Subscriber access provided by Service des bibliothèques | Université de Sherbrooke
Article
Hyperspectral deep UV autofluorescence of muscle fibers is affected by postmortem changes Caroline Chagnot, Annie Venien, Frédéric Jamme, Matthieu Refregiers, Mickaël Desvaux, and Thierry Astruc J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.5b00668 • Publication Date (Web): 27 Apr 2015 Downloaded from http://pubs.acs.org on May 3, 2015
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 30
Journal of Agricultural and Food Chemistry
Hyperspectral deep UV autofluorescence of muscle fibers is affected by postmortem changes
1 2 3 4 5 6 7 8
Caroline Chagnot1, 2, Annie Vénien1, Frédéric Jamme3, 4, Matthieu Réfrégiers4,
9
Mickaël Desvaux2 and Thierry Astruc1
10
1
11
France.
12
2
13
3
14
UR370 Qualité des Produits Animaux, INRA, F-63122 Saint-Genès Champanelle,
UR454 Microbiologie, INRA, F-63122 Saint-Genès Champanelle, France Synchrotron SOLEIL, BP48, L’Orme des Merisiers, F-91120 Gif-sur-Yvette, France
4
INRA, UAR1008 CEPIA, rue de la Géraudière, F-44316 Nantes, France
15 16 17 18
Corresponding author: Thierry Astuc
19
E-mail:
[email protected] 20
Tel. +33 4 73 62 41 56
21
Fax +33 4 73 62 42 68
22 23 24
Short title: Postmortem muscle fibers fluorescence spectroscopy
25 26
1 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 2 of 30
27
Abstract
28
After slaughter, muscle cells undergo biochemical and physicochemical changes that
29
may affect their autofluorescence characteristics. The autofluorescent response of
30
different rat Extensor Digitorum Longus (EDL) and soleus muscle fiber types was
31
investigated by deep UV synchrotron microspectroscopy immediately after animal
32
sacrifice and after 24 hours storage in a moist chamber at 20 °C. The glycogen
33
content decreased from 23 to 18 micromoles / g fresh muscle in 24 hours
34
postmortem. Following a 275 nm excitation wavelength, the spectral muscle fiber
35
autofluorescence response showed discrimination depending on postmortem time (t0
36
versus t24h) on both muscles at 346 nm and 302 nm, and to a lesser extent at 408 nm
37
and 325 nm. Taken individually, all fiber types were discriminated, but with variable
38
accuracy, type IIA showing better separation of t0 / t24h than other fiber types.
39
These results suggest the usefulness of the autofluorescent response of muscle cells
40
for rapidly meat aging characterization.
41 42 43
Key words: rat muscle fibers; postmortem; meat; UV spectroscopy; histology ; fiber
44
types
45 46 47 48 49 50 51
2 ACS Paragon Plus Environment
Page 3 of 30
Journal of Agricultural and Food Chemistry
52
Introduction
53
Meat comes from skeletal muscle, which generally contains different proportions of
54
the four pure myofiber types: Type I, Type IIA, Type IIX and Type IIB. These contain
55
respectively the myosin heavy chain isoforms (MyHC) I, IIa, IIx and IIb
56
fibers intermediate between two pure types and containing two different myosin
57
isoforms (I-IIA, IIA-IIX, IIX-IIB) are also found. The muscle fibers are characterized by
58
their metabolism (oxidative or glycolytic) and their contraction speed (slow or fast).
59
Type I are slow-twitch fibers and all Types II are fast-twitch fibers. Type I and IIA
60
fibers have an oxidative metabolism; IIX and IIB fibers have a glycolytic one 1, 2.
61
After slaughter of farm animals, muscle undergoes significant metabolic, physical,
62
structural and biochemical changes that determine the quality of meat and meat
63
products
64
cells will try to survive by degrading their glycogen stock through anaerobic
65
glycolysis. Lactate accumulation in cells is accompanied by a drop in pH. When the
66
energy reserves are depleted, muscle pH stabilizes at values generally between 5.7
67
and 6.2, termed the ultimate pH. The kinetics of change in pH is species-dependent,
68
and the time to reach the ultimate pH can range from 2 h in poultry to 24 h in cattle.
69
After this first phase of postmortem changes, additional biochemical changes and
70
significant ultrastructural alterations are observed, related to the improvement of
71
meat quality, especially meat tenderness
72
changes (also called meat maturation) is dependent on the species, and mostly on
73
the metabolic and contractile types of the muscles considered. Degradation of
74
myofibrillar structure is faster in white than in red fibers 5, 7.
75
Postmortem changes can be assessed by mechanical measurements and
76
biochemical and/or ultrastructural analysis, but these methods are usually destructive
1, 2
. Hybrid
3, 4
. After bleeding, the muscle is deprived of oxygen and nutrients. Muscle
5, 6
. The speed of these postmortem
3 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 4 of 30
77
and time-consuming. Some authors have exploited the variation in the physical
78
properties of muscles to characterize differences in composition by fluorescence
79
spectroscopy 8-10. This highly sensitive method detects fluorophores naturally present
80
in muscle whose properties are very sensitive to changes in their environment.
81
Collagen,
82
autofluorescent molecules in muscle 8. Several years ago, the autofluorescence of
83
tryptophan signal following meat sample excitation at 290 nm, allowed a correct
84
discrimination of non aged from aged meat samples
85
performed on a macroscopic scale and did not address the metabolic and contractile
86
properties of muscles. The analysis at the cell or even the ultrastructural scale is
87
more difficult because of the low availability fluorescence microscopes that allow an
88
excitation in the deep UV which is relevant for studying biomacromolecules that have
89
a strong absorption in this energy range. Indeed, although spectrometers can often go
90
down to 250 nm, microscopes are usually very limited in this range with either 350 nm
91
cut-off or access to only a limited number of excitation wavelengths. However, the
92
combination of a microscope with a UV fluorescence spectrometer coupled to
93
synchrotron radiation allows an excitation range of 200 to 600 nm that has been used
94
to characterize animal tissues at the ultrastructural level 12-15.
95
Thanks to the availability of these tools, it becomes theoretically possible to detect
96
small changes of the tissue and cells and therefore characterize the degree of
97
postmortem
98
autofluorescence changes of individual muscle fibers from rat skeletal muscles
99
stored under anoxia during 24 hours, regarding their metabolism and ultrastructure
100
tryptophan,
change
tyrosine
in
muscle
or
NADH
cells.
are
Our
among the
11
most
abundant
. However, the analysis was
goal
was
to
investigate
the
changes.
4 ACS Paragon Plus Environment
Page 5 of 30
Journal of Agricultural and Food Chemistry
101
Materials and Methods
102
Animals and samples
103
The rat model was chosen because it allows a better control of rearing and slaughter
104
conditions; it is easy to extract entire muscles, and obtain early samples in the
105
laboratory by controlling the best kinetic sampling and temperature conditions.
106
Twenty-four male Wistar rats aged 5 months and weighing about 500 g were bought
107
from Janvier (Saint-Berthevin, France) and housed at 22 °C under 12 h:12 h
108
dark:light cycles with water and food ad libitum at an INRA animal facility (Installation
109
Expérimentale de Nutrition, Unité de Nutrition Humaine, INRA Theix, agreement
110
No. C 63345.14). Rats were anaesthetized by isoflurane gas and decapitated with a
111
guillotine according to the recommendation of our Regional Ethics Committee
112
(C2E2A No. 2) and French national legislation (JO 87-848). As no act of pain was
113
performed, since the sampling took place after the death of animals, the experimental
114
project was outside the scope of the European Union Directive 2010/63 and therefore
115
did not require ethics review. However, the experimental project was followed by the
116
animal welfare structure integrated in the INRA animal facility installation according to
117
French law (Article JO R. 214-103).
118
Immediately after death, the lower limbs were severed and dissected under sterile
119
conditions. Extensor digitorum longus (EDL), tibialis anterior, gastrocnemius and
120
soleus muscles from the right and left hindlegs were extracted from tendon to tendon
121
avoiding any lesions.
122 123 124
5 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 6 of 30
125
Muscle postmortem kinetics
126
Muscles from three rats were immediately cryofixed post-dissection (10 minutes after
127
decapitation). For the other 21 rats, muscles were suspended at 20 °C in a sterile
128
moist chamber to prevent drying of the muscle and contamination from
129
microorganisms. Lead ballast was attached at the caudal tendon of each muscle
130
(1.5 g for soleus and EDL, 4.5 g for tibialis and gastrocnemius) to preserve the
131
anatomical muscle length (Figure 1).
132
After respectively 30 min, 1 h, 2 h, 4 h, 8 h, 12 h and 24 h postmortem, muscles of
133
three rats were removed from their moist chambers and prepared for biochemical
134
and imaging analysis.
135 136
pH measurements
137
As EDL and soleus muscles are too small, pH measurements were made on tibialis
138
anterior and gastrocnemius muscles localized around EDL and soleus muscle. pH
139
measurements were made according to Fernandez et al., (2002)
140
small quantity of muscle available. Briefly, muscles were weighed and ground in
141
5 mM idoacetate buffer (2.5 ml/g of muscle), and pH was measured in the muscle
142
suspension at room temperature.
16
, adapting to the
143 144
Glycogen determination
145
Glycogen content was determined on each right EDL and soleus rat muscle at each
146
time postmortem by enzymatic procedures according to Bergmeyer, (1974)17, with
147
slight modifications. About 0.15 g of muscle tissue was homogenized with 1.8 mL of
148
0.55 M perchloric acid. On two 0.5 ml aliquots, glycogen was hydrolysed in glucose
149
with amyloglycosidase. Glucose was converted into 6-phosphogluconolactone after
6 ACS Paragon Plus Environment
Page 7 of 30
Journal of Agricultural and Food Chemistry
150
hexokinase and G-6-P dehydrogenase action accompanied by NADH H+ production.
151
NADH, proportional to glucose content, was determined by spectrophotometry at 340
152
nm and compared with standards to calculate the glucose concentration. Results are
153
expressed in micromoles per gram of fresh tissue.
154 155
Histology
156
Cryofixation and histological sections: Parts of left EDL and soleus muscles at 0 (t0)
157
and 24 h (t24h) postmortem were positioned on a cork plate with embedding medium
158
(Tissue-Tek) and cryofixed by immersion at −160 °C in isopentane cooled with liquid
159
nitrogen (−196 °C). Serial cross-sections of entire muscles (10 µm thick) were cut
160
using a cryostat (Microm, HM 560) and collected on glass slides for histological
161
stains, and on quartz coverslips for DUV microspectroscopy fluorescence analyses.
162
Sections were stored at −20 °C under vacuum until use. Periodic acid Schiff (PAS)
163
staining was performed for glycogen characterization 18.
164 165
Muscle fiber type determination: Fiber types were identified by highlighting the
166
different myosin heavy-chain isoforms (MyHC) using specific mouse monoclonal
167
antibodies BA-D5, SC-71 and BFF3 (AGRO-BIO France). The different primary
168
MyHC antibodies were visualized by an Alexa Fluor 488 labelled goat anti-mouse
169
IgG secondary antibody (A 11001, Invitrogen). In addition, the ECM protein laminin,
170
surrounding the muscle fiber, was stained using an anti-laminin primary polyclonal
171
antibody (L9393 Sigma) and a Cyanine 3-labelled secondary antibody (111-165-008,
172
Jackson). The myofiber response to the different MyHC antibodies identified fiber
173
Types I, IIA, IIB and hybrid IIX-IIB. IIX fiber types corresponded to the remaining
7 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 8 of 30
174
unmarked cells. Controls were performed with no primary antibody to validate the
175
results.
176 177
Image acquisitions: Observations and image acquisitions were performed using a
178
light microscope (Olympus BX 61) coupled to a high resolution digital camera
179
(Olympus DP 71) and the Cell F software. PAS-stained sections were examined, and
180
images were acquired in bright field mode. Immunohistochemistry images were
181
acquired in fluorescence mode (Alexa Fluor 488: 495/519 nm, Cyanine 3: 550/570
182
nm).
183 184
Transmission electron microscopy: Analysis was conducted on EDL and soleus
185
muscles at 0 and 24 h postmortem from 6 rats (3 rats per postmortem time). Pieces
186
of EDL and soleus muscles of about 4 × 1 mm were fixed overnight at 4 °C by
187
immersion in 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer pH 7.2, taking
188
into account muscle fiber direction. Small cubes of about 1 mm3 were post-fixed in
189
1% osmium tetroxide in 0.1 M sodium cacodylate buffer for 1 h at room temperature.
190
These cubes were dehydrated through a graded ethanol series and embedded in
191
epoxy resin (TAAB, Eurobio France). Ultra-thin longitudinal sections (90 nm) (parallel
192
to the fiber direction) were stained with uranyl acetate and lead citrate, and examined
193
with a transmission electron microscope (Hitachi HM 7650) using a 80 kV
194
acceleration voltage at the CICS cellular imaging lab on the Clermont-Ferrand
195
University campus (France). Micrographs were acquired with a Hamamatsu AMT
196
digital camera system coupled with the microscope.
8 ACS Paragon Plus Environment
Page 9 of 30
Journal of Agricultural and Food Chemistry
197
Deep UV fluorescence microspectrocopy
198
Spotting of cell types on unstained muscle serial cross-section: From two rat soleus
199
and EDL muscles (0 and 24 h postmortem), an unstained serial section was
200
deposited
201
immunohistofluorescence on other serial sections, 20 Soleus muscle fibers (10 Type
202
I and 10 Type IIA) and 21 EDL muscle fibers (3 Type I, 2 Type IIA, 5 Type IIX, 8 Type
203
IIX-IIB and 3 Type IIB) were spotted for fluorescence spectra acquisition by deep UV
204
(DUV) microspectroscopy.
on
quartz
coverslips,
and
using
fiber
type
identification
by
205 206
Spectrum acquisition: Synchrotron DUV microspectroscopy was performed on the
207
DISCO beamline of the SOLEIL synchrotron radiation facility (Saint-Aubin, France) 19,
208
20
. DUV monochromatized light was used at 275 nm excitation wavelength to excite
209
tissue sections. The emission spectra were acquired from 290 to 540 nm. For each
210
excited pixel a fluorescence spectrum was recorded. On each spotted cell, 20
211
acquisitions were made in the intracellular space.
212 213
Spectrum processing and statistical analysis: Autofluorescence spectra were spike-
214
and noise-filtered using an in-house program written in MATLAB version 7.3 (The
215
MathWorks, Natick, MA). The Unscrambler software (v9.8, Camo Software AS,
216
Norway) was used to perform a baseline adjustment to zero, apply unit vector
217
normalization that normalizes sample-wise the spectral data to unit vectors, and
218
analyze the processed spectra by principal component analysis (PCA). After
219
analysis, the family label of each spectrum was revealed and the two first of the 10
220
components tested were plotted. The mean spectrum of each group was also plotted
221
for better referencing of the spectral discriminants. Score plots were used to show
9 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 10 of 30
222
similarity maps for comparison of spectra regardless of sample categories. Loading
223
plots derived from the first (PC1) and second principal components (PC2) were used
224
to reveal and identify characteristic fluorescence peaks.
225
Numerical variables of fluorescence intensity for wavelengths of interest identified on
226
the PCA loadings were expressed as mean ± standard error of the mean (s.e.m.).
227
Variance analysis and mean comparisons were performed using one-way variance
228
analysis and the Student-Newman-Keuls test under the XLSTATsoftware 2010
229
(Microsoft office, Redmond, USA).
10 ACS Paragon Plus Environment
Page 11 of 30
Journal of Agricultural and Food Chemistry
230
Results and Discussion
231
Postmortem metabolism time course
232
The level of EDL and soleus muscles glycogen was about 23 µmol / g fresh muscle
233
at 10 minutes postmortem, and decreased slightly with time for both muscles (Figure
234
2). The decrease in contrast of PAS staining with time postmortem reflects a
235
decrease in intracellular glycogen, in agreement with our biochemical data and the
236
literature
237
on the fiber type. Glycolytic fibers (IIX and IIB) of the EDL were systematically richer
238
in glycogen than oxidative fibers (I and IIA), fully consistent with the literature on the
239
subject
240
a heterogeneous distribution. In this muscle, the IIA fibers contained less glycogen
241
than Type I fibers. This result is surprising at first sight because generally IIA fast-
242
twitch fibers have more glycogen reserves than Type I slow-twitch fibers to cope with
243
high energy consumption during vigorous exercise. This finding is explained by a
244
specific feature of rodent muscles (rats and mice), whose Type IIA fibers are richer in
245
mitochondria, and thus more oxidative than their Type I fibers
246
logical for it to be less richly supplied with glycogen. The initial level of glycogen
247
(around 23 µmol / g fresh muscle) was lower than reported by others who found 30–
248
40 µmol / g of fresh muscle in the rat muscle
249
the delay of 10 minutes necessary for muscle dissection when part of the glycogen
250
was depleted; 60% of muscle glycogen is degraded within 5 minutes after the death
251
of the animal
252
and gastrocnemius muscles), slightly lower than usual in vivo pH (7.2), as a
3, 4
3, 4
. However, soon after animal sacrifice, PAS staining intensity depended
. The soleus, which contains only oxidative fibers (I and IIA), also showed
23
23, 24
21, 22
. It is therefore
. This difference is probably due to
. This hypothesis is in accordance with the initial pH (6.9 in Tibialis
ACS Paragon Plus Environment
11
Journal of Agricultural and Food Chemistry
Page 12 of 30
253
consequence of glycogen degradation associated with a release of protons in the
254
tissue.
255
The rat muscle glycogen content appeared to be considerably lower than seen in
256
farm animal species where values of glycogen are about 60–70 µmol / g in rabbits 25,
257
85–100 µmol / g in pigs
258
29, 30
26, 27
, 45–50 µmol / g in turkeys
28
, 80–100 µmol / g in cattle
and about 70 µmol / g in sheep 31. Glycogen reserves remained relatively high in
259
rat muscle (Figure 2), unlike in farm animals whose meat glycogen reserves fall to
260
less than 10 µmol / g after rigor mortis, with a pH stabilization around 5.6–6.0
261
depending on the type of muscle 3, 4.
262
The initial pH of the gastrocnemius and tibialis muscle decreased and has reached
263
its minimum (6.06 ± 0.12 and 6.15 ± 0.03 respectively) after 2.5 hours postmortem,
264
which is consistent with the postmortem time of the stopped of glycogen consumption
265
in soleus and EDL muscles (Figure 2).
266 267
Postmortem ultrastructural changes
268
Representative images of ultrastructural analysis are shown in Figure 3. Shortly after
269
animal sacrifice, the myofibrils were well-preserved with well-defined sarcomeres and
270
aligned Z lines. However, the soleus showed contracted myofibrils that prevented the
271
I bands being distinguished. This contraction may be related to mechanical
272
stimulation during sampling, but also to cold shortening despite our precautions to
273
avoid this problem (fixative immersion at room temperature). The fact that only the
274
soleus muscle showed this state leads us to believe that it was due to cold
275
shortening as only red muscles are sensitive to cold shortening 3.
276
After 24 h postmortem, the myofibrils showed no significant change in their structure.
277
Soleus myofibrils were still contracted, but Z lines remained aligned, and
ACS Paragon Plus Environment
12
Page 13 of 30
Journal of Agricultural and Food Chemistry
278
mitochondria were well-preserved despite the storage temperature of 20 °C, which
279
promotes the action of endogenous proteases.
280
Appreciable changes were observed at 48 h postmortem on muscle sample prepared
281
exactly in the same conditions (see supplementary Figure S1).
282 283
Fluorescence response of muscle fibers according to time postmortem
284
Figure 4 shows the profile of fluorescence spectra acquired in soleus muscle fibers
285
at death (t0) and at 24 hours postmortem (t24h) (spectra acquired on the EDL muscle
286
show the same general pattern). The spectra show a peak at 332 nm characteristic
287
of tryptophan and a shoulder at 302 nm assigned to tyrosine
288
that emits fluorescence at 408 nm could not be identified with certainty. Collagen and
289
elastin fluoresce in this wavelengths range
290
extracellular matrix, which excludes the involvement of these compounds in the
291
observed fluorescence, as our spectral acquisitions were strictly intracellular. The
292
fluorescence peak at 408 nm could be attributed to NADH, since in living cells this
293
compound is suspected to fluoresce in this wavelength range 20.
294
The superposition of the spectra acquired at t0 and t24h shows some shifts which are
295
however difficult to see with the naked eye. However, these differences are
296
highlighted by the PCA and variance analysis that are presented in Figures 5–7.
297
PCA and loadings indicate a separation of the two postmortem times (t0 and t24h)
298
essentially in the PC2 on the fluorescence intensities at 302 and 346 nm, and to a
299
lesser extent on the PC1 at 325 nm and 408 nm, whichever the muscle considered
300
(Figures 5 and 6). In general, 302 nm and 408 nm fluorescence intensities were
301
higher, and 346 nm and 325 nm fluorescence intensities lower for t0 than for t24h
302
samples. Separation on the EDL PCA was less clearcut than in the soleus, mainly
14
12, 20, 32
. The compound
, but these proteins are located in the
ACS Paragon Plus Environment
13
Journal of Agricultural and Food Chemistry
Page 14 of 30
303
because the EDL is composed of four types of fiber plus IIX-IIB hybrid fibers, while
304
the soleus is composed of only I and IIA fibers.
305
Some cell types allow a better separation of postmortem times, such as Type IIA
306
fibers, which show excellent discrimination of postmortem times on both the EDL and
307
soleus muscles. IIX-IIB hybrid fibers and IIX fibers (absent in the soleus muscle) also
308
show a good separation as regards postmortem time. Postmortem time of soleus
309
muscle Type I fiber discrimination was somewhat less marked than for Type IIA, but
310
still very sharp. By contrast, postmortem time discrimination of EDL Type I and Type
311
IIB fibers was not as sharp on PC2. For these fiber types, the postmortem time was
312
discriminating on both PC2 at 302 nm and 346 nm and PC1 at 408 nm and 325 nm.
313
Discriminatory wavelengths of PC1 and PC2 were exactly the same for the two
314
muscles (EDL and soleus) and all types of fibers: 325 nm and 408 nm for PC1, 302
315
and 346 for PC2. Hence it seems that only few fluorophores are concerned in
316
postmortem time discrimination.
317
Significant differences in emission fluorescence intensities regarding the postmortem
318
time were amplified by the 346/302 and 408/325 ratios, which were systematically
319
lower and higher respectively for t0 than t24h muscle fibers (Figure 7). Soleus t0 fibers
320
I and IIA have a lower 408/325 ratio than EDL t0 I and IIA fibers (Figure 7). The
321
fluorescence intensity at 408 nm is significantly higher (p