Subscriber access provided by the University of Exeter
Article
Evidence for the initial steps of DHA biohydrogenation by mixed ruminal microorganisms from sheep involves formation of conjugated fatty acids Noelia Aldai, Pierluigi Delmonte, Susana Alves, Rui J.B. Bessa, and John Kramer J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.7b04563 • Publication Date (Web): 01 Jan 2018 Downloaded from http://pubs.acs.org on January 1, 2018
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 46
Journal of Agricultural and Food Chemistry
Evidence for the initial steps of DHA biohydrogenation by mixed ruminal microorganisms from sheep involves formation of conjugated fatty acids
Noelia Aldai1*, Pierluigi Delmonte2, Susana P. Alves3, Rui J. B. Bessa3 and John K.G. Kramer4
1
Department of Pharmacy and Food Sciences, University of the Basque Country (UPV/EHU),
Paseo de la Universidad 7, 01006 Vitoria-Gasteiz, Spain; 2Office of Regulatory Science, Centre for Food Safety and Applied Nutrition, Food and Drug Administration, College Park, MD, USA; 3
CIISA, Faculty of Veterinary Medicine, University of Lisbon, Av. da Universidade Técnica, 1300-
477 Lisbon, Portugal; 4Guelph Food Research Centre, Agriculture and Agri-Food Canada, Guelph N1G 5C9, Ontario, Canada (retired).
*Corresponding author (Tel: (+34) 945 014501; Fax: (+34) 945 013014; E-mail:
[email protected],
[email protected])
1 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
1
ABSTRACT
2
Incubation of DHA with sheep rumen fluid resulted in 80% disappearance in 6h. The products were
3
analyzed as their fatty acid (FA) methyl esters by GC-FID on SP-2560 and SLB-IL111 columns.
4
The GC-online reduction×GC and GC-MS techniques demonstrated that all DHA metabolites
5
retained the C22 structure (no evidence of chain-shortening). Two new transient DHA products
6
were identified; mono-trans methylene interrupted-DHA and mono-conjugated DHA (MC-DHA)
7
isomers. Identification of MC-DHA was confirmed by their predicted elution using equivalent chain
8
length differences from C18 FA, their molecular ions, and the 22:5 products formed which were the
9
most abundant at 6h. The 22:5 structures were established by fragmentation of their 4,4-
10
dimethyloxazoline derivatives, and all 22:5 products contained an isolated double bond, suggesting
11
formation via MC-DHA. The most abundant c4,c7,c10,t14,c19-22:5 appeared to be formed by
12
unknown isomerases. Results suggest that the initial biohydrogenation of DHA was analogous to
13
that of C18 FA.
14 15 16
KEYWORDS
17
Rumen biohydrogenation, biohydrogenation intermediates, docosahexaenoic acid, in vitro.
18 19
2 ACS Paragon Plus Environment
Page 2 of 46
Page 3 of 46
Journal of Agricultural and Food Chemistry
20
Introduction
21
Unsaturated fatty acids (FAs) are extensively reduced to more saturated FAs in the rumen. The
22
rumen biohydrogenation (RBH) process of the C18 unsaturated FAs (oleic [18:1n-9], linoleic
23
[18:2n-6] and linolenic [18:3n-3] acids) and the intermediates formed have been extensively
24
reviewed 1-3. The major RBH pathway of 18:2n-6 and 18:3n-3 involves a concerted enzymatic cis-
25
trans isomerizations to conjugated FAs (CFA) that are subsequently reduced via proton
26
abstractions, double bond migrations, and exchanges with water to form trans containing
27
intermediates
28
recently shown that the mixture of intermediates and metabolites formed from 18:3n-6 6 and 18:3n-
29
3 7 may be even more complex than first reported.
30
The long-chain polyunsaturated FAs (PUFAs) like eicosapentaenoic (EPA; 20:5n-3) and
31
docoxahexaenoic (DHA; 22:6n-3) acids are also known to undergo extensive RBH as evidenced by
32
the low transfer efficiency of these PUFAs into ruminant-derived products 8. However, to date only
33
few studies have reported the products formed from the BH of DHA and evaluated the mechanism.
34
Kairenius et al. 9 examined the products of DHA and other PUFAs in the omasal digesta from cows
35
fed fish oil. They resolved many products by gas chromatography (GC) and characterized one 22:5
36
product (∆5,10,13,16,19-22:5). Based on this structure and the lack of finding conjugated
37
intermediates, they proposed a different pathway for the BH of DHA compared to that of C18
38
PUFAs, suggesting a reduction of the cis double bond closest to the carboxyl group. Escobar et al.
39
10
40
intermediates of DHA. Therefore they assumed that the previously described pathway 9 was correct
41
while they did not identify additional DHA metabolites. Jeyanathan et al.
42
two well-known BH bacteria Butyrivibrio fibrisolvens D1 and Butyrivibrio proteoclasticus P18 for
43
up to 48 h. They resolved a number of DHA products by GC and identified one major
44
(∆5,10,13,16,19-22:5) and two minor 22:5 products (∆4,10,13,16,19-22:5; ∆7,10,13,16,19-22:5).
45
Since they were not able to find conjugated intermediates of DHA, they supported the pathway
4,5
. With the use of isotope labeling technique and mass spectrometry (MS) it was
incubated DHA with rumen fluid from adult sheep and likewise could not find conjugated
3 ACS Paragon Plus Environment
11
incubated DHA with
Journal of Agricultural and Food Chemistry
46
reported by Kairenius et al. 9. Even though Klein and Jenkins
12
47
DHA metabolites, they found that 13C enriched DHA was only metabolized to C22 FAs, and none
48
of the C18 FAs contained the label. This put to rest the earlier speculation that DHA could be chain-
49
shortened to trans-18:1 isomers that increased when DHA was included in the diet 13 or incubation
50
14
51
It is not the purpose of this study to investigate the effect of adding DHA on the metabolism of
52
other PUFA or conditions in the rumen
53
metabolism of DHA. Furthermore, it is not clear whether DHA by itself or the trans containing
54
products formed during the BH of DHA, identified in this study, might affect the metabolism of
55
other PUFA. For this reason it is important to actually characterize the products of DHA formed
56
during the BH of DHA.
57
In this manuscript we investigated the products formed during the initial stages of DHA BH using
58
mixed ruminal microorganisms from sheep at conditions which were previously found to give
59
appreciate levels of DHA intermediates 15. A combination of methods was used to help identify the
60
products, including different highly polar GC columns, a newly developed GC- online reduction
61
×GC (GC-ORxGC) method
62
determine whether the RBH of DHA involved the formation of CFA as intermediates as in the case
63
of the C18 PUFA, and to confirm that the metabolism of DHA does not involve chain-shortening.
64
An identification of the intermediates formed during the incubation of DHA should provide a better
65
understanding of how DHA is metabolized, since not all metabolites should be considered
66
beneficial to health 17.
67
MATERIALS AND METHODS
68
Solutions
69
Phosphate-bicarbonate buffer solution (PBBS) was prepared as described by Goering & Van Soest
70
18
71
10 µg per µL of unesterified DHA (U-84-A from Nu-Chek Prep Inc., Elysian, MN) dissolved in
did not characterize additional
.
1-3
, nor whether BH of other PUFAs will affect the
16
, GC-MS and synthetic preparations. It was of particular interest to
with the exception that no trypticase was added. The concentration of the DHA solution was of
4 ACS Paragon Plus Environment
Page 4 of 46
Page 5 of 46
Journal of Agricultural and Food Chemistry
72
ethanol. Methyl tricosanoate (23:0) obtained from Nu-Chek Prep Inc. was used as an internal
73
standard and the concentration was 1 µg per µL in toluene.
74
In vitro incubations
75
Three mature sheep (3 males of Merino Branco breed) averaging 57±4.6 kg body weight were fitted
76
with ruminal cannula and served as donors of rumen contents. Animals were handled and cared for
77
in accordance with the EU Directive 86/609/EEC (Protocol # FMV/CEBEA 007/2016). Sheep were
78
fed a maintenance diet consisting of 0.5 kg of hay and 0.5 kg of concentrate (50% barley, 20%
79
sunflower meal, 17% ground corn, 10% soya meal, 1% calcium carbonate, 1% salt, 0.5% sodium
80
bicarbonate, 0.5% vitatec or vitamin premix) per day for over 2 weeks in morning and afternoon
81
portions. The chemical composition (g per kg of dry matter) of the concentrate and hay are 264 and
82
65 for crude protein, 14 and 9.0 for total fat, 241 and 622 for neutral detergent fiber, 123 and 475
83
for acid detergent fiber, and 79 and 91 for ashes, respectively.
84
Between 800-1000 mL of rumen content were collected from each sheep before morning feeding,
85
strained through four layers of gauze and transferred to the laboratory in three (one per animal) pre-
86
warmed thermo flasks. Then, 300 mL of strained rumen fluid of each thermo were pooled together
87
and kept at 39 ºC with continuous CO2 flushing. Fifty mL of the pooled and strained rumen fluid
88
were mixed with 100 mL of PBBS and used in further incubations.
89
Batch-cultures were performed using 15 mL Hungate tubes. The incubated substrate was composed
90
by a proportional mixture of glucose, cellobiose, maltose, soluble starch, and casein, similar to other
91
reports 19,20, and each tube contained 25 mg of the substrate (5 mg of each ingredient). Twenty µL
92
of the DHA solution were added to each tube which gave a final DHA content of 200 µg per tube
93
which was within the appropriate range to observe the formation of DHA metabolites within the
94
first 6 h of incubation as previously reported 15. In order to have the same amount of ethanol in all
95
tubes, 20 µL of ethanol were added to the control tubes containing no DHA. As reported by
96
Morgavi et al. 21, low volumes of ethanol did not affect fermentation, and was not considered toxic
97
to the rumen microorganisms. 5 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
98
For incubations, volumes of 6 mL were used that were flushed with CO2, capped with a rubber
99
stopper and mixed. Three tubes per time period were run under anaerobic conditions for 0, 1, 2, 3,
100
and 6 h in a water bath set at 39 ºC, and tubes were individually agitated every hour. After each
101
incubation time, the reaction was stopped by placing the tubes into boiling water for 5 min and
102
immediately stored at -80 ºC. An exception was made for tubes at 0 h in which the appropriate
103
amount of DHA was added just when the tubes were placed into the boiling water in order to avoid
104
any possible reaction. Additionally, another set of tubes with no DHA were also frozen at 0 h
105
(control) to determine the natural (baseline) content of DHA and other FAs in the samples. Two
106
repetitions of the aforementioned design were run on separate days.
107
Transesterification of rumen samples
108
Samples were freeze-dried, and transesterified to FA methyl esters (FAME) in the same tube using
109
a combined basic followed by acid catalysis adapted from Alves et al. 22. Briefly, 200 µL of internal
110
standard solution (1 µg/µL of 23:0; methyl tricosanoate) was added to each tube containing the
111
rumen fluid. Then 1.5 mL of methanol was added to each tube and sonicated for 10 min to rupture
112
cells. After vortexing, tubes were placed in the sonicator for another 5 min and then 1 mL of 0.5 M
113
sodium methoxide in methanol was added, vortexed and tubes were placed in a water bath at 50 ºC
114
for 30 min. After cooling, 1.5 mL of 10 % HCl in methanol (v/v) was added, vortexed and heated at
115
80 ºC for 15 min. After cooling, 1 mL of an aqueous solution of potassium carbonate (6%, w/w)
116
was added and the methylated lipids were extracted by adding 2 mL of hexane. After mixing, the
117
tubes were centrifuged at 500 g for 5 min (20 ºC). The upper hexane layer was transferred to a clean
118
test tube containing approximately 1 g of anhydrous sodium sulphate. The extraction step was
119
repeated with another 2 mL of hexane. The combined hexane layers were evaporated under a stream
120
of N2 at 37 ºC, and the FAMEs reconstituted in about 200 µL of dichloromethane for subsequent
121
purification from other constituents. Purification of the methylated lipids was performed using thin
122
layer chromatography (TLC) and dichloromethane as developing solvent in order to resolve the
123
FAMEs, dimethylacetals, and the FA dimethylesters and oxo-FAMEs 22. Each isolated fraction was 6 ACS Paragon Plus Environment
Page 6 of 46
Page 7 of 46
Journal of Agricultural and Food Chemistry
124
then stored at -20 ºC until further analysis.
125
Reduction of DHA with hydrazine
126
The FAME of DHA was partially reduced with hydrazine to produce all-cis 22:5, 22:4, 22:3, 22:2
127
and 22:1 FAMEs as described by Delmonte et al. 23.
128
Preparation of 4,4-dimethyloxazoline (DMOX) derivatives
129
Approximately 1-5 mg of the BH products as FAME were hydrolyzed to free FAs (FFA) by
130
reaction with 1 mL of 0.1 M KOH in 90% ethanol, at 50°C for 2 hours. The solution was acidified
131
with 0.5 mL of 1 M HCl. After addition of 5 mL of water and 3 mL of hexane:diethylether (1:1),
132
the FFA were recovered. The FFAs were reduced to dryness at the bottom of a 1 mL conical screw
133
cap reaction vial, and then 0.5 mL of 2-amino-2-methyl-1-propanol was added. The container was
134
purged with argon and heated for 2 h at 190 ºC 24. The reaction products were transferred into a 20
135
mL test tube containing 5 mL of hexane:diethylether (1:1) and 5 mL of water. The test tube was
136
gently mixed and let rest until the layers separated. The organic layer was washed with another 5
137
mL of water, passed through 500 mg of sodium sulfate, and concentrated to 50 µL with a stream of
138
nitrogen.
139
GC-FID analysis
140
The FAMEs were analyzed using a gas chromatograph (HP 6890A; Hewlett-Packard, Avondale,
141
PA) equipped with a fame-ionization detector (GC-FID) and a Supelco SP-2560 capillary column
142
(100 m x 0.25 mm i.d. x 0.20 µm film thickness; Bellefonte, PA). The injector and detector
143
temperatures were set at 250 ºC and 280 ºC, respectively. The FAME samples were analyzed by
144
applying a temperature program with a plateau at 175 ºC and at 150 ºC to enhance the resolution of
145
the trans-18:1 isomers as previously described 25. The equivalent chain lengths (ECL) values of
146
selected FAMEs from the incubation mixtures and bovine milk were calculated 26 from separations
147
acquired using an isothermal temperature at 200 ºC. Hydrogen was used as carrier gas at a flow rate
148
of 1 mL / min, and 1 µL of sample was injected with a split ratio of 50:1. For FAME identification,
149
reference standards #463 and #603, plus individual FAMEs of 21:0, 23:0, 26:0, 28:0 and CLA 7 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
150
mixture #UC-59M were used (Nu-Chek Prep Inc., Elysian, MN). Branched-chain FAs were
151
identified using a bacterial FAME mixture purchased from Matreya (Pleasant Gap, PA).
152
Selected samples were also analyzed using the Supelco SLB-IL111 ionic capillary column (200 m x
153
0.25 mm i.d. x 0.20 µm film thickness; Bellefonte, PA) 27,28. In this case, the injector and detector
154
temperatures were set at 300 ºC and 250 ºC, respectively. Initial oven temperature of 170 ºC was
155
held for 50 min, increased at 6 ºC / min to 185 ºC and held for 35 min. Hydrogen was used as
156
carrier gas at a flow rate of 1.6 mL / min for 35 min, then increased 0.3 mL / min / min to a final
157
flow rate of 3 mL / min. One µL of sample was injected and the split ratio was 100:1.
158
GC-online reduction × GC (GC-OR×GC) analysis
159
The GC-OR×GC separations were acquired by using the instrumental configuration and conditions
160
previously described with minor modifications 16. The apparatus consisted of an Agilent 7890A GC-
161
FID (Wilmington, DE) equipped with a dual stage thermal modulator Zoex ZX2 (Houston, TX),
162
and a split/splitless injection port. The column set was sequentially composed of a SLB-IL111
163
capillary column (100 m x 0.25 mm i.d. x 0.25 µm film thickness; Supelco, Inc., Bellefonte, PA), a
164
0.20 m x 0.18 mm i.d. capillary tube coated with palladium, a 2 m x 0.10 mm deactivated uncoated
165
capillary tube (modulator loop), and a SLB-IL111 capillary column (1 m x 0.10 mm i.d., 0.08 mm
166
film thickness; Supelco). The modulation spots were set at 0.05 m after the beginning of the
167
modulation loop and 0.05 m from the beginning of the second separation column. The modulation
168
time was set to 8 sec. The oven was maintained at the constant temperature of 170 ºC, and the
169
injector port at 300 ºC. The FID was maintained at 250 ºC and fed with 30 mL / min of N2 make up
170
gas, 400 mL / min air, and 30 mL / min of H2. Hydrogen was used as carrier gas at a constant
171
pressure of 60 psi. The split flow was set to 26.88 mL / min, and the injection volume to 1 µL. The
172
acquisition rate was set to 200 hz, and data were processed with the GC Image GC×GC software
173
(Version 2.1, GC Image, LLC, Lincoln, Nebraska). All capillary connections were made with Micro
174
Unions (SGE Analytical Science, Australia).
175
GC-MS separation of FAMEs in the CI+ mode 8 ACS Paragon Plus Environment
Page 8 of 46
Page 9 of 46
Journal of Agricultural and Food Chemistry
176
GC-MS separations in the chemical ionisation (CI+) mode were performed with an Agilent 7200A
177
gas chromatography quadrupole time-of-flight mass spectrometry (GC-QTOF-MS), equipped with
178
a SP-2560 capillary column (100 m x 0.25 mm, 0.20 µm film thickness) followed by a purged union
179
and a 2 m x 0.10 mm deactivated retention gap connecting to the MS. The carrier gas was He at the
180
constant flow of 1.1 mL / min increased to 1.2 mL / min after the purged union. The oven
181
temperature was maintained at 45 ºC for 4 min, ramped to 175 ºC at 13 ºC / min, maintained for 27
182
min, then ramped at 4 ºC / min to 225 ºC and maintained for 100 min. The injection volume was 1
183
µL in splitless mode, and the injection port was purged after 2 min with 70 mL / min of carrier gas.
184
The injection port and transfer line temperature was set to 250 ºC, and the ion source to 300 ºC. The
185
MS was operated in CI+ mode with NH3 as chemical ionization reagent. The filament current was
186
set to 100 µA.
187
GC-MS separation of DMOX derivative in the EI+ mode
188
Experimental conditions were the same as for the CI+ analyses of FAMEs with the following
189
exceptions. The oven temperature was maintained at 190 ºC for the entire separation (200 min), 1
190
µL was injected in split mode with a split ratio of 1:15. The MS was operated in EI+ mode with the
191
electron energy of 50 eV and the filament current of 100 µA.
192
Statistical analysis
193
Data were analyzed using the PROC MIXED from SAS 9.4 29 considering the incubation run as
194
random block and the incubation time as a fixed categorical effect. When significant effects
195
(P 0.05).
759 760 761
Figure 10. Proposed pathways of DHA metabolism using sheep rumen fluid. One of the five
762
possible MTMI-DHA and two of the MC-DHA metabolites are shown as representative products,
763
as well as the five 22:5 products identified, all of which contain an isolated trans double bond. (*)
764
Members of the pathway of DHA metabolism that resulted in the formation of the major 22:5
765
product in the incubation mixture. (**) Product reported by Jeyanathan et al.(11) that could not be
766
confirmed.
767
MTMI-DHA, mono trans methylene interrupted-DHA; MC-DHA, mono-conjugated DHA.
768
31 ACS Paragon Plus Environment
Journal of Agricultural and Food Chemistry
Page 32 of 46
Table 1. Products of DHA incubation with sheep mixed rumen fluid for 0, 1, 2, 3, and 6 hours.
µg / tube 22:6 n-3 (DHA) MC-DHA MTMI-DHA 22:x unidentified sum of DHA products DHA + DHA products 22:5 22:4 22:3 22:2 22:1 cis-11 and cis-13 22:0
0 190a 0.00c 0.91c 9.20d 10.1d 200a 0.88c 0.01d 0.89b 1.01 1.29 12.7bc
Incubation time (h) 1 2 b 146 102c 5.96b 9.70a b 8.53 11.46a c 31.9 54.9b c 46.4 76.1b 192ab 178b bc 1.27 1.93ab c 0.52 0.76c a 1.19 1.37a 1.10 0.92 1.11 1.34 12.5c 13.0ab
3 83.3d 9.91a 11.6a 69.6b 91.1b 174bc 2.37a 1.10b 1.23a 1.16 1.36 12.9abc
6 36.2e 7.61ab 9.49b 102a 119a 156c 2.22a 1.74a 1.40a 1.44 1.20 13.1a
SEM 3.88 1.07 2.16 6.75 7.07 6.93 0.36 0.13 0.30 0.18 0.07 0.62
P-value