1
2
Interaction between fish oil and plant oils or starchy concentrates in the diet:
3
Effects on dairy performance and milk fatty acid composition in goats
4
P.G. Torala,b, J. Rouela,b, L. Bernarda,b,*, Y. Chilliarda,b,*
5
6
7
8
a
INRA, UMR1213 Herbivores, Site de Theix, F-63122 Saint-Genès-Champanelle, France
b
Clermont Université, VetAgro Sup, BP 10448, F-63000 Clermont-Ferrand, France
9
10
11
12
13
14
15
16
17
18
* Corresponding authors.
19
INRA, UMR1213 Herbivores, Site de Theix, F-63122 Saint-Genès-Champanelle, France. Tel:
20
+33 473 62 41 14 and 40 51; fax: +33 473 62 45 19.
21
E-mail addresses: yves.chilliard@clermont.inra.fr and laurence.bernard@clermont.inra.fr
22
23
24
Abbreviations: CLA, conjugated linoleic acid; FA, fatty acid; FAME, fatty acid methyl ester;
25
PC, principal component; PCA, principal component analysis.
1
26
Abstract
27
Two experiments were performed to investigate the effects of supplementing a diet
28
with fish oil either alone or with plant oils on dairy goat performance and milk fatty acid (FA)
29
profile (Experiment 1), and the interaction between fish oil with the type and level of starch
30
concentrate (Experiment 2). In Experiment 1, 84 goats were allocated to 7 experimental diets
31
without a lipid supplement or with a low or high dose of fish oil (20 or 40 g/d, respectively)
32
either alone or with linseed or sunflower-seed oils (the combinations included 130 g/d of oil
33
supplements). In Experiment 2, 72 goats were allocated to 6 experimental diets without a lipid
34
supplement and with a concentrate rich in corn and barley grain starch or with the high dose
35
of fish oil (40 g/d) and concentrates rich in starch from barley grain, corn grain or both; or
36
that were low in starch from barley grain or corn grain. In contrast to cows, in goats, fish oil
37
supplements modulated milk FA composition without decreasing the milk fat content; this
38
result may have been related to specific milk FA responses, such as a lack of or a small
39
reduction in 18:0 and c9-18:1 (Experiment 1) or a moderate reduction compensated through
40
increases in short-chain FA (Experiment 2) and limited increases in t10-18:1. The
41
combinations of fish oil with sunflower-seed oil were more efficient than either fish oil plus
42
linseed oil or fish oil alone at increasing (P<0.05) milk c9,t11-18:2 and t11-18:1
43
concentrations (up to 24- and 35-fold increases, respectively), simultaneously decreasing
44
(P<0.05) medium-chain saturated FA (on average, 40% decrease). Based on the milk FA
45
changes (e.g., 18:2n-6, 18:3n-3, t11-18:1 and 18:0) in Experiment 2, diets rich in barley grain
46
starch with fish oil would induce less extensive ruminal dietary FA biohydrogenation and
47
better inhibit the trans 18:1 reduction than diets that are supplemented with the same level of
48
fish oil but are low in barley grain starch or rich in corn grain starch. The apparent transfer
49
rates of 20:5n-3 and 22:6n-3 from fish oil to milk were low (on average, 2.8 and 2.4%,
50
respectively; Experiment 1) but were slightly higher in barley-grain treatments (on average,
2
51
5.1 and 7.6%, respectively) compared with corn-grain-only treatments (on average, 2.5 and
52
3.7%, respectively; Experiment 2). The milk t10-18:1 concentration remained low (≤1.1%
53
total FA) and the t10-18:1/t11-18:1 ratio was much lower than in cows fed fish oil with plant
54
oils.
55
56
Key words: goat, fish oil, linseed oil, sunflower-seed oil, milk fatty acid, starch concentrate
3
57
1. Introduction
58
The potential benefits to human health have generated great interest in developing
59
nutritional strategies that enhance the concentration of bioactive fatty acids (FA) in ruminant
60
milk (Lock and Bauman, 2004; Chilliard et al., 2007). Diet supplementation with plant oils
61
has been reported to be a good strategy for increasing milk c9,t11-conjugated linoleic acid
62
(CLA) levels in goats (Mele et al., 2008; Bernard et al., 2009; Martínez Marín et al., 2011).
63
Additionally, multiple studies have attempted to increase the concentration of 20:5n-3 and
64
22:6n-3 in ruminant milk by adding fish oil to the diet, but the apparent transfer rate of these
65
FA from diet to milk is relatively low (Kitessa et al., 2001; Loor et al., 2005a; Toral et al.,
66
2010a). However, as a rumen biohydrogenation modulator, fish oil yields large increases in
67
milk c9,t11-CLA and t11-18:1 concentrations, particularly when combined with plant oils
68
either in goats, cows, or sheep (Gagliostro et al., 2006; Shingfield et al., 2006; Toral et al.,
69
2010a).
70
By contrast, the effect of fish oil supplementation on dairy performance largely
71
depends on the ruminant species. In cows and ewes fed fish oil, lower milk fat content and
72
yield are frequently observed (Chilliard et al., 2001; Loor et al., 2005a; Shingfield et al.,
73
2006; Toral et al., 2010a). Goats are less prone to milk fat depression (Chilliard et al., 2003;
74
Shingfield et al., 2010), but the data on fish oil supplementation in this species are too limited
75
to offer a full picture (Kitessa et al., 2001; Gagliostro et al., 2006; Sanz Sampelayo et al.,
76
2007; Bernard et al., 2010).
77
Furthermore, the response to lipid supplements is strongly influenced by the basal diet
78
composition, and in cows and goats fed oils, enhanced dietary starch levels shift the ruminal
79
biohydrogenation toward the t10-pathway at the expense of the t11-pathway (Shingfield et al.,
80
2005; Mele et al., 2008; Bernard et al., 2009), which could detrimentally affect animal
81
performance and the nutritional quality of ruminant-derived products (Roy et al., 2007;
4
82
Shingfield et al., 2008). However, despite the volume of research on the effects of dietary
83
concentrate levels on the milk FA profile (Chilliard et al., 2007), surprisingly few data
84
describe the extent to which starch concentrates with different degradability affect milk FA
85
composition (Jurjanz et al., 2004; Cabrita et al., 2009; Bernard et al., 2012). Interestingly,
86
replacing corn grain with wheat grain as a dietary starch source influenced the goat response
87
to sunflower-seed oil through decreasing the level of c9,t11-CLA and t11-18:1 as well as
88
increasing the t10-18:1 and de novo synthesized FA in milk (Bernard et al., 2012). However,
89
the effect from the interaction between fish oil and dietary starch sources on milk FA
90
composition remains unknown.
91
Thus, this study included two experiments to investigate the effects of supplementing
92
a diet with two doses of fish oil either alone or with plant oils (linseed or sunflower-seed oil)
93
on dairy goat performance and milk FA profile (Experiment 1) as well as the interaction
94
between fish oil and the type and level of starch concentrate (corn and barley grains;
95
Experiment 2), with the objective of establishing the framework necessary to better control
96
the FA profile and the levels of bioactive FA in goat milk without impairing animal
97
performance, in particular milk fat content and yield. In contrast to cows, plant oil
98
supplements increase milk fat yield in goats (Chilliard et al., 2003, 2007); however, it is
99
unknown whether fish oil supplements can reverse the effects of plant oils. In cows fed fish
100
oil, supplementation with sunflower-seed oil has been associated with greater milk CLA
101
concentrations compared with linseed oil (AbuGhazaleh et al., 2003); however, information is
102
currently unavailable for goats fed fish oil with plant oils. By contrast, corn and barley grains
103
are sources of slowly and rapidly degradable starch in the rumen of goats, respectively
104
(Archimède et al., 1996), and have been selected to determine whether the differences in the
105
nature of starch affect the dairy goat response to fish oil supplements.
106
5
107
2. Material and methods
108
2.1. Animals, experimental diets and management
109
The experimental procedures were approved by the Animal Care Committee of INRA
110
in accordance with the “Use of Vertebrates for Scientific Purposes Act” of 1985. In both
111
experiments, the goats were housed in the INRA herd at Lusignan (France) in a barn with
112
paddocks that included 12 animals and straw bedding. The animals were allocated to
113
treatment groups based on milk yield, milk fat and protein content, parity, and lactation stage.
114
For Experiment 1, 84 Alpine goats with a mean parity of 2.4 ± 1.3 (28 primiparous
115
and 56 multiparous) and 88 ± 15.4 days in milk at the beginning of the experiment were used.
116
The goats were randomly allocated to 7 experimental diets (12 animals/diet) based on alfalfa
117
hay and concentrate without a lipid supplement (Control diet) or with a low dose of fish oil
118
(20 g/d; LFO diet), a low dose of fish oil (20 g/d) with linseed oil (110 g/d; LFLO diet), a low
119
dose of fish oil (20 g/d) with sunflower-seed oil (110 g/d; LFSO diet), a high dose of fish oil
120
(40 g/d; HFO diet), a high dose of fish oil (40 g/d) with linseed oil (90 g/d; HFLO diet), or a
121
high dose of fish oil (40 g/d) with sunflower-seed oil (90 g/d; HFSO diet). The ingredient
122
details of the diets are provided in Table 1. Before the experiment began, the goats were fed
123
the Control diet for a 3-week adaptation period.
124
For Experiment 2, 72 multiparous Alpine goats with a mean parity of 3.1 ± 1.1 and
125
112 ± 12.3 days in milk at the beginning of the experiment were used. The goats were
126
randomly allocated to 6 experimental treatments (12 animals/treatment) based on alfalfa hay
127
and a concentrate without a lipid supplement that was rich in corn and barley grain starch
128
(Control diet) or concentrates with a high dose of fish oil (40 g/d) that were rich in corn and
129
barley grain starch (CBFO diet), rich in corn grain starch (HCFO diet), rich in barley grain
130
starch (HBFO diet), low in corn grain starch (LCFO diet) or low in barley grain starch (LBFO
131
diet). The ingredient details of the diets are provided in Table 2. Before the experiment began,
6
132
the goats were fed a diet with an intermediate starch level (140 g/kg DM) during a 3-week
133
adaptation period; each lot then received an experimental diet (60-200 g starch/kg DM)
134
without the oil supplement for a 4-week pre-experimental period.
135
In both experiments, goats were group-fed. Alfalfa hay was offered ad libitum, and the
136
quantity of concentrate distributed in each paddock was calculated based on the predicted
137
forage consumption (INRA, 1989; Rouel et al., 1997) to cover 115% of the metabolizable
138
energy requirements for maintenance and milk yield. The concentrates were offered in two
139
equal meals at 09:30 and 14:30 h; the animals were blocked into the feeding rack to support
140
similar individual consumption. The oils were manually mixed with the concentrate
141
ingredients immediately before feeding, and clean water was always available. The
142
experimental periods were 4-weeks long, and the goats were milked daily at 08:00 and 16:30
143
h. The experiments were conducted during spring (from March through the beginning of
144
June).
145
146
2.2. Measurements and sampling procedures
147
Representative hay and concentrate samples were collected during the last week of
148
each experiment, and one subsample was used to determine the DM content after 48 h at
149
103°C. Additional subsamples were submitted for chemical composition and FA analyses.
150
The individual goat milk yields were recorded and individual milk subsamples were collected
151
over 4 consecutive milkings, which began at 16:30 h on day 26 of the experimental period.
152
The individual milk samples were stored at 4ºC with 2-bromo-2-nitro-1,3-propanediol (Acros
153
Organics, Geel, Belgium) until they were analyzed for fat, protein and lactose contents at
154
LILCO (Surgères, France). The free FA concentrations resulting from post-milking lipolysis
155
were measured in unpreserved samples collected over 2 consecutive milkings, which began at
156
16:30 h on day 26 of the experimental periods, and stored at 4ºC for 34 h. Additional aliquots
7
157
of unpreserved milk samples were stored at –20ºC, freeze-dried and composed in accordance
158
with a.m. and p.m. milk production for FA analyses. The goats’ live weights were measured
159
at the end of the experimental periods. The same measurements and sampling procedures
160
were used at the end of the adaptation period to generate the covariates (see the statistical
161
analysis section).
162
163
2.3. Chemical analysis
164
The feed ingredients were analyzed for organic matter, ether extract, NDF (Van Soest
165
et al., 1991), ADF (AOAC, 1990; method 973.18), total N (AOAC, 1990; method 988.05),
166
and starch (Faisant et al., 1995). The NDF was assayed without sodium sulfite and α-amylase,
167
and it was expressed with residual ash (the latter was also used for the ADF).
168
The milk fat, protein, and lactose contents were determined through infrared
169
spectrophotometry (AOAC, 1997; method 972-16). The free FA concentrations resulting from
170
post-milking lipolysis were determined using the copper soaps method from Koops and
171
Klomp (1977).
172
Lipid fatty acid methyl esters (FAME) in feed samples were prepared using 1-step
173
extraction-transesterification (Sukhija and Palmquist, 1988) with 23:0 (Sigma, Saint-Quentin
174
Fallavier, France) as an internal standard. For the milk FA composition analysis, the lipid in
175
100 mg of freeze-dried milk was directly methylated using 2 mL of 0.5 M sodium methoxide
176
in anhydrous methanol with 1 mL of hexane at 50ºC for 15 min; after cooling, 1 mL of
177
methanol/HCl (95:5 vol/vol) was added, and the samples were incubated at 50°C for 15 min.
178
The FAME were recovered in 1.5 mL of hexane, washed with 3 mL of aqueous (6% wt/wt)
179
K2CO3, and analyzed using gas chromatography. The FAME profile for a 0.6-µL sample at a
180
split ratio of 1:50 was generated using a Trace-GC 2000 Series gas chromatograph equipped
181
with a flame ionization detector (Thermo Finnigan, Les Ulis, France), a 100-m fused silica
8
182
capillary column (i.d. 0.25 mm) coated with a 0.2-μm cyanopropyl polysiloxane film (CP-Sil
183
88; Chrompack Nederland BV, Middelburg, the Netherlands) and H2 as the carrier and fuel
184
gas. The FAME were separated using a temperature gradient program (Chilliard et al., 2013),
185
and the peaks were identified based on comparing retention times with authentic standards
186
(NCP #463 and #603, Nu-Chek Prep Inc., Elysian, MN, USA; Supelco #37, Supelco Inc.,
187
Bellefonte, PA, USA; L8404 and O5632; Sigma) and by comparing milk samples for which
188
the FA composition was determined through FAME gas chromatography analysis and a gas
189
chromatography-mass spectrometry analysis of the corresponding 4,4-dimethyloxazoline
190
derivatives (Shingfield et al., 2006). Correction factors were estimated using reference butter
191
oil with a known composition to account for the carbon deficiency in the flame ionization
192
detector response for FAME with 4 to 10 carbon atoms (CRM 164; Commission of the
193
European Communities, Community Bureau of Reference, Brussels, Belgium).
194
195
2.4. Calculations and statistical analyses
196
The milk FA melting point was estimated as the sum of the FA melting points
197
weighted by their respective molar proportions (Jensen and Patton, 2000; Toral et al., 2013).
198
The melting points for each FA were obtained from Gunstone et al. (1994) or, if unavailable,
199
from the LipidBank database (http://www.lipidbank.jp; Japanese Conference on the
200
Biochemistry of Lipids). Where the FA melting points were not reported in the two
201
aforementioned sources, they were evaluated using neighboring isomers. The apparent
202
transfer rates of 20:5n-3, 22:5n-3, and 22:6n-3 from fish oil-supplemented diets to milk were
203
calculated as follows: [g milk fat yield × (% FA in milk fat − % FA in control milk fat) / (DM
204
intake × % FA in the diet)] × 100.
205
The data were analyzed by ANOVA using the GLM procedure of the SAS software
206
package (version 9.2, SAS Institute Inc., Cary, NC) with a model that includes the fixed effect
9
207
of diet, the initial measurements at the end of the adaptation periods (covariate) and the
208
random effect from the goats nested within the diet. The covariate term was centered by parity
209
(primiparous or multiparous). Contrast statements were included to compare treatment effects.
210
In Experiment 1, planned comparisons were conducted to test the effects of 1) low vs. high
211
dose of fish oil (LFO + LFLO + LFSO vs. HFO + HFLO + HFSO), 2) linseed oil vs. no
212
linseed oil (LFO + HFO vs. LFLO + HFLO), and 3) sunflower-seed oil vs. no sunflower-seed
213
oil (LFO + HFO vs. LFSO + HFSO). In Experiment 2, planned comparisons were conducted
214
to test the effects of 1) starch source (HCFO + LCFO vs. HBFO + LBFO), and 2) starch level
215
(HCFO + HBFO vs. LCFO + LBFO). The least square means adjusted for the covariance are
216
reported. The differences were considered significant if P<0.05. The relationships between the
217
oil supplements, starch concentrate and milk production and composition were assessed
218
through a principal component analysis (PCA) using the ‘R-project’ software (http://www.r-
219
project.org, version 3.0.1).
220
221
3. Results
222
3.1. Experiment 1
223
The chemical compositions of the diets are provided in Table 1. The supplemented
224
diets provided higher fat and lower starch levels (up to 37% decrease in diets with plant oils)
225
compared with the Control. The protein and energy balances were not negative for any
226
treatment, although the calculated values (mean values for the groups of goats) for DM intake
227
were lower in goats fed the high dose of fish oil alone (–70 g/d) or fish oil with plant oils (on
228
average, –240 g/d; Table 3). Including fish oil markedly increased the 14:0, c9-16:1, and 20-
229
and 22-carbon FA intake, whereas sunflower-seed oil provided the highest levels of dietary
230
18:2n-6 and c9-18:1, and linseed oil provided the most 18:3n-3.
231
The milk, protein and lactose yields were unaffected (P>0.10) by the treatments (Table
10
232
3); however, the combination of fish oil with plant oils increased (P<0.05) milk fat content
233
and yield, with increases in >16-carbon FA secretion (P<0.001) that were greater than
234
decreases in secretion of FA with 16 or fewer carbons (P<0.001). By contrast, supplementing
235
the diet with only fish oil did not affect (P>0.10) the milk fat content or yield. The post-
236
milking free FA concentration decreased in response to the low doses of fish oil with plant
237
oils (LFLO and LFSO) compared with the Control and HFO treatments (P=0.04).
238
As shown in Table 4 and Supplementary Table S1, inclusion of fish oil in the diet
239
altered the milk FA composition, particularly in combination with plant oils. Compared with
240
the Control, diets supplemented with fish oil and plant oils lowered (P<0.01) the
241
concentrations of even-chain saturated FA as well as odd- and branched-chain FA in milk;
242
however, the differences between fish oil plus plant oils and fish oil alone were limited to the
243
first group of FA, and the dose of fish oil had no effect on these 2 groups (P>0.10). The low
244
dose of fish oil with plant oils increased the milk 18:0 concentration compared with the
245
Control and the high dose of fish oil (P<0.001).
246
Regarding monounsatured FA, the concentration of c9-18:1 was affected by the level
247
of fish oil and the presence linseed oil in the diet (P<0.01), with lower milk contents for the
248
HFO, LFLO and, particularly, the HFLO and HFSO treatments (P<0.001), whereas c11-18:1
249
concentration increased with supplemented diets, in particular with fish oil plus plant oils
250
(P<0.001). Fish oil alone induced dose-dependent increases in several trans 18:1 and 18:2
251
isomers in milk (P<0.001); however, greater variations were observed in goats fed fish oil and
252
plant oils, which distinctly responded to each type of vegetable lipid. Thus, the highest levels
253
of t11-18:1 and c9,t11-CLA in milk were observed in goats fed fish and sunflower-seed oils
254
(P<0.001); no differences were observed between the LFSO and HFSO groups (P>0.10; Table
255
4) despite the positive effects of the level of fish oil on these FA (P<0.05). By contrast, the
256
fish and linseed oil diets more efficiently increased non-conjugated 18:2 isomers, such as
11
257
t9,c12-, t11,c15-, t9,t12, and t11,t15-18:2 (P<0.01), often more markedly with the high dose
258
of fish oil (HFLO) than with the low dose (LFLO). Similarly, the t10-, t11-18:1 and c9,t11-
259
CLA concentrations were greater with the HFLO diet than with the LFLO diet (P<0.001).
260
Linseed oil enhanced the 18:3n-3 concentrations in milk and lowered the 18:2n-6
261
concentrations compared with the other treatments, whereas the latter FA was more abundant
262
with the LFSO diet compared with the other treatments (P<0.001). Increases in the proportion
263
of 20:1 and 22:1 isomers, and very long-chain n-3 polyunsaturated FA in milk were related to
264
fish oil level (P<0.001), but the effects of the presence of linseed oil or sunflower-seed oil in
265
the diet was negligible in most cases, except for linseed-oil on the high dose of fish oil.
266
Decreases in the n-6:n-3 FA ratio with fish oil supplementation were greater with the high
267
dose (P<0.001), being partly promoted or counteracted by the presence of linseed oil or
268
sunflower-seed oil in the diet, respectively (P<0.001). The apparent transfer rates of 20:5n-3
269
and 22:6n-3 from the diet to milk were always very low (≤3.4%), while the 22:5n-3 transfer
270
rate showed great variability and ranged from 0 to 10.7% (Table 4).
271
The calculated milk FA melting point was not affected by fish oil level (P>0.10),
272
although it was slightly reduced with the HFO diet and, to a greater extent, with the
273
sunflower-seed and linseed oil supplements (P<0.001; Table 4).
274
275
3.2. Experiment 2
276
The chemical compositions of the diets are shown in Table 2. By design, the corn and
277
barley grains contributed equally to the starch levels in the Control and CBFO diets, although
278
the starch levels were slightly lower when the fish oil was included. Compared with the
279
CBFO diet, the HCFO and HBFO diets provided similar starch levels, but using either corn or
280
barley grain, respectively, whereas the LCFO and LBFO diets provided only 35% of the
281
starch levels compared with the HCFO and HBFO diets, respectively. The calculated energy
12
282
and protein balances were positive for each group, and DM intake numerical values (mean
283
values for the groups) were greater in the CBFO and HCFO treatments (on average, +245 g/d;
284
Table 5). Fish oil supplements increased the 14:0, 16:0, c9-16:1, c11-18:1, c11-20:1, 20:5n-3,
285
22:5n-3, and 22:6n-3 intake, while the starch type and level were primarily related to changes
286
in c9-18:1 and 18:2n-6 intake.
287
As shown in Table 5, the milk yield was unaffected (P=0.19) by the diets, which
288
increased the daily lactose and fat yield compared with the Control (P=0.03). Although corn
289
grain led to slightly higher milk fat contents than barley grain (P=0.03), neither the type nor
290
the level of starch concentrate in the diet modified milk fat yield (P>0.10). The secretion of
291
milk FA with 16 or fewer carbons was greater in the diets with fish oil (P<0.001), but that of
292
>16 carbons only increased with the HCFO treatment (P<0.001). The post-milking free FA
293
concentration was unaffected by fish oil or the starch concentrate level and type (P=0.54).
294
Dietary supplementation with fish oil altered the milk FA composition, and the
295
response varied due to the type and level of starch concentrate (Table 6 and Supplementary
296
Table S2). Thus, although fish oil decreased the 18:0 concentration in milk (P<0.001), the
297
lowest levels were observed from the diets with barley grain as the sole starch source, with no
298
differences compared to the CBFO diet (P>0.10). In general, high levels of starch tended to
299
decrease milk 18:0 contents to a greater extent than low starch levels (P=0.08). The changes
300
in milk short- and medium-chain FA were limited; the concentration of 6:0 + 8:0 + 10:0 were
301
slightly affected by starch source, with lower values in diets with corn grain than in those
302
based on barley grain (P<0.001). However, the 12:0 +14:0 + 16:0 concentrations were not
303
affected by starch source (P>0.10) and slightly affected by starch level (P<0.01). Regarding
304
the effect of starch level, the proportion of odd- and branched-chain FA was greater with the
305
low-starch diets (P<0.001) although there were distinct effects on individual FA (see
306
Supplemental Table S2).
13
307
Changes in c9-18:1 mirrored those of 18:0, with greater decreases in the diets with
308
barley grain as the sole starch source (P<0.001), whereas the contrary was observed for c11-
309
18:1 (P<0.001). However, starch level had no effect on these two cis 18:1 isomers (P>0.10).
310
Supplementing the diet with fish oil increased the concentrations of trans 18:1 and 18:2
311
isomers in milk (P<0.001), and the greatest levels of t11-18:1 and c9,t11-CLA were observed
312
with the HBFO treatment (P<0.001), due to the effects of both starch source and level
313
(P<0.001), but concentrations were not different compared with the CBFO treatment. On the
314
other hand, the increases in t10-18:1 concentration were always limited even though
315
significant (P<0.001), and no effects of starch source or level were observed. Generally, the
316
goats fed fish oil and barley grain as the sole starch concentrate had higher (P<0.001) t,c- and
317
t,t-18:2, which in most cases were not affected by starch level (P>0.05). The milk t10,c12-
318
CLA proportions remained low (≤0.015 g/100 g FA); the greatest increase was observed with
319
the diets that only included barley grain (P=0.002).
320
Compared with the Control, fish oil decreased the levels of 18:2n-6 in milk, and the
321
effect of starch source resulted in lower concentrations in treatments containing only corn
322
grain than in those based on barley grain (P<0.001). Similarly, the 18:3n-3 concentration was
323
lower with the fish oil and corn-based concentrates (P<0.001). However, fish oil
324
supplementation enhanced the proportion of most 20- and 22-carbon FA in milk (P<0.001),
325
although the 20:5n-3, 22:5n-3, and 22:6n-3 concentrations were higher with diets that
326
included barley grain (CBFO, HBFO, and LBFO; P<0.001). No effect of starch level was
327
observed on milk n-3 PUFA (P>0.10). Diets containing barley grain and fish oil yielded
328
higher apparent transfer rates from the diets to the milk (5.1, 35.7, and 7.6% for 20:5n-3,
329
22:5n-3, and 22:6n-3, respectively) than the diets that only included corn grain (2.5, 18.7, and
330
3.7%, respectively).
331
14
332
3.3. Principal component analysis
333
A PCA of the data from both experiments discriminated two principal components
334
(PC) that described 36.6 (PC1) and 22.8% (PC2) of the total variation in milk yield and
335
composition as well as milk FA composition, whereas the third principal component (PC3)
336
accounted for 7.0% of the total variability. The scores plot (Figure 1a) showed that PC1
337
discriminates based on oil supplementation; the samples from goats fed fish and plant oils
338
were in the negative range. Additionally, samples in the positive range were primarily
339
distributed in two clusters: the Control and the low dose of fish oil (20 g/d) vs. the high dose
340
of fish oil (40 g/d). In the latter cluster, we only observed discrimination for samples from
341
diets rich in corn grain starch. Two major FA groups were apparent from the loading plot PC1
342
× PC2 (Figure 1b), which primarily included 18-carbon FA with a trans double bond, c11-
343
and c13-18:1, 10-O-18:0, and 22:1 FA (corresponding to biohydrogenation intermediates and
344
Δ9-desaturase products with two double bonds) that clustered as a group, negatively
345
correlated with PC1 and positively correlated with PC2. However, most FA with 16 or fewer
346
carbons (including branched-chain FA) clustered as a group that positively correlated with
347
PC1. In addition, 20:3n-3, 20:5n-3, 22:5n-3, 22:6n-3, and 16:1 isomers formed a third group
348
that positively correlated with PC2, whereas 18:0 and c9-18:1 negatively correlated with PC2.
349
350
4. Discussion
351
Despite the ample data on including plant lipids in dairy goat diets (Chilliard et al.,
352
2007; Mele et al., 2008; Bernard et al., 2009), few studies have investigated the effects of fish
353
oil on milk production and composition in this species (Kitessa et al., 2001; Sanz Sampelayo
354
et al., 2007; Bernard et al., 2010). Thus, the present study aimed to comprehensively evaluate
355
the effects of dietary fish oil on dairy performance and the milk FA profile in goats, with
356
particular emphasis on the interaction between fish oil and either plant oils or starch
15
357
concentrates in the diet. The two experiments were performed using the same herd at the same
358
time of year, under identical conditions and with equivalent forage types and concentrate
359
proportions, to support comparisons between experiments. The data were also analyzed
360
through PCA to summarize the relationships between oil treatments, starch concentrates and
361
milk production and composition.
362
363
4.1. Effects from dietary fish oil alone
364
Dietary fish oil alone (0.8-1.6% diet DM) had either no effect (Experiment 1) or a
365
positive effect (Experiment 2) on dairy performance (including milk fat yield). Although a
366
previous report on goats (Kitessa et al., 2001) showed a negative impact from unprotected fish
367
oil (3% diet DM) on DM intake, milk yield and milk fat content, the discrepancy between the
368
two studies may be due to differences in lipid dosage. In cows and ewes, similar levels of fish
369
oil as in the present study did not affect milk yield but have consistently been associated with
370
reduced milk fat levels (Offer et al., 1999; Loor et al., 2005a; Capper et al., 2007; Toral et al.,
371
2010a; Table 7). A marked shortage of 18:0 for endogenous c9-18:1 synthesis has been
372
proposed to explain fish oil-induced milk fat reduction in cows and ewes (Loor et al., 2005a;
373
Shingfield et al, 2006; Toral et al., 2010a) due to its potentially negative impact on
374
maintaining milk fat fluidity, which would be aggravated by a concomitant increase in milk
375
trans 18:1 concentrations (with melting points higher than body temperature; Gunstone et al.,
376
1994). Thus, the reasons that could explain the absence of milk fat depression in the present
377
study may be due to a particular milk FA profile response in goats. In Experiment 1, the
378
proportion of c9-18:1 was only slightly lower with the HFO diet, and trans 18:1 increases
379
were limited with the LFO and HFO treatments. In Experiment 2, lower milk c9-18:1
380
concentrations were not accompanied by large increases in trans 18:1 isomers; increases in
381
milk FA with low melting points (namely, short-chain FA and c9,t11-CLA) may have
16
382
compensated for their potentially negative effect on milk fat fluidity, which is supported by
383
the stable calculated milk FA melting point.
384
By contrast, the bases for the different milk FA responses in Experiments 1 and 2 are
385
not readily apparent but may be related to the different oil compositions because goats in the
386
second trial ingested, on average, 62% more 20:5n-3 than goats fed the high dose of fish oil in
387
the first trial. This FA seems more toxic to ruminal bacteria than 22:6n-3 (Maia et al., 2007),
388
and a meta-analysis of cows fed fish oil showed that 20:5n-3 intake was a better predictor of
389
lower milk fat than 22:6n-3 intake (Chilliard et al., 2001). Although we did not observe lower
390
milk fat in the present study, the aforementioned differences in milk 18:0 and c9-18:1 (and
391
higher concentration of milk t11-18:1 and c9,t11-CLA in Experiment 2) are consistent with
392
the notion that fish oil had a greater impact on the hydrogenation of trans 18:1 to 18:0 in
393
Experiment 2 compared with Experiment 1. In contrast, the differences in fish oil composition
394
(or dosage) produced few obvious effects on the apparent transfer rates of 20:5n-3 and 22:6n-
395
3 from the diet to milk, which were low (1.4-3.4% in Experiment 1 and 2.1-8.5% in
396
Experiment 2). Because the milk 20:5n-3, 22:5n-3, and 22:6n-3 concentrations were low but
397
detectable without fish oil supplementation in this and previous studies (e.g., Loor et al.,
398
2005a; Shingfield et al., 2006; Toral et al., 2010a), the calculated apparent transfer rates were
399
corrected for the respective FA concentrations in the control milk. The transfer rates were
400
within the same range or slightly higher than the rates calculated using the same approach in
401
cows that were fed fish oil (2.0-4.1%; Chilliard et al., 2001; Loor et al., 2005a; Gama et al.,
402
2008) and relatively lower than the rates in ewes (9.2-15.5%; Capper et al., 2007; Toral et al.,
403
2010a). However, the apparent transfer rates of 22:5n-3 are often higher in the three species
404
(up to 38% in Experiment 2), which may be explained, at least in part, by its lower
405
disappearance in the rumen (Lee et al., 2008; Toral et al., 2010b) and its more efficient uptake
406
and secretion by the mammary gland (Chilliard et al., 2000; Loor et al., 2005a) compared
17
407
with 20:5n-3 and 22:6n-3.
408
409
4.2. Effects from the interaction between fish oil and starch level
410
Starch level had no effect on milk yield and fat content in goats fed fish oil, which is
411
consistent with data from cows that received diets with different forage:concentrate ratios and
412
fish oil alone (Gama et al., 2008) or with sunflower-seed oil (Shingfield et al., 2005). In the
413
scores plot from the PCA, a clear discrimination was not observed based on starch level.
414
However, milk FA composition changes suggest that high dietary starch levels affected the
415
ruminal environment in goats fed fish oil. First, the general decrease in milk odd- and
416
branched-chain FA concentrations from high-starch diets compared with low-starch diets
417
suggests that the rumen bacterial biomass was lower (Fievez et al., 2012). In addition, higher
418
starch levels from barley grain with fish oil enhanced the c9,t11-CLA and t11-18:1
419
concentrations (but not for t10-18:1) in goat milk (Experiment 2), which suggests enhanced
420
inhibition for the last step of ruminal biohydrogenation, which did not stimulate t10-18:1
421
formation at the expense of t11-18:1. The opposite result (i.e., lower milk c9,t11-CLA and
422
t11-18:1 but higher t10-18:1 levels) has been observed in cows fed fish oil and sunflower-
423
seed oil when the proportion of wheat-based concentrates was increased, regardless of the
424
forage source (Shingfield et al., 2005). However, increasing the corn grain starch levels with
425
fish oil enhanced milk 18:0 and cis-9 18:1 concentrations for goats in the present study, which
426
might be related to enhanced 18-carbon FA intake (36%) in the HCFO compared with the
427
LCFO (41 vs. 30 g/d, respectively; Table 2) treatments. However, increasing the dietary
428
starch from corn grain with fish oil did not affect the milk trans FA concentrations for goats
429
in Experiment 2 or for cows in a previous study (Gama et al., 2008), which suggests enhanced
430
stability in ruminal biohydrogenation processes for diets with starchy concentrates of slower
431
degradability. Overall, the results indicated relevant interactions between oils, starch
18
432
concentrates and ruminant species; however, the available information remains limited, and
433
further research (including direct inter-species comparisons) is warranted.
434
435
4.3. Effects from the interaction between fish oil and the type of starch concentrate
436
Studies investigating the effect from the type of starch concentrate in dairy ruminants
437
have been limited to cows fed diets without lipid supplements (Jurjanz et al., 2004; Cabrita et
438
al., 2009) and goats fed sunflower-seed oil (Bernard et al., 2012). The lower milk fat contents
439
in the HBFO diet compared with the HCFO diet are consistent with results from dairy cows
440
fed rapidly or slowly degradable starch without fish oil (wheat grain or corn grain and
441
potatoes, respectively; Jurjanz et al., 2004; Cabrita et al., 2009). However, in goats fed high-
442
concentrate diets with sunflower-seed oil alone (Bernard et al., 2012), the starch source
443
(wheat or corn grain) did not affect milk and fat yield. In the milk FA profile, compared with
444
diets that only included corn grain, barley grain enhanced the concentrations of 18:2n-6 and
445
18:3n-3 as well as a range of trans 18:1 and 18:2 intermediates in the milk from goats fed fish
446
oil at the expense of 18:0 and c9-18:1; however, we did not observe differences in the low
447
t10-18:1 concentration increases. Similar variations in the 18-carbon FA (except for the t10-
448
18:1 isomer) have been demonstrated in cows (Jurjanz et al., 2004) and goats (Bernard et al.,
449
2012) fed starch concentrates with different degradability (wheat grain vs. potatoes or corn
450
grain) without fish oil. Overall, these data suggest slower and less extensive biohydrogenation
451
in the rumen from animals fed rapidly degradable starch, which could be explained by the
452
effects of starch source on the rumen bacterial community (Weimer et al., 2010). The higher
453
apparent transfer rates of n-3 polyunsaturated FA from fish oil to milk in goats fed barley
454
grain compared with goats that only received corn grain further support this hypothesis. The
455
low t10-18:1 concentrations suggest that 18 carbon FA biohydrogenation in the rumen was
456
not accompanied by shifts towards t10-18:1 in goats fed fish oil, regardless of the type and
19
457
level of dietary starch.
458
459
4.4. Effects from the interaction between fish oil and plant oils
460
Among the dietary treatments tested in the present study, the fish oil (either at low or
461
high dose) and plant oil combinations produced the greatest effects on milk FA composition.
462
This observation is summarized in the scores plot from the PCA, which clearly discriminates
463
goats that received fish oil with sunflower-seed oil or linseed oil from goats without
464
supplemental lipids or with fish oil alone (Figure 1a). Plant oils increase milk fat yield in
465
goats (Chilliard et al., 2007; Mele et al., 2008; Bernard et al., 2009), which was the prevailing
466
observation for the LFSO, LFLO and HFLO treatments due to the higher long-chain FA
467
output, which more than compensated for the moderately inhibited de novo FA synthesis
468
(Table 3). Accordingly, most biohydrogenation-derived FA with 18 carbons (such as trans
469
18:1 and 18:2, c13-18:1, and 10-O-18:0) were grouped in the loading plot from the PCA and
470
opposite the cluster with de novo synthesized FA (composed of even-, odd- and branched-
471
chain FA; Figure 1b).
472
Fish oil with sunflower-seed oil combinations were more efficient at increasing milk
473
c9,t11-CLA and t11-18:1 concentrations than fish oil with linseed oil; greater proportions of
474
these FA in cow milk have been observed for diets supplemented with fish oil and sunflower
475
seeds compared with fish oil and linseeds due to differences in the accumulation of t11-18:1,
476
but not c9,t11-CLA, in ruminal digesta (AbuGhazaleh et al., 2003). Thus, we could
477
hypothesize that responses to fish oil with sunflower-seed oil or linseed oil may also be
478
mediated by mechanisms other than inhibiting trans 18:1 saturation (Wąsowska et al., 2006).
479
Because the level of t11,c15-18:2 in milk was greater for the HFLO diet than for the LFLO
480
diet, regardless of the lower 18:3n-3 intake (17% decrease) in the HFLO diet, the fish oil also
481
likely interferes with reduction of this 18:2 isomer, as suggested in previous studies on cows
20
482
(Loor et al., 2005b; Wąsowska et al., 2006).
483
Greater increases in milk t10-18:1 concentrations were observed in response to diets
484
supplemented with fish oil and plant oils compared with those supplemented with fish oil
485
alone. Nevertheless, the proportion of this trans 18:1 isomer remained low (≤1.1% of total
486
FA); the ratio t10-18:1/t11-18:1 was much lower than in observations on cows fed similar
487
levels of fish oil and plant lipids (AbuGhazaleh et al., 2003; Shingfield et al., 2006; Cruz-
488
Hernandez et al., 2007). This difference confirms the low propensity for the t10 shift in goats
489
(Shingfield et al., 2010; Table 7), which may be related to the absence of milk fat depression
490
in goats fed fish oil and plant oils, which is in contrast to cows (Shingfield et al., 2006; Cruz-
491
Hernandez et al., 2007) and ewes (Toral et al., 2010a).
492
Finally, the negative effect on post-milking free FA concentrations from the LFSO and
493
LFLO treatments in milk is consistent with the lower levels of spontaneous lipolysis in goat
494
milk with diets supplemented with 18:2n-6 and 18:3n-3-rich lipids (Chilliard et al., 2003,
495
2013; Eknæs et al., 2009). This reduction could impact the milk sensory quality by reducing
496
the development of goat flavor (Chilliard et al., 2003); however, this reduction was not
497
observed with the HFSO and HFLO treatments or after adding fish oil alone (Experiments 1
498
and 2), which suggests that FO has no effect per se or may compensate for the effects from
499
plant oils.
500
501
5. Conclusions
502
In contrast to cows, adding fish oil to low- or high-starch diets with or without plant
503
oil supplements changes the levels of bioactive FA in milk without decreasing milk fat
504
content and yield in goats. This observation may be related to specific milk FA responses in
505
goats; for example, the fish oil-mediated decrease in 18:0 and c9-18:1 and increase in t10-
506
18:1 (≤1.1% total FA) concentrations is lower in goats, and a shift in ruminal
21
507
biohydrogenation from t11 towards the t10 pathways is not observed with fish oil and plant
508
oils (in contrast to observations from cows fed diets with similar oil supplements and starch
509
levels). The results from Experiment 1 indicate that fish oil and sunflower-seed oil
510
combinations more efficiently increase milk c9,t11-CLA and t11-18:1 concentrations
511
compared with fish oil and linseed oil or fish oil alone. In addition, using high vs. low levels
512
of fish oil in combination with sunflower-seed oil also decreases the n-6:n-3 FA ratio in milk.
513
In Experiment 2, 18-carbon FA changes (particularly 18:2n-6, 18:3n-3, c9,t11-CLA,
514
t11-18:1, and 18:0) suggest that diets rich in barley grain starch with fish oil induce less
515
extensive ruminal biohydrogenation of dietary FA and better inhibit the trans 18:1 saturation
516
compared with diets supplemented with the same levels of fish oil and low barley grain starch
517
levels or high corn grain starch levels, with greater increases in the bioactive FA with 18
518
carbons. Similarly, the apparent 20:5n-3 and 22:6n-3 transfer rates from fish oil to milk were
519
also higher for treatments with barley grain compared with the corn grain only treatments, but
520
the values remained low (<9%). However, the level and type of starch concentrate in the diet
521
has no effect on milk and milk components yield in goats fed fish oil. Direct inter-species
522
comparisons are required to more precisely characterize the different responses to dietary fish
523
oil supplements and their interaction with plant oils and starch concentrates.
524
Acknowledgments
525
P.G. Toral was granted a post-doctoral fellowship from Fundación Alfonso Martín
526
Escudero (Madrid, Spain). The authors gratefully acknowledge P. Capitan, E. Tixier, C.
527
Delavaud, and K. J. Shingfield for help with milk FA analyses; D. Bany for feedstuff FA
528
analyses; and P. Guillouet, E. Bruneteau, B. Ranger, F. Dugué, N. Riquet, and M. Van
529
Pethegen for help with animal experiments.
530
References
531
AbuGhazaleh, A.A., Schingoethe, D.J., Hippen, A.R., Kalscheur, K.F., 2003. Conjugated
22
532
linoleic acid and vaccenic acid in rumen, plasma, and milk of cows fed fish oil and fats
533
differing in saturation of 18 carbon fatty acids. J. Dairy Sci. 86, 3648–3660.
534
535
536
537
AOAC, 1990. Official Methods of Analysis of the Association of Official Analytical
Chemists, 15th ed. AOAC International, Arlington, VA, USA.
AOAC, 1997. Official Methods of Analysis of the Association of Official Analytical
Chemists, 16th ed. AOAC International, Gaithersburg, MD, USA.
538
Archimède, H., Sauvant, D., Hervieu, J., Ternois, F., Poncet, C., 1996. Effects of the nature of
539
roughage and concentrate and their proportion on ruminal characteristics of non
540
lactating goats, consequences on digestive interactions. Anim. Feed Sci. Technol. 58,
541
267–282.
542
Bernard, L., Leroux, C., Rouel, J., Bonnet, M., Chilliard, Y., 2012. Effect of level and type of
543
starchy concentrate on tissue lipid metabolism, gene expression, and milk fatty acid
544
secretion in Alpine goats receiving a diet rich in sunflower oil. Br. J. Nutr. 107, 1147–
545
1159.
546
Bernard, L., Mouriot, J., Rouel, J., Glasser, F., Capitan, P., Pujos-Guillot, E., Chardigny, J.M.,
547
Chilliard, Y., 2010. Effects of fish oil and starch added to a diet containing sunflower-
548
seed oil on dairy goat performance, milk fatty acid composition and in vivo Δ9-
549
desaturation of [13C] vaccenic acid. Br. J. Nutr. 104, 346–354.
550
Bernard, L., Shingfield, K.J., Rouel, J., Ferlay, A., Chilliard, Y., 2009. Effect of plant oils in
551
the diet on performance and milk fatty acid composition in goats fed diets based on
552
grass hay or maize silage. Br. J. Nutr. 101, 213–224.
553
Cabrita, A.R.J., Vale, J.M.P., Bessa, R.J.B., Dewhurst, R.J., Fonseca, A.J.M., 2009. Effects of
554
dietary starch source and buffers on milk responses and rumen fatty acid
555
biohydrogenation in dairy cows fed maize silage-based diets. Anim. Feed Sci.
556
Technol. 152, 267–277.
23
557
Capper, J.L., Wilkinson, R.G., Mackenzie, A.M., Sinclair, L.A., 2007. The effect of fish oil
558
supplementation of pregnant and lactating ewes on milk production and lamb
559
performance. Animal 1, 889–898.
560
Chilliard, Y., Ferlay, A., Doreau, M., 2001. Effect of different types of forages, animal fat or
561
marine oils in cow's diet on milk fat secretion and composition, especially conjugated
562
linoleic acid (CLA) and polyunsaturated fatty acids. Livest. Prod. Sci. 70, 31–48.
563
Chilliard, Y., Ferlay, A., Mansbridge, R.M., Doreau, M., 2000. Ruminant milk fat plasticity:
564
nutritional control of saturated, polyunsaturated, trans and conjugated fatty acids. Ann.
565
Zootech. 49, 181–205.
566
Chilliard, Y., Ferlay, A., Rouel, J., Lamberett, G., 2003. A review of nutritional and
567
physiological factors affecting goat milk lipid synthesis and lipolysis. J. Dairy Sci. 86,
568
1751–1770.
569
Chilliard, Y., Glasser, F., Ferlay, A., Bernard, L., Rouel, J., Doreau, M., 2007. Diet, rumen
570
biohydrogenation and nutritional quality of cow and goat milk fat. Eur. J. Lipid Sci.
571
Technol. 109, 828–855.
572
Chilliard, Y., Rouel, J., Guillouet, P., 2013. Goat alpha-s1 casein genotype interacts with the
573
effect of extruded linseed feeding on milk fat yield, fatty acid composition and post-
574
milking lipolysis. Anim. Feed Sci. Technol. 185, 140–149.
575
Cruz-Hernandez, C., Kramer, J.K.G., Kennelly, J.J., Glimm, D.R., Sorensen, B.M., Okine,
576
E.K., Goonewardene, L.A., Weselake, R.J., 2007. Evaluating the conjugated linoleic
577
acid and trans 18:1 isomers in milk fat of dairy cows fed increasing amounts of
578
sunflower oil and a constant level of fish oil. J. Dairy Sci. 90, 3786–3801.
579
Eknæs, M., Havrevoll, Ø., Volden, H., Hove, K., 2009. Fat content, fatty acid profile and off-
580
flavours in goats milk - Effects of feed concentrates with different fat sources during
581
the grazing season. Anim. Feed Sci. Technol. 152, 112–122.
24
582
Faisant, N., Plachot V., Kozlowski F., Pacouret M.P., Colonna P., Champ M., 1995. Resistant
583
starch determination adapted to products containing high level of resistant starch. Sci.
584
Aliments 15, 83–89.
585
Fievez, V., Colman, E., Castro-Montoya, J.M., Stefanov, I., Vlaeminck, B., 2012. Milk odd-
586
and branched-chain fatty acids as biomarkers of rumen function – An update. Anim.
587
Feed Sci. Technol. 172, 51–65.
588
Gagliostro, G.A., Rodríguez, A., Pellegrini, P.A., Gatti, P., Muset, G., Castañeda, R.A.,
589
Colombo, D., Chilliard, Y., 2006. Effects of fish oil or sunflower plus fish oil
590
supplementation on conjugated linoleic acid (CLA) and omega 3 fatty acids in goat
591
milk. Rev. Argent. Prod. Anim. 26, 71–87.
592
Gama, M.A.S., Garnsworthy, P.C., Griinari, J.M., Leme, P.R., Rodrigues, P.H.M., Souza,
593
L.W.O., Lanna, D.P.D., 2008. Diet-induced milk fat depression: Association with
594
changes in milk fatty acid composition and fluidity of milk fat. Livest. Sci. 115, 319–
595
331.
596
597
598
599
600
601
602
603
Gunstone, F.D., Harwood, J.L., Padley, F.B., 1994. The Lipid Handbook. 2nd ed. Chapman
and Hall, New York, NY, USA.
INRA, 1989. Ruminant nutrition. In: Jarrige, R. (Ed.), Recommended allowances and feed
tables. John Libbey Eurotext, London, England, Paris, France.
Jensen, R.G., Patton, S., 2000. The effect of maternal diets on the mean melting points of
human milk fatty acids. Lipids 35, 1159–1161.
Jurjanz, S., Monteils, V., Juaneda, P., Laurent, F., 2004. Variations of trans octadecenoic acid
in milk fat induced by feeding different starch-based diets to cows. Lipids 39, 19–24.
604
Kitessa, S.M., Gulati, S.K., Ashes, J.R., Fleck, E., Scott, T.W., Nichols, P.D., 2001.
605
Utilisation of fish oil in ruminants - II. Transfer of fish oil fatty acids into goats' milk.
606
Anim. Feed Sci. Technol. 89, 201–208.
25
607
608
Koops, J., Klomp, H., 1977. Rapid colorimetric determination of free fatty acids (lipolysis) in
milk by the copper soap method. Milk Dairy J. 31, 56–72.
609
Lee, M.R.F., Shingfield, K.J., Tweed, J.K.S., Toivonen, V., Huws, S.A., Scollan, N.D., 2008.
610
Effect of fish oil on ruminal biohydrogenation of C18 unsaturated fatty acids in steers
611
fed grass or red clover silages. Animal 2, 1859–1869.
612
613
Lock, A.L., Bauman, D.E., 2004. Modifying milk fat composition of dairy cows to enhance
fatty acids beneficial to human health. Lipids 39, 1197–1206.
614
Loor, J.J., Doreau, M., Chardigny, J.M., Ollier, A., Sébédio, J.L., Chilliard, Y., 2005a. Effects
615
of ruminal or duodenal supply of fish oil on milk fat secretion and profiles of trans-
616
fatty acids and conjugated linoleic acid isomers in dairy cows fed maize silage. Anim.
617
Feed Sci. Technol. 119, 227–246.
618
Loor, J. J., Ferlay, A., Ollier, A., Ueda, K., Doreau, M., Chilliard, Y., 2005b. High-
619
concentrate diets and polyunsaturated oils alter trans and conjugated isomers in bovine
620
rumen, blood, and milk. J. Dairy Sci. 88, 3986–3999.
621
Maia, M.R.G., Chaudhary, L.C., Figueres, L., Wallace, R.J., 2007. Metabolism of
622
polyunsaturated fatty acids and their toxicity to the microflora of the rumen. Antonie
623
van Leeuwenhoek 91, 303–314.
624
Martínez Marín, A.L., Gómez-Cortés, P., Gómez Castro, A.G., Juárez, M., Pérez Alba, L.M.,
625
Pérez Hernández, M., de la Fuente, M.A., 2011. Animal performance and milk fatty
626
acid profile of dairy goats fed diets with different unsaturated plant oils. J. Dairy Sci.
627
94, 5359–5368.
628
Mele, M., Serra, A., Buccioni, A., Conte, G., Pollicardo, A., Secchiari, P., 2008. Effect of
629
soybean oil supplementation on milk fatty acid composition from Saanen goats fed
630
diets with different forage:concentrate ratios. Ital. J. Anim. Sci. 7, 297–311.
631
Offer, N.W., Marsden, M., Dixon, J., Speake, B.K., Thacker, F.E., 1999. Effect of dietary fat
26
632
supplements on levels of n-3 poly-unsaturated fatty acids, trans acids and conjugated
633
linoleic acid in bovine milk. Anim. Sci. 69, 613–625.
634
Rouel, J., Bocquier, F., Chilliard, Y., 1997. Effets du niveau de production laitière et de
635
l'apport d'aliment concentré sur l'ingestion volontaire d'ensilage de maïs chez la chèvre
636
laitière. Rencontres Recherches Ruminants 4, 92.
637
Roy A., Chardigny J.M., Bauchart D., Ferlay A., Lorenz S., Durand D., Gruffat D.,
638
Faulconnier Y., Sébédio J.L., Chilliard Y., 2007. Butters rich either in trans-10-C18:1
639
or in trans-11-C18:1 plus cis-9, trans-11 CLA differentially affect plasma lipids and
640
aortic fatty streak in experimental atherosclerosis in rabbits. Animal. 1, 467-476.
641
Sanz Sampelayo, M.R., Chilliard, Y., Schmidely, P., Boza, J., 2007. Influence of type of diet
642
on the fat constituents of goat and sheep milk. Small Rumin. Res. 68, 42–63.
643
Shingfield, K.J., Ahvenjärvi, S., Toivonen, V., Äröla, A., Nurmela, K.V.V., Huhtanen, P.,
644
Griinari, J.M., 2003. Effect of dietary fish oil on biohydrogenation of fatty acids and
645
milk fatty acid content in cows. Anim. Sci. 77:165–179.
646
647
648
649
Shingfield, K.J., Bernard, L., Leroux, C., Chilliard, Y., 2010. Role of trans fatty acids in the
nutritional regulation of mammary lipogenesis in ruminants. Animal 4, 1140–1166.
Shingfield K.J., Chilliard Y., Toivonen V., Kairenius P., Givens D.I., 2008. Trans fatty acids
and bioactive lipids in ruminant milk. Adv. Exp. Med. Biol. 606, 3-65.
650
Shingfield, K.J., Reynolds, C.K., Hervás, G., Griinari, J.M., Grandison, A.S., Beever, D.E.,
651
2006. Examination of the persistency of milk fatty acid composition responses to fish
652
oil and sunflower oil in the diet of dairy cows. J. Dairy Sci. 89, 714–732.
653
Shingfield, K.J., Reynolds, C.K., Lupoli, B., Toivonen, V., Yurawecz, M.P., Delmonte, P.,
654
Griinari, J.M., Grandison, A.S., Beever, D.E., 2005. Effect of forage type and
655
proportion of concentrate in the diet on milk fatty acid composition in cows given
656
sunflower oil and fish oil. Anim. Sci. 80, 225–238.
27
657
Sukhija, P.S., Palmquist, D.L., 1988. Rapid method for determination of total fatty-acid
658
content and composition of feedstuffs and feces. J. Agric. Food Chem. 36, 1202–1206.
659
Toral, P.G., Bernard, L., Chilliard, Y., Glasser, F., 2013. Short communication: Diet-induced
660
variations in milk fatty acid composition have minor effects on the estimated melting
661
point of milk fat in cows, goats, and ewes: insights from a meta-analysis. J. Dairy Sci.
662
96, 1232–1236.
663
Toral, P.G., Frutos, P., Hervás, G., Gómez-Cortés, P., Juárez, M., de la Fuente, M.A., 2010a.
664
Changes in milk fatty acid profile and animal performance in response to fish oil
665
supplementation, alone or in combination with sunflower oil, in dairy ewes. J. Dairy
666
Sci. 93, 1604–1615.
667
Toral, P. G., Shingfield, K. J., Hervás, G., Toivonen, V., Frutos, P., 2010b. Effect of fish oil
668
and sunflower oil on rumen fermentation characteristics and fatty acid composition of
669
digesta in ewes fed a high concentrate diet. J. Dairy Sci. 93, 4804–4817.
670
Van Soest, P.J., Robertson, J.B., Lewis, B.A., 1991. Methods for dietary fiber, neutral
671
detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J. Dairy
672
Sci. 74, 3583–3597.
673
Weimer, P.J., Stevenson, D.M., Mertens, D.R., 2010. Shifts in bacterial community
674
composition in the rumen of lactating dairy cows under milk fat-depressing conditions.
675
J. Dairy Sci. 93:265–278.
676
Wąsowska, I., Maia, M.R.G., Niedźwiedzka, K.M., Czauderna, M., Ribeiro, J., Devillard, E.,
677
Shingfield, K.J., Wallace, R.J., 2006. Influence of fish oil on ruminal
678
biohydrogenation of C18 unsaturated fatty acids. Br. J. Nutr. 95, 1199–1211.
28
679
680
681
Table 1
Ingredients and chemical composition of the ingested diets, and energy and protein balance in
lactating goats (Experiment 1).
Dieta
Low dose of fish oil
LFO LFLO LFSO
High dose of fish oil
HFO HFLO HFSO
54.3
29.4
10.4
5.1
0
0
0
0.8
55.2
27.6
9.9
5.7
0.8
0
0
0.8
57.5
18.4
8.3
9.4
0.9
4.7
0
0.9
56.7
18.7
8.4
9.6
0.9
0
4.8
0.9
56.0
25.0
10.1
6.5
1.6
0
0
0.8
56.5
18.8
8.5
9.6
1.8
4.0
0
0.9
58.4
18.0
8.1
9.2
1.7
0
3.8
0.8
918
413
266
142
200
18
6.05
917
415
269
144
188
25
6.05
914
415
275
158
125
71
6.55
915
412
273
158
128
72
6.12
916
418
273
147
170
33
6.47
915
411
272
158
128
72
6.55
913
419
278
157
123
70
6.40
97
1.64
64
98
1.07
59
107
0.28
33
107
0.14
28
100
0.28
50
107
0.00
28
107
0.71
42
Control
Ingredients (g/kg DM)
Alfalfa hay
Corn grain
Beet pulp
Soybean cake
Fish oilb
Linseed oilc
Sunflower-seed oild
Minerals-vitaminse
Chemical composition (g/kg DM)
Organic matter
Neutral detergent fibre
Acid detergent fibre
Crude protein
Starch
Ether extract
Net energy for lactationf (MJ/kg DM)
Protein digestible in the intestinef
(g/kg DM)
Energy balancef (MJ/d)
Protein balancef (g/d)
682
683
684
685
686
687
688
689
690
691
692
693
694
695
a
Goats received diets based on alfalfa hay without a lipid supplement (Control), or with a low dose of
fish oil alone (LFO), a low dose of fish oil with linseed oil (LFLO), a low dose of fish oil with
sunflower-seed oil (LFSO), a high dose of fish oil alone (HFO), a high dose of fish oil with linseed oil
(HFLO) or a high dose of fish oil with sunflower oil (HFSO).
b
Fish oil (SA Daudruy, Dunkerque, France) contained (g/kg fatty acids): 14:0 (85.1), 15:0 (5.7), 16:0
(174.8), c9-16:1 (85.7), 18:0 (29.2), c9-18:1 (147.2), c11-18:1 (30.4), 18:2n-6 (38.7), 18:3n-3 (17.0),
c11-20:1 (30.5), 20:2n-6 (5.7), 20:4n-6 (7.0), 20:5n-3 (90.7), 22:5n-3 (19.6), and 22:6n-3 (72.7).
c
Linseed oil (SA Vandeputte, Muscron, Belgique) contained (g/kg fatty acids): 16:0 (59.6), 18:0
(39.5), c9-18:1 (194.3), 18:2n-6 (166.0), and 18:3n-3 (522.9).
d
Sunflower-seed oil (Huileries de Lapalisse, Lapalisse, France) contained (g/kg fatty acids): 16:0
(70.0), 18:0 (30.8), c9-18:1 (281.5), c11-18:1 (8.0), 18:2n-6 (578.7), and 18:3n-3 (14.5).
e
Minerals-vitamins mix (Centre Atlantique Aliments, Bayers, France) declared as containing (g/kg
pre-mix): Ca, 200; P, 60; Mg, 50; Na, 40; Zn, 6; Mn, 4; Cu, 0.7.
f
Calculated according to INRA (1989).
29
696
697
698
Table 2
Ingredients and chemical composition of the ingested diets, and energy and protein balance in
lactating goats (Experiment 2).
Dieta
Control
Ingredients (g/kg DM)
Alfalfa hay
Corn grain
Barley grain
Beet pulp
Soybean cake
Fish oilb
Mineral-vitaminsc
Chemical composition (g/kg DM)
Organic matter
Neutral detergent fibre
Acid detergent fibre
Crude protein
Starch
Starch from corn grain
Starch from barley grain
Ether extract
Net energy for lactationd (MJ/kg DM)
Protein digestible in the intestined
(g/kg DM)
Energy balanced (MJ/d)
Protein balanced (g/d)
699
700
701
702
703
704
705
706
707
708
709
710
High starch
CBFO
HCFO
HBFO
Low starch
LCFO
LBFO
55.6
14.5
17.9
6.0
5.3
0.0
0.7
60.5
12.0
14.7
5.1
5.8
1.2
0.6
62.0
23.9
0.0
7.4
4.9
1.2
0.6
56.3
0.0
31.5
4.5
5.7
1.3
0.6
55.6
8.6
0.0
25.3
8.6
1.3
0.6
54.5
0.0
10.7
23.7
9.2
1.3
0.7
934
402
277
125
204
103
101
21
5.86
933
420
294
127
168
85
83
32
5.81
935
428
301
120
170
170
0
33
5.83
931
405
280
130
178
0
178
32
5.88
928
457
314
135
61
61
0
29
5.93
927
451
309
140
60
0
60
29
5.91
86
3.13
64
87
3.63
76
82
3.34
55
88
2.42
54
93
2.42
68
96
1.64
68
a
Goats received diets based on alfalfa hay containing no lipid supplement and a concentrate rich in
starch from corn and barley grains (Control) or containing a fish oil supplement and concentrates rich
in starch from corn and barley grains (CBFO), rich in starch from corn grain (HCFO), rich in starch
from barley grain (HBFO), low in starch from corn grain (LCFO) or low in starch from barley grain
(LBFO).
b
Fish oil (SA Daudruy, Dunkerque, France) contained (g/kg fatty acids): 14:0 (87.7), 16:0 (18.8), c916:1 (88.5), 18:0 (34.0), c9-18:1 (101.4), c11-18:1 (30.8), 18:2n-6 (18.3), 18:3n-3 (6.1), c11-20:1
(12.2), 20:2n-6 (1.4), 20:4n-6 (8.1), 20:5n-3 (147.1), 22:5n-3 (20.1), and 22:6n-3 (71.3).
c
Minerals-vitamins mix (Centre Atlantique Aliments, Bayers, France) declared as containing (g/kg
pre-mix): Ca, 200; P, 60; Mg, 50; Na, 40; Zn, 6; Mn, 4; Cu, 0.7.
d
Calculated according to INRA (1989).
e
No statistical analysis for intake and balance values were conducted because goats were group-fed.
30
711
712
713
Table 3
Effects of diet supplementation with fish oil alone or combined with linseed or sunflower-seed oil on calculated intake, animal performance and post-milking
lipolysis in dairy goats (Experiment 1).
Dieta
Low dose of fish oil
LFO
LFLO LFSO
High dose of fish oil
HFO HFLO HFSO
2,554
0.12
8.27
0.19
1.11
8.42
0.42
19.56
2.22
0.00
0.00
0.00
0.00
2,536
1.74
11.36
1.82
1.66
10.75
0.99
19.38
2.54
0.58
1.72
0.37
1.38
2,324
1.78
16.08
1.80
5.67
28.58
1.63
31.73
57.03
0.67
1.72
0.37
1.38
2,280
1.82
17.08
1.92
4.74
37.68
1.75
74.80
3.85
0.58
1.72
0.37
1.38
2,478
3.36
14.26
3.44
2.17
12.80
1.55
18.62
2.82
1.16
3.45
0.74
2.76
2,269
3.39
18.16
3.42
5.45
27.67
2.07
29.23
47.35
1.16
3.45
0.74
2.76
3.04
88
138
101b
35.5a
29.9a
28.8d
3.07
91
142
103b
36.4a
29.5a
32.5cd
2.97
90
141
114a
29.7b
22.9b
58.9a
2.99
92
140
118a
29.5b
24.5b
59.7a
3.17
92
148
109ab
37.1a
29.0a
35.8c
29.1b
47.5a
33.5c
29.6ab
46.4e
33.3c
30.5a
47.6bd
38.2ab
30.7a
46.9de
39.8a
29.1b
46.8de
34.4c
Control
SED
Pb
1
Contrastsc
2
2,374
3.43
19.26
3.53
4.75
35.15
2.19
64.67
4.00
1.16
3.45
0.74
2.76
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
3.11
92
152
116a
30.0b
24.9b
59.0a
2.90
86
138
108ab
29.2b
24.9b
50.9b
0.075
2.2
4.0
3.5
1.37
0.92
1.86
0.28
0.41
0.19
0.02
<0.001
<0.001
<0.001
0.42
0.68
0.16
0.87
0.79
0.05
0.22
0.33
0.97
0.71
0.02
<0.001
<0.001
<0.001
0.04
0.32
0.17
0.06
<0.001
<0.001
<0.001
30.2a
49.0c
37.3b
29.6ab
47.8b
37.5b
0.38
0.31
0.77
0.03
<0.001
<0.001
0.06
0.002
0.33
0.02
<0.001
<0.001
0.06
0.03
<0.001
3
d
Average nutrient intake (g/d)
Dry matter
14:0
16:0
c9-16:1
18:0
c9-18:1
c11-18:1
18:2n-6
18:3n-3
c11-20:1
20:5n-3
22:5n-3
22:6n-3
Yield (g/d)
Milk
Protein
Lactose
Fat
<16 carbon FA
16 carbon FA
>16 carbon FA
Composition (g/kg)
Protein
Lactose
Fat
31
Free fatty acids (mmol/ 100 g fat)
Live weight (kg)
714
715
716
717
718
719
720
0.99a
58.8
0.85ab
58.7
0.57b
58.9
0.46b
58.3
1.24a
58.2
0.82ab
58.6
0.85ab
58.5
0.15
0.53
0.04
0.97
0.02
0.65
0.05
0.62
0.03
0.88
a
Goats received diets based on alfalfa hay without a lipid supplement (Control), or with a low dose of fish oil alone (LFO), a low dose of fish oil with linseed
oil (LFLO), a low dose of fish oil with sunflower-seed oil (LFSO), a high dose of fish oil alone (HFO), a high dose of fish oil with linseed oil (HFLO) or a
high dose of fish oil with sunflower oil (HFSO).
b
Probability of significant effects due to experimental diet. Within a row, mean values with different superscript differ significantly (P<0.05).
c
Contrasts are designed as: 1) low vs. high dose of fish oil (LFO + LFLO + LFSO vs. HFO + HFLO + HFSO), 2) linseed oil vs. no linseed oil (LFO + HFO
vs. LFLO + HFLO), and 3) sunflower-seed oil vs. no sunflower-seed oil (LFO + HFO vs. LFSO + HFSO).
d
No statistical analysis were conducted because goats were group-fed.
32
721
722
723
Table 4
Effects of diet supplementation with fish oil alone or combined with linseed or sunflower-seed oil on milk fatty acid (FA) composition (g/100 g FA) in dairy
goats (Experiment 1).
∑ even-chain saturated FA
4:0
6:0 + 8:0 + 10:0
12:0 + 14:0 + 16:0
18:0
∑ odd- and branched-chain FA
∑ cis-monounsaturated FA
c9-18:1d
c11-18:1
∑ 20:1
∑ 22:1
∑ trans FA
t10-18:1
t11-18:1
t9,c12-18:2
t11,c15-18:2
t9,t12-18:2
t11,t15-18:2
c9,t11-CLAe
∑ CLA
∑ n-6 Polyunsaturated FA
18:2n-6
∑ n-3 Polyunsaturated FA
18:3n-3
20:5n-3
22:5n-3
Control
72.61a
2.01a
15.06a
48.22a
6.99cd
5.41a
17.23ab
15.41a
0.48e
0.04e
<0.01d
1.41e
0.10c
0.33e
<0.001d
0.01d
0.02c
<0.001c
0.18e
0.18e
2.15b
1.93b
0.43c
0.27cd
0.03e
0.09b
Dieta
Low dose of fish oil
LFO LFLO LFSO
70.26b 53.59c 51.61cd
1.76b 1.99a 1.95a
14.90bc 11.40ac 10.62f
45.64b 31.18d 30.62d
7.55bc 8.67a 7.96ab
5.52a 4.13c 4.16bc
16.95ab 15.80b 17.50a
14.88ab 13.78b 15.36a
0.50de 0.63c 0.70b
0.13cd 0.11d 0.17c
0.12c 0.11c 0.11c
3.21d 18.28b 17.40b
0.25c 0.65b 0.67ab
0.96de 8.80c 11.46a
0.01cd 0.09b 0.02c
0.06d 2.45b 0.13cd
0.02c 0.06b 0.01c
0.02c 0.52a 0.04bc
0.36e 3.04c 4.41a
0.36e 3.22c 4.45a
2.10b 1.65d 2.51a
1.91b 1.49d 2.35a
0.56c 1.02a 0.53c
0.32c 0.78a 0.23d
0.07c 0.07c 0.05d
0.11b 0.09b 0.09b
High dose of fish oil
HFO
HFLO HFSO
b
68.29
51.43d 53.63c
a
2.01
1.88ab 1.93a
14.33b 11.10e 11.52d
45.13b 31.82d 33.61c
6.29d
6.26d
6.19d
a
b
5.33
4.41
4.18bc
16.09ab 14.18c 14.26c
13.71b 11.62c 11.93c
0.55d
0.80a
0.74b
0.41a
0.35b
0.35b
b
a
0.23
0.28
0.25b
5.15c
20.67a 17.94b
bc
0.36
1.04a 0.74ab
1.72d
9.99b 11.66a
c
0.03
0.12a
0.03c
0.11cd
2.72a
0.30c
c
a
0.01
0.07
0.02c
0.03c
0.51a
0.07b
d
b
0.91
3.61
4.05a
d
b
0.91
3.67
4.08a
1.94c
1.61d
2.13b
c
d
1.74
1.42
1.93b
0.73b
1.12a
0.73b
c
b
0.32
0.65
0.26cd
0.10b
0.12a
0.12a
a
a
0.15
0.14
0.15a
SED
0.72
0.047
0.38
0.54
0.298
0.098
0.53
0.51
0.019
0.017
0.009
0.511
0.135
0.368
0.006
0.079
0.005
0.015
0.134
0.137
0.055
0.053
0.047
0.027
0.004
0.008
b
P
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
1
0.24
0.35
<0.001
0.02
<0.001
0.67
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.09
0.02
<0.001
0.02
0.21
0.44
0.03
0.06
<0.001
<0.001
<0.001
0.15
<0.001
<0.001
Contrastsc
2
<0.001
0.40
<0.001
<0.001
0.08
<0.001
0.005
0.003
<0.001
0.03
0.03
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.06
0.05
3
<0.001
0.29
<0.001
<0.001
0.61
<0.001
0.23
0.21
<0.001
0.41
0.41
<0.001
0.005
<0.001
0.59
0.10
0.78
0.04
<0.001
<0.001
<0.001
<0.001
0.85
0.006
0.79
0.14
33
22:6n-3
n-6:n-3 FA ratio
PUFA:saturated FA ratio
Calculated milk FA melting point (ºC)
Apparent transfer ratef
20:5n-3
22:5n-3
22:6n-3
724
725
726
727
728
729
730
731
732
733
0.02d 0.05b
5.23a 3.84b
0.039e 0.049de
39.1a 39.1a
–
2.4
–
6.8
–
2.4
0.03c 0.04bc
1.65d 5.32a
0.181b 0.153c
36.2d 36.8c
2.9
1.7
0
0
1.4
1.9
0.08a
2.69c
0.061d
38.5b
2.9
10.7
2.7
0.08a
1.49d
0.199a
36.3d
3.4
8.5
2.7
0.09a
2.92c
0.143c
37.3c
3.3
10.1
3.1
0.004
0.185
0.006
0.19
–
–
–
<0.001
<0.001
<0.001
<0.001
–
–
–
<0.001
<0.001
0.15
0.92
–
–
–
0.20
<0.001
<0.001
<0.001
–
–
–
0.55
<0.001
<0.001
<0.001
–
–
–
a
Goats received diets based on alfalfa hay without a lipid supplement (Control), or with a low dose of fish oil alone (LFO), a low dose of fish oil with linseed
oil (LFLO), a low dose of fish oil with sunflower-seed oil (LFSO), a high dose of fish oil alone (HFO), a high dose of fish oil with linseed oil (HFLO) or a
high dose of fish oil with sunflower oil (HFSO).
b
Probability of significant effects due to experimental diet. Within a row, mean values with different superscript differ significantly (P<0.05).
c
Contrasts are designed as: 1) low vs. high dose of fish oil (LFO + LFLO + LFSO vs. HFO + HFLO + HFSO), 2) linseed oil vs. no linseed oil (LFO + HFO
vs. LFLO + HFLO), and 3) sunflower-seed oil vs. no sunflower-seed oil (LFO + HFO vs. LFSO + HFSO).
d
Contains t13-, t14-, t15-, and c10-18:1 as minor components.
e
Contains t7,c9- and t8,c10-conjugated linoleic acid (CLA) as minor components.
f
Calculated as: [g milk fat yield × (% FA in milk fat − % FA in control milk fat) / (DM intake × % FA in the diet)] × 100. No statistical analyses were
conducted because goats were group-fed.
34
734
735
736
Table 5
Effects of starch concentrate level and type and supplementation with fish oil on calculated intake, animal performance and post-milking lipolysis in dairy
goats (Experiment 2).
Dieta
High starch
CBFO
HCFO
HBFO
Low starch
LCFO
LBFO
3.04
0.19
9.93
0.05
1.13
6.84
0.33
20.02
3.35
0.00
0.00
0.00
0.00
3.30
3.53
17.21
3.41
2.50
10.29
1.49
19.89
3.86
0.46
5.59
0.76
2.71
3.32
3.49
16.45
3.41
2.51
11.85
1.50
21.45
3.45
0.46
5.59
0.76
2.71
3.09
3.56
17.48
3.41
2.38
8.56
1.46
17.65
4.00
0.46
5.59
0.76
2.71
3.09
3.47
15.01
3.42
2.24
8.25
1.46
14.75
3.28
0.46
5.59
0.76
2.71
3.73
105b
163b
105b
37.3b
33.2b
26.7b
4.02
117a
178a
118a
43.2a
37.4a
28.9ab
3.90
112ab
175a
120a
42.1a
37.9a
31.7a
4.04
116a
184a
116a
43.1a
36.9a
27.8b
28.4c
28.5
43.7c
29.9abc
29.5
44.3bc
31.1a
29.0
45.2ab
28.8bc
28.8
45.9a
Control
Contrastc
SED
b
P
1
2
3.04
3.50
15.43
3.42
2.22
7.18
1.45
13.58
3.52
0.46
5.59
0.76
2.71
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
3.97
114a
184a
120a
44.7a
38.7a
28.3b
4.07
117a
182a
122a
45.8a
40.2a
27.2b
0.10
3.0
4.9
3.7
1.43
1.31
1.05
0.19
0.07
0.03
0.03
<0.001
<0.001
<0.001
0.26
0.22
0.99
0.73
0.005
0.06
0.003
0.63
0.58
<0.001
0.45
<0.001
0.010
<0.001
30.6ab
29.0
46.0a
29.9abc
28.7
45.1ab
0.65
0.30
0.44
0.04
0.28
0.002
0.03
0.57
0.76
0.62
0.91
0.95
d
Average nutrient intake (g/d)
Dry matter
14:0
16:0
c9-16:1
18:0
c9-18:1
c11-18:1
18:2n-6
18:3n-3
c11-20:1
20:5n-3
22:5n-3
22:6n-3
Yield (g/d)
Milk
Protein
Lactose
Fat
<16 carbon FA
16 carbon FA
>16 carbon FA
Concentration (g/kg)
Fat
Protein
Lactose
35
Free fatty acids (mmol/100 g fat)
Live weight (kg)
737
738
739
740
741
742
1.87
62.9
1.49
62.4
1.25
62.3
1.47
62.5
1.35
62.6
1.78
62.7
0.27
0.68
0.54
0.99
0.24
0.84
0.45
0.72
a
Goats received diets based on alfalfa hay without a lipid supplement and with a concentrate rich in corn and barley grain starch (Control) or with a fish oil
supplement and concentrates rich in corn and barley grain starch (CBFO), rich in corn grain starch (HCFO), rich in barley grain starch (HBFO), low in corn
grain starch (LCFO) or low in barley grain starch (LBFO).
b
Probability of significant effects due to experimental diet. Within a row, mean values with different superscript differ significantly (P<0.05).
c
Contrasts are designed as: starch source (HCFO + LCFO vs. HBFO + LBFO), and 2) starch level (HCFO + HBFO vs. LCFO + LBFO).
d
No statistical analysis were conducted because goats were group-fed.
36
743
744
Table 6
Effects of starch concentrate level and type and supplementation with fish oil on milk fatty acid (FA) composition (g/100 g FA) in dairy goats (Experiment 2).
Dieta
∑ even-chain saturated FA
4:0
6:0 + 8:0 + 10:0
12:0 + 14:0 + 16:0
18:0
∑ odd- and branched-chain FA
∑ cis-monounsaturated FA
c9-18:1d
c11-18:1
∑ 20:1
∑ 22:1
∑ trans FA
t10-18:1
t11-18:1
t9,c12-18:2
t11,c15-18:2
t9,t12 18:2
t11,t15-18:2
c9,t11-CLAe
t10,c12-CLA
∑ CLA
∑ n-6 Polyunsaturated FA
18:2n-6
∑ n-3 Polyunsaturated FA
18:3n-3
20:5n-3
22:5n-3
Control
73.35a
2.36ab
15.83b
48.87ab
5.94a
4.47c
15.88a
13.73a
0.39c
0.06d
<0.01d
1.08c
0.10c
0.44d
0.01d
0.03c
0.008d
<0.01c
0.27c
<0.0001d
0.28c
2.63a
2.41a
0.74c
0.53ab
0.06c
0.12c
High starch
CBFO
HCFO
bc
69.81
70.35bc
ab
2.35
2.42a
16.82a
15.96b
bc
48.11
47.42c
2.35cd
4.09b
bc
4.58
4.47c
cd
9.97
12.16b
6.94cd
9.35b
a
0.66
0.53b
0.39ab
0.40ab
ab
0.13
0.11c
6.29a
5.33b
a
0.45
0.39ab
3.46ab
2.68c
b
0.07
0.05c
a
0.43
0.26b
0.026b
0.021c
b
0.04
0.03b
2.02a
1.50b
c
0.009
0.009c
2.09a
1.58b
c
2.06
1.92cd
1.76c
1.65cd
a
1.68
1.21b
0.51bc
0.48c
a
0.28
0.15b
a
0.35
0.21b
HBFO
69.79c
2.21c
17.33a
48.04bc
1.85d
4.80b
8.80e
5.89d
0.66a
0.34c
0.13ab
6.82a
0.44a
4.04a
0.08a
0.51a
0.028ab
0.05a
2.05a
0.014ab
2.13a
2.34b
2.03b
1.79a
0.57a
0.33a
0.35a
Low starch
LCFO
LBFO
b
71.37
71.14bc
ab
2.34
2.28bc
17.01a
17.48a
abc
48.64
49.49a
2.93c
1.99d
a
5.26
5.23a
c
10.67
9.10de
7.75c
6.09d
b
0.55
0.66a
0.44a
0.38bc
bc
0.11
0.15a
4.79b
5.38b
b
0.37
0.39ab
2.26c
2.83bc
b
0.06
0.08a
b
0.31
0.51a
0.025bc 0.032a
0.04b
0.06a
1.27b
1.49b
bc
0.010
0.015a
1.36b
1.59b
d
1.87
1.94cd
1.56d
1.61d
b
1.38
1.72a
0.46c
0.51bc
b
0.19
0.28a
b
0.26
0.35a
Contrastsc
SED
0.58
0.039
0.24
0.49
0.283
0.09
0.41
0.399
0.029
0.018
0.007
0.32
0.025
0.244
0.003
0.031
0.0017
0.004
0.149
0.0013
0.16
0.06
0.050
0.07
0.020
0.019
0.0171
b
P
<0.001
<0.01
<0.001
0.07
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.004
<0.001
<0.001
1
0.48
<0.001
<0.001
0.14
<0.001
0.07
<0.001
<0.001
<0.001
0.001
<0.001
0.002
0.19
<0.001
<0.001
<0.001
<0.001
<0.001
0.02
0.002
0.02
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
2
0.05
0.95
0.02
0.008
0.08
<0.001
0.16
0.10
0.13
0.05
0.05
0.003
0.15
0.002
0.07
0.43
0.03
0.22
0.03
0.43
0.03
<0.001
<0.001
0.47
0.06
0.77
0.19
37
22:6n-3
n-6:n-3 FA ratio
PUFA:saturated FA ratio
Calculated milk FA melting point (ºC)
Apparent transfer ratef
20:5n-3
22:5n-3
22:6n-3
745
746
747
748
749
750
751
752
753
0.02d
3.66a
0.053e
37.8
–
–
–
0.17b
1.29c
0.095ab
37.2
4.5
33.4
6.7
0.08c
1.57b
0.076cd
37.4
2.1
15.1
3.2
0.21a
1.29c
0.100a
37.1
5.8
35.9
8.6
0.11c
1.37bc
0.075d
37.6
2.9
22.3
4.1
0.18ab
1.18c
0.086bc
37.4
5.1
37.9
7.5
0.013
0.086
0.004
0.18
–
–
–
<0.001
<0.001
<0.001
0.08
–
–
–
<0.001
0.008
<0.001
0.14
–
–
–
0.68
0.08
0.07
0.16
–
–
–
a
Goats received diets based on alfalfa hay without a lipid supplement and with a concentrate rich in corn and barley grain starch (Control) or with a fish oil
supplement and concentrates rich in corn and barley grain starch (CBFO), rich in corn grain starch (HCFO), rich in barley grain starch (HBFO), low in corn
grain starch (LCFO) or low in barley grain starch (LBFO).
b
Probability of significant effects due to experimental diet. Within a row, mean values with different superscript differ significantly (P<0.05).
c
Contrasts are designed as: starch source (HCFO + LCFO vs. HBFO + LBFO), and 2) starch level (HCFO + HBFO vs. LCFO + LBFO).
d
Contains t13-, t14-, t15-, and c10-18:1 as minor components.
e
Contains t7,c9- and t8,c10-conjugated linoleic acid (CLA) as minor components.
f
Calculated as: [g milk fat yield × (% FA in milk fat − % FA in control milk fat) / (DM intake × % FA in the diet)] × 100. No statistical analyses were
conducted because goats were group-fed.
38
754
755
756
Table 7
Effects of diet supplementation with fish oil (FO) alone or combined with sunflower-seed oil (SO) on the changes in milk fat content and yield and milk fatty
acid concentration in dairy goats and cows.
Species
Goat
Cow
Oil supplement,
% DM
0.8% FO
0.6% FO
Foragea,
% DM
55
59
Starchb,
% DM
18.8
10.1
Change in fat
content, %
-0.6
-8.4
Change in fat
yield, %
+2.0
-0.9
Change in
18:0, Δ
+0.56
-1.80
Change in
t11-18:1, Δ
+0.63
+1.62
Change in
t10-18:1, Δ
+0.15
+0.38
Experiment 1
Offer et al. (1999)
Shingfield et al. (2003)
Goat
Cow
Cow
1.6% FO
1.6% FO
1.6% FO
56
58
58
17.0
10.4
14.0
+2.7
-24.0
-7.0
+7.9
-22.6
-23.6
-0.70
-5.57
-15.11
+1.39
+6.47
+7.59
+0.26
+1.61
+0.80
Experiment 1
Shingfield et al. (2006)
Goat
Cow
1.7% FO + 3.8% SO
1.5% FO + 3.0% SO
58
65
12.3
15.7
+11.9
-37.0
+6.9
-38.2
-0.80
-2.34
+11.33
+6.19
+0.64
+7.40
Study
Experiment 1
Offer et al. (1999)
757
758
759
a
Forage source: alfalfa hay (Experiment 1), grass silage (Offer et al., 1999; Shingfield et al., 2003) or corn silage (Shingfield et al., 2006).
Starch source: corn grain (Experiment 1), barley grain (Offer et al., 1999; Shingfield et al., 2003) or wheat middlings (Shingfield et al., 2006). In italics,
value estimated according to INRA (1989).
b
39
760
Figure 1
761
Principal component analysis of the data derived from the analysis of 156 milk samples (Experiments 1 and 2). (a) Sample distribution based on the first two
762
principal components (PC1 and PC2); each point represents an individual goat. (b) Plot of experimental variables projected based on the first two principal
763
components (PC1 and PC2) that describes the association between milk yield, milk composition and milk fatty acid concentrations.
40
764
FIGURE 1a
a)
No lipid supplementation (exp. 1 and 2; n = 24)
5
Low dose of fish oil (exp. 1; n = 6)
Fish oil and linseed oil (2 doses of fish oil, exp. 1; n = 24)
0
PC2 (22.8%)
Fish oil and sunflower-seed oil (2 doses of fish oil, exp. 1; n = 24)
High dose of fish oil:
High starch from corn grain (exp. 1 and 2; n = 24)
High starch from barley grain (exp. 2; n = 12)
-5
High starch from corn and barley grains (exp. 2; n = 12)
Low starch from corn grain (exp. 2; n = 12)
Low starch from barley grain (exp. 2; n = 12)
-10
765
-5
0
5
PC1 (36.6%)
41
FIGURE 1b
b)
1.0
766
c11-16:1
20:5n-3
t9-16:1
t,t-CLA
22:6n-3
t9c12-18:2
c13-18:1
∑20:1
20:3n-3
22:5n-3
t12-18:1
c11-18:1
20:4n-3 19:0
c9-16:1
20:2n-6 22:3n-3
c9t11-CLA
20:3n-6
10-O-18:0 ∑22:1
5:0
7:0
c9t12-18:2
9:0
c9t13-18:2 t9t12-18:2
t9-18:1
18:0 iso
t10-18:1
18:3n-3
t11-18:1
t11c15-18:2
11:0
milk yield
t6+7+8-18:1
c9t14-18:2
13:0 6:0
f at yield
8:0
c9t15-18:2
t4-18:1
t5-18:1
4:0 15:0 iso
17:0
t11t15-18:2
15:0 anteiso
10:0
21:0
12:0
15:0
c9-12:1
c9-10:1
20:4n-6
f at content
24:0
c9-14:1 16:0
22:0
16:0 iso
13:0 iso
14:0
22:4n-6
c12-18:1
20:0
18:2n-6 18:3n-6
14:0 iso
0
–0.5
PC2 (22.8%)
0.5
22:5n-6
17:0 anteiso
c9-17:1
18:0
–1.0
c9-18:1
–1.0
767
–0.5
0
0.5
1.0
PC1 (36.6%)
42