF as a covariate. Similarly, for breed effects, AngusGrass was compared to
241
AngusGrain using ‘breed’ and ‘panelist’ as fixed factors and IMF as a covariate. Similar
11 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 12 of 39
Intramuscular fat and beef flavor 242
MANOVA comparisons were made for various replicate chemical data using fixed effects of
243
production type, sample type, feed and breed and the covariate IMF. Mean sensory and
244
chemical data values were used to determine Pearson’s correlation coefficients and
245
subjected to a two-sided test for significance.
246
RESULTS AND DISCUSSION
247
Carcass Characteristics
248
All carcasses were eligible for MSA grading according to specified criteria12,
249
covered a typical range of values (Table 1). In accordance with the experimental design,
250
differences in MSA-MB and IMF were measured between the nominal marbling bands.
251
AngusGrass represented the lower end, AngusGrain the middle, whereas WagyuGrass
252
samples were at the high end of the marbling range. Hot carcass weight (HCWT) and eye
253
muscle area (EMA) were significantly different between production types. The Wagyu breed
254
was generally lower in carcass weight than Angus. EMA generally increased with marbling
255
level. AngusGrain carcasses were heaviest in carcass weight, consistent with being from
256
animals on a higher energy diet during the finishing phase. Dentition and ossification scores
257
indicated that WagyuGrass heifers were older than the AngusGrain steers and AngusGrass
258
yearlings. Other factors being equal, marbling typically increases with maturity; age
259
differences are inevitable across such a broad marbling range and are typical for beef
260
production in Australia.23, 29, 30 The upH was < 5.7 for all carcasses, eliminating high pH dark
261
cutting meat as a potential negative factor affecting meat sensory quality.31,
262
(a* and b* values) indicated typical values and did not vary significantly according to sample
263
type, except for (L*), which was higher in the AngusGrain especially compared to the
264
AngusGrass.15
265
TBARS values were higher in WagyuGrass (p < 0.001); for the WagyGrass, TBARS was
266
positively correlated with IMF (r = 0.62, P < 0.007) (Table 1). There were no feed-related
32
13, 28
and
Meat color
12 ACS Paragon Plus Environment
Page 13 of 39
Journal of Agricultural and Food Chemistry
Intramuscular fat and beef flavor 267
difference in TBARS between the AngusGrain and AngusGrass and a correlation between
268
TBARS and IMF was not found in these samples. The higher TBARS for the WagyuGrass
269
but not the AngusGrass compared to the AngusGrain, may have been due to differences in
270
overall antioxidant status in these samples, e.g. selenium and vitamin E (not measured).33
271
Although no differences were observed in upH, there were significant differences in the
272
muscle glycogen stores at 24-hrs post-mortem. This is expected as usually residual
273
glycogen remains in the muscle after glycolysis ceases post-mortem.34, 35 Average residual
274
glycogen content for AngusGrass, AngusGrain and WagyuGrass was 16.5 mg/g, 11.9 mg/g
275
and 14.42 mg/g, respectively (p = 0.033).
276
Warner-Bratzler Shear Force, Soluble Collagen and Insoluble Collagen
277
As expected, WBS was negatively correlated with MSA-MB (r = -0.66, p < 0.001) and IMF (r
278
= -0.53, p < 0.001). WagyuGrass had lower WBS values compared to AngusGrass (p <
279
0.001, breed effect), when IMF was used as a covariate (Table 1). No feed related
280
differences were measured. Most samples could be classified as either ‘tender’ (31.4 N <
281
PF < 38.3 N) or ‘very tender’ (PF < 31.4) (Table 1).19 The AGL samples were classified as
282
“intermediate tenderness” (38.3 N < PF < 45.1 N). The amount of collagen and connective
283
tissue is known to affect beef palatability.36 Total collagen and heat soluble collagen did not
284
differ significantly between sample types; there were no breed or feed effects (Table 1). A
285
negative correlation between IMF and total collagen was measured for AngusGrain (r = -
286
0.55, p = 0.04) and AngusGrass samples (r = -0.48, p = 0.05). WBS cook loss differed
287
between sample types and was inversely correlated with IMF (Table 2), with significantly
288
greater losses from lower IMF meat (r = -0.56, p < 0.001).
289
Relationship between Intramuscular Fat and MSA-MB
290
As expected, IMF was strongly and positively correlated with MSA-MB, however the
291
relationship varied slightly depending on production type. Correlations and linear equations 13 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 14 of 39
Intramuscular fat and beef flavor 292
for the relationship between IMF(%) and MSA-MB were: all samples together (n=42, r =
293
0.79, p < 0.001, IMF(%) = 0.0148*(MSA-MB) + 0.4977), for AngusGrain (n=14, r = 0.89, p
294
