Get documents about "Uterine, Ovarian, and Production Responses of Lactating Dairy Cows - PDF
Document Sample
J. Dairy Sci. 85:755–764
American Dairy Science Association, 2002.
Uterine, Ovarian, and Production Responses of Lactating Dairy Cows
to Increasing Dietary Concentrations of Menhaden Fish Meal
R. Mattos,* C. R. Staples,* J. Williams,* A. Amorocho,*
M. A. McGuire,† and W. W. Thatcher*
*Department of Animal Sciences
University of Florida, Gainesville FL 32611-0920
†University of Idaho, Moscow, ID
ABSTRACT (Key words: fish meal, prostaglandin, milk, n-3
fatty acid)
The primary objective was to determine whether the
dietary polyunsaturated fatty acids, eicosapentaenoic Abbreviation key: CLA = conjugated linoleic acid,
(EPA, C20:5, n-3) and docosahexaenoic (DHA, C22:6, COX = cyclooxygenase, DHA = docosahexaenoic acid,
n-3), present in fish meal (FM) can attenuate uterine EPA = eicosapentaenoic acid, FM = fish meal, PGFM
secretion of PGF2α in response to a challenge with = PGF2α metabolite, PUFA = polyunsaturated fatty
estradiol and oxytocin in lactating dairy cows. Cycling acid.
multiparous cows (n = 32) were fed diets containing 0
(0FM), 2.6 (2.6FM), 5.2 (5.2FM), or 7.8% menhaden INTRODUCTION
FM (7.8FM). The diet consisting of 7.8FM also con-
Pregnancy rate is an important determinant of
tained fish oil (0.28% of dietary dry matter) to increase
profitability in dairy operations. High herd pregnancy
intake of EPA and DHA. Average dry matter intake
rate increases the lifetime milk production of cows and
was 24.9 kg/d and unaffected by diet. Combined in-
reduces the costs associated with AI and replacement
takes of EPA and DHA averaged 0, 12.8, 24.1, and 54.0
animals. Anestrus, poor estrous detection, and low
g/d from the 0FM, 2.6FM, 5.2FM, and 7.8FM diets,
conception rates are factors that contribute to low
respectively. At 30 to 34 d after initiation of dietary
pregnancy rates in dairy herds. Several management
treatments, cows received an i.m. injection of 100 µg
and pharmacological practices have been developed to
of GnRH followed by i.m. administration of 25 and 15 reduce the proportion of anestrous cows and to improve
mg of PGF2α after 7 and 8 d, respectively. Synchronous estrous detection. Increased adoption of protocols for
ovulation was induced by an injection of 3000 IU of synchronization of ovulation that provide acceptable
human chorionic gonadotropin (hCG) given 24 h later conception rates after insemination without detection
on d 9. Subsequent luteal phase increases in plasma of estrus has the potential to reduce the impact of poor
progesterone concentrations did not differ (0.88 ng/ml estrous detection on herd fertility. This study will focus
per day). At 15 d after hCG injection, cows were in- on a potential strategy to reduce embryonic mortality
jected with estradiol-17β (3 mg, i.v.) at 0900 h and and increase pregnancy rates.
oxytocin (100 IU, i.v.) at 1300 h. Plasma PGF2α metab- Up to 40% of total embryonic losses are estimated
olite concentrations after oxytocin injection were re- to occur between d 8 and 17 of pregnancy (Thatcher
duced in cows fed diets containing FM compared with et al., 1994). This high proportion of losses is coincident
those fed 0FM. Milk production (39.1 kg/d) and concen- with the period of embryonic inhibition of uterine
trations of fat, protein, or urea nitrogen in milk were PGF2α secretion, suggesting that some loss may be
not affected by diet. Feeding fish meal and fish oil occurring because certain embryos are unable to in-
containing eicosapentaenoic acid and docosahexaenoic hibit secretion of PGF2α. Therefore strategies to fur-
acid reduced the proportion of n-6 fatty acids and in- ther inhibit secretion of PGF2α may result in increased
creased that of n-3 fatty acids in milk in a dose-respon- embryonic survival and pregnancy rates.
sive manner. Menhaden fish meal (FM) is used in dairy cow ra-
tions as a source of RUP. Fish meal contains oil (8%
of DM) with relatively high concentrations of two poly-
unsaturated fatty acids (PUFA) of the n-3 family, eico-
Received June 26, 2001. sapentaenoic acid (EPA, C20:5) and docosahexaenoic
Accepted November 5, 2001.
Corresponding author: W. W. Thatcher; e-mail: thatcher@animal. acid (DHA, C22:6). Concentrations of EPA and DHA
ufl.edu. in fish oil have been reported to be 10.8 and 11.1% of
755
756 MATTOS ET AL.
total fatty acids (Donovan et al., 2000). Eicosapentae- compositions of the diets are described in Tables 1 and
noic acid and DHA have inhibited secretion of PGF2α 2. The fatty acid composition of the experimental diets
in various animal cell culture systems (Levine and is reported in Table 3. Menhaden fish oil was added
Worth, 1984; Achard et al., 1997), including bovine to the 7.8FM diet (0.28% of dietary DM) to further
endometrial cells (Mattos et al., 2001). Inhibiting uter- increase the intake of EPA and DHA. Cows assigned
ine secretion of PGF2α by feeding EPA and DHA may to the 7.8FM diet were fed the 5.2FM diet for 4 d before
prevent regression of the CL and increase fertility being offered the 7.8FM diet in order to minimize pos-
rates (Burke et al., 1997; Staples et al., 1998; Mattos sible intake problems. The day of introduction of the
et al., 2000). Feeding FM to lactating multiparous cows 7.8FM diet was considered the first experimental day
for 29 d reduced the secretion of PGF2α induced by (d 0).
estradiol and oxytocin injected on d 15 of a synchro- Cows were housed in a free-stall, open-sided barn
nized estrous cycle (Thatcher et al., 1997). In humans, fitted with Calan gates (American Calan, Inc., North-
consumption of high amounts of fish oil was reported wood, NH) to allow measurement of individual feed
to prolong gestation (Olsen et al., 1992). Also, infusing intake. The barn was equipped with sprinklers and
ewes with a suspension containing n-3 fatty acids fans that operated when the air temperature exceeded
blocked a betamethasone-induced increase in plasma 25°C. The study was conducted from April 29 to June
concentrations of PGE2 and delayed occurrence of par- 30, 1999. The concentrate portions of the diets were
turition in prepartum sheep (Baguma-Nibasheka et mixed and stored in metal bins of 1.8-tonne capacity.
al., 1999). Concentrate mixtures and forage sources were mixed
Depression in feed intake and milk fat and protein in a weighing and mixing unit (American Calan, Inc.)
concentrations can occur in cows fed fish oil (Cant et and offered twice daily at 0830 and 1330 h to allow 5
al., 1997). Feeding FM has resulted in either an in- to 10% orts (as-fed basis). Orts were collected once
crease (Calsamiglia et al., 1999) or no change in milk daily and weighed. The DM concentration of silage
protein concentrations (Spain et al., 1995). was monitored once weekly to maintain the proper
Feeding high concentrations of n-3 fatty acids usu- forage-to-concentrate ratio of diets. Samples of forage
ally results in increased concentrations of these fatty and concentrate mixes were collected weekly, compos-
acids in plasma (Spain et al., 1995) and milk lipids ited monthly, and analyzed for CP, NDF, ADF, ether
(Jones et al., 1998). Evidence for the incorporation of extract, and minerals (Dairy One, DHIA Forage Test-
dietary fatty acids into uterine lipids has been reported ing Laboratory, Ithaca, NY) using wet chemistry. Cows
by Burns et al. (2000). Feeding FM (5% DM basis) for were milked three times daily at 0500, 1200, and 1900
approximately 60 d to nonlactating beef cows resulted h. Milk weights were recorded by calibrated electronic
in an increase in the proportion of n-3 fatty acids pres- milk meters at each milking.
ent in caruncular endometrium when compared to a
control group not fed FM.
Synchronization of the Estrous Cycle
Our hypothesis was that feeding EPA and DHA us-
ing FM and fish oil would reduce uterine secretion of Cows were submitted to a protocol for synchroniza-
PGF2α in lactating dairy cows. The objectives were to tion of estrus so that all animals started an estrous
determine the effect of increasing dietary concentra- cycle on the same day. Because it was not physically
tion of FM on uterine secretion of PGF2α induced by feasible to conduct the oxytocin challenge and serial
estradiol and oxytocin, and to evaluate the effect of collection of blood samples using all cows in one day,
diet on plasma progesterone concentrations, DMI, half of the cows initiated the synchronization proce-
milk production, and milk composition. dure on experimental d 23 and the other half on d 27.
The synchronization protocol was initiated with an
MATERIALS AND METHODS i.m. injection of 100 µg of GnRH (Cystorelin, Merial,
Duluth, GA) followed by i.m. administration of PGF2α
Cows and Diets (Lutalyse, Pharmacia Corporation, Kalamazoo, MI)
Cycling multiparous Holstein cows (n = 32), averag- after 7 (25 mg) and 8 d (15 mg). Synchronous ovulation
ing 116 ± 12.5 DIM, with healthy reproductive tracts, was induced by an injection of 3000 IU of hCG (Choru-
were assigned randomly to diets containing Menhaden lon, Intervet, Netherlands) given 24 h later (Figure 1).
FM (Sealac, Omega Protein, Reedville, VA) at the fol- On d 15 of the synchronized estrous cycle, (47 to 51
lowing four concentrations (DM basis): 0 (0FM, n = 8), d of feeding dietary treatments), cows were injected
2.6 (2.6FM, n = 7), 5.2 (5.2FM, n = 8), or 7.8% (7.8FM, intravenously with 6 ml of a solution containing 0.5
n = 9). Rations were calculated to contain similar con- mg/ml estradiol-17β [saline/ethanol (50:50); Sigma, St.
centrations of CP and NEL. Ingredient and chemical Louis, MO] at 0900 h and with 100 IU of oxytocin
Journal of Dairy Science Vol. 85, No. 4, 2002
REPRODUCTIVE AND PRODUCTION RESPONSES TO FISH MEAL 757
Table 1. Ingredient composition of experimental diets containing increasing concentrations of Menhaden
fish meal (FM) fed to lactating dairy cows.
Diet
Item 0%FM 2.6%FM 5.2%FM 7.8%FM
% of dietary DM
Corn silage 24.9 24.9 24.9 24.9
Alfalfa hay 10.0 10.0 10.0 10.0
Shelled corn, ground 27.7 28.8 30.3 31.5
Citrus pulp 9.7 9.7 9.7 9.6
Soybean meal 12.2 10.1 7.1 5.9
Fish meal 0.0 2.6 5.2 7.8
Fish oil 0.0 0.0 0.0 0.28
Whole cottonseeds 5.1 5.1 5.1 5.1
Dried distillers grains with solubles (ethanol) 5.2 4.0 2.8 0.0
Blood meal 0.4 0.0 0.0 0.0
Mineral mix1 4.8 4.8 4.8 4.8
1
Mineral and vitamin mix contained 26.4% CP, 1.74% fat, 10.15% Ca, 0.90% P, 3.1% Mg, 8.6% Na, 5.1%
K, 1.5% S, 4.1% Cl, 2231 mg/kg of Mn, 1698 mg/kg of Zn, 339 mg/kg of Fe, 512 mg/kg of Cu, 31 mg/kg of
Co, 26 mg/kg of I, 7.9 mg/kg of Se, 67,021 IU/kg of vitamin A, 19,845 IU/kg of vitamin D, and 357 IU/kg of
vitamin E (DM basis).
(Phoenix Pharmaceuticals, St. Joseph, MO) at 1300 product of PGF2α metabolism, was assayed directly
h. Using indwelling jugular catheters, blood samples using a polyethylene glycol RIA procedure described
were collected at 15-min intervals from 1 h before to by Eley et al. (1981) and Guilbault et al. (1984). Inter-
3 h after the oxytocin injection (−60, −45, −30, −15, 0, and intra-assay coefficients of variation were 13.9 and
15, 30, 45, 60, 75, 90, 105, 120, 135, 150, 165, and 14.6%, respectively.
180 min) and at 30-min intervals from 3 to 4 h after Blood (10 ml) was obtained daily from the day of hCG
oxytocin injection (210, 240 min) to monitor uterine injection (d 32) to d 61 for determination of plasma
secretion of PGF2α. Ethylenediaminetetraacetic acid concentration of progesterone. Blood samples also
was used as an anticoagulant. Plasma was separated were taken at the time of each hormone injection of
immediately by centrifugation (2600 × g, 30 min) at the estrous synchronization protocol (Figure 1). Blood
4°C and frozen at −20°C until analyzed. Plasma con- was collected from the coccygeal vessels into evacuated
centration of 13,14-dihydro 15-keto PGF2α (PGFM), a blood tubes containing EDTA as an anticoagulant
Table 2. Chemical composition of experimental diets containing increasing concentrations of Menhaden
fish meal fed to lactating dairy cows.
Item 0%FM 2.6%FM 5.2%FM 7.8%FM
CP, % of DM 18.4 18.6 18.9 19.1
RUP, % of DM1 6.3 6.8 7.4 8.1
NEL, Mcal/kg of DM2 1.73 1.72 1.73 1.76
NDF, % of DM 31.7 28.9 31.1 28.9
ADF, % of DM 20.4 18.6 19.8 18.5
Ether extract, % of DM 5.1 4.5 5.0 5.3
NSC, % of DM3 38.2 38.9 36.7 37.3
Ca, % of DM 1.07 1.25 1.31 1.59
P, % of DM 0.41 0.50 0.55 0.64
Mg, % of DM 0.37 0.37 0.36 0.36
K, % of DM 1.31 1.33 1.26 1.25
Na, % of DM 0.45 0.54 0.49 0.58
S, % of DM 0.19 0.23 0.21 0.25
Cl, % of DM 0.44 0.48 0.48 0.53
Fe, mg/kg of DM 258 272 310 360
Zn, mg/kg of DM 117 123 129 129
Mn, mg/kg of DM 81 77 79 101
Cu, mg/kg of DM 27 32 29 41
1
Calculated using ruminally undegradable protein values for individual feedstuffs as given by 1989 NRC.
2
Calculated by DHI Forage Testing Laboratory, Ithaca, NY.
3
Calculated as (% NSC = 100% − %CP − %NDF − % fat − % ash).
Journal of Dairy Science Vol. 85, No. 4, 2002
758 MATTOS ET AL.
Table 3. Fatty acid composition of experimental diets.
Diet
Fatty acid1 0%FM 2.6%FM 5.2%FM 7.8%FM
(g/100 g of fatty acids)
14:0 7.03 0.83 3.49 2.66
14:1 0.05 0.06 0.05 0.06
16:0 15.93 16.17 16.78 17.78
16:1 0.47 0.81 1.24 1.90
18:0 1.96 2.12 2.16 2.54
18:1(c9) 15.92 16.25 15.87 14.93
18:1(c11) 0.96 1.10 1.12 1.30
18:2(c9,c12) 29.12 27.19 26.81 24.22
18:3, n6 0.78 0.79 0.81 0.79
18:3, n3 4.05 4.10 3.33 3.87
20:1 0.20 0.16 0.14 0.12
20:5, n3 0.00 0.57 1.14 1.74
22:5 0.02 0.12 0.06 0.25
22:6, n3 0.00 0.46 0.92 1.40
1
Expressed as number of carbons:double bonds.
(10.5 mg, Monoject, Sherwood Medical, St. Louis, MO). trations. A real-time ultrasound scanner equipped
Samples were maintained in ice until plasma was sep- with a 7.5 MHz rectal probe was used (Aloka SSD 500
arated by centrifugation (2600 × g, 30 min) at 4°C V, Aloka Co. Ltd., Tokyo, Japan).
within 1 h of collection, and stored at −20°C until ana-
lyzed. Progesterone concentration in plasma was de- Milk Fatty Isolation and Analysis
termined using a single antibody radioimmunoassay
procedure (Knickerbocker et al., 1986a). Inter- and Milk samples for the determination of fat, protein,
intraassay coefficients of variation were 9.0 and urea nitrogen, and somatic cells (Southeast Dairy Lab,
9.9%, respectively. McDonough, GA) were collected at two consecutive
Ultrasound scanning of the ovaries was performed milkings on experimental d 49, 50, 61, and 62. Milk
at the time of each injection of the synchronization samples were collected during two consecutive milk-
protocol to determine the number and size of ovarian ings on each of the two last experimental days for
structures, and to assess the recruitment of a new analysis of fatty acids using GLC. Samples were refrig-
follicular wave. Cows also were scanned 3 d after the erated at 4°C and subsequently composited according
hCG injection to determine the occurrence of ovula- to milk yield recorded for each of the milkings. Milk
tion. One cow (2.6FM) did not respond to the synchro- samples were composited after being manually in-
nization protocol and was dropped from the statistical verted gently several times to resuspend components.
analyses of plasma PGFM and progesterone concen- Milk fat was extracted using a detergent solution con-
taining 3% triton X-100 (wt/vol) and 7% sodium hexa-
metaphosphate in distilled water. Methyl esters of the
fatty acids were formed using a methanolic sodium
methoxide solution (Christie, 1982). Analysis of the
methyl esters was performed on a gas-liquid chromato-
graph (Hewlett Packard 6890 Series with auto injec-
tor) fitted with a flame-ionization detector. Fatty acid
profile was determined by split injection (20:1) onto a
CP-Sil 88 fused silica capillary column (100 m × 0.25
mm, Chrompack, Raritan, NJ) using a programmed
temperature gradient method. The hydrogen carrier
gas pressure was constant, and the injector and detec-
tor temperatures were 255°C. Initial oven tempera-
Figure 1. Sequence of injections and collection of samples. Cows ture was 70°C. Following injection of sample, oven
were split into two groups, which initiated the estrous synchroniza- temperature was increased at 4°C/min to 175°C and
tion procedure at a 4-d interval. Experimental day of each procedure
is indicated for both groups. BS = Blood sample, US = ultrasound held for 3 min. Oven temperature was then raised at
scanning of the ovaries. 1°C/min to 185°C and held for 20 min. Oven tempera-
Journal of Dairy Science Vol. 85, No. 4, 2002
REPRODUCTIVE AND PRODUCTION RESPONSES TO FISH MEAL 759
ture was then increased at 3°C/min to 215°C, followed Data on DMI and milk fatty acids were analyzed
by an increase at 10°C/min to 255°C. That oven tem- using the split-plot approach for analysis of repeated
perature was held for 5 min until completion of data measures of the GLM procedure of SAS. Model in-
collection, after which oven temperature was returned cluded the effects of treatment, cow within treatment,
to 70°C. Individual fatty acids were identified by com- time, and time × treatment interaction. Cow within
parison of retention times to those of pure standards treatment was used as the error term to test the effect
(Matreya Inc., Pleasant Gap, PA). trans-Fatty acid of treatment. Differences were considered significant
profile was determined using a constant oven tempera- at the 5% level. Orthogonal contrasts were run in all
ture of 160°C (Molkentin and Precht, 1995) with the analyses to compare the treatment means and to test
same gas chromatograph and column mentioned for linear, quadratic, and cubic effects of increasing
above. A response correction factor for each fatty acid concentrations of FM in the diet. The GLM procedure
methyl ester was used to convert peak area percentage also was used to analyze length of the lifespan of the
to weight percentage. Correction factors were deter- CL. The model included the effect of treatment only.
mined by analyzing butter oil of a known fatty acid The length of the lifespan of the CL for a given cow
profile with certified values (CRM 164; European Com- was defined as the number of days between the hCG
munity Bureau of Reference, Brussels). This standard injection and the first day when plasma progesterone
does not include EPA and DHA, and correction factors concentration was 50% or less than that detected on
for this portion of the chromatogram were not used. the previous day.
However, the correction factor for the nearest certified
value near EPA and DHA was approximately 1 so little
correction was necessary. For trans 10, cis12-18:2 con- RESULTS AND DISCUSSION
jugated linoleic acid (CLA), the correction factor gen-
Increasing FM in the diet did not affect DMI (Table
erated for c9, t11-18:2 CLA was used. The correction
4). Estimated concentrations of fish oil in the diets
factors during a run are very similar for fatty acids
provided by FM and supplemental fish oil were 0, 0.21,
that elute in the same region.
0.42, and 0.91% (DM basis) for the 0FM, 2.6FM,
5.2FM, and 7.8FM diets, respectively. Estimated com-
Statistical Analyses bined intakes of EPA and DHA were 0, 12.8, 24.1,
Plasma concentrations of PGFM and progesterone, and 54.0 g/d from the 0FM, 2.6FM, 5.2FM, and 7.8FM
and milk variables such as production, fat, protein, diets, respectively. Reports indicate that supplemen-
and urea nitrogen, were analyzed using the repeated tation with fish oil at 2 to 3% of dietary DM but not
measures analysis of the Mixed Procedure of SAS (Lit- at 1% of dietary DM reduced feed intake in lactating
tell et al., 1996). This method applies methods based dairy cows (Donovan et al., 2000). Cant et al. (1997)
on the mixed model with special parametric structure observed a 10% decrease in DMI with the addition of
on the covariance matrices. The dataset was tested to fish oil at 2% of diet (DM basis). Concentrations of
determine the covariance structure that provided the fish oil in the present diets were likely insufficient to
best fit for the data. Covariance structures tested in- reduce feed intake.
cluded compound symmetry, autoregressive 1, and un- Milk production was unaffected by the experimental
structured. The model included effects of treatment, diets (Table 4). Milk protein concentrations tended to
cow within treatment, time, and treatment-by-time in- increase linearly (P = 0.07) with increasing content of
teraction. FM in the diet (Table 4). This effect was likely a result
Data on plasma PGFM concentrations also were an- of the increasing concentrations of RUP (Table 2) ob-
alyzed using the test of homogeneity of polynomial served with increasing concentrations of FM in the
regression of the general linear models (GLM) proce- diet. It is possible that the profile of essential AA pro-
dure of SAS. This approach determines whether indi- vided by the increasing concentrations of FM was more
vidual curves for each treatment provide a better fit balanced and resulted in increased milk protein con-
for the data than a single pooled curve (Wilcox et al., centrations. Previous reports of effects of FM on milk
1990). Once a significant effect was detected, orthogo- protein concentrations have shown inconsistent re-
nal contrasts (0FM vs. 2.6FM, 5.2FM, and 7.8FM; sults. Feeding FM at the same concentrations used in
2.6FM vs. 5.2FM and 7.8FM; 5.2FM vs. 7.8FM) were this study did not affect milk protein concentrations
run to detect differences between specific treatment (Spain et al., 1995). However, other studies have re-
curves. The model included effects of treatment, cow ported increased milk protein concentrations in re-
within treatment, time, and the interaction between sponse to the addition of FM to the diet (Calsamiglia
treatment and time. et al., 1995; Metcalf et al., 1994).
Journal of Dairy Science Vol. 85, No. 4, 2002
760 MATTOS ET AL.
Table 4. Effect of increasing intake of fish meal (FM) on lifespan of the induced corpus luteum (CL), DMI,
milk production, and milk composition of Holstein cows.
Diet
Measurement 0%FM 2.6%FM 5.2%FM 7.6%FM SE
Length of CL lifespan, d 15.7 16.8 14.8 15.6 0.6
DMI, kg/d 25.4 25.4 23.8 25.1 0.9
Milk, kg/d 40.1 40.8 35.1 40.2 2.2
Corrected milk, kg/d1 39.1 38.9 38.8 40.2 0.5
Milk fat, % 2.88 2.86 2.90 2.66 0.13
Milk protein, %† 3.10 3.22 3.24 3.30 0.07
4% FCM, kg/d 32.5 32.2 32.4 32.1 0.2
Milk fat, kg/d 1.16 1.17 1.02 1.07 0.06
Milk protein, kg/d 1.24 1.31 1.14 1.33 0.07
Milk urea nitrogen, mg% 15.8 15.4 14.9 14.7 0.8
1
Milk weights of each cow were corrected using the individual milk production averages recorded on the
week before the introduction of the experimental diets.
†P = 0.07.
Milk fat percentage was not affected by diet (Table nic acid (C18:3). Similar results were reported by Don-
4). Milk fat depression occurs often in cows fed FM or ovan et al. (2000) and Cant et al. (1997). The reduction
fish oil (Hussein and Jordan, 1991; Cant et al., 1997). in the proportion of n-6 fatty acids was due largely to
Feeding FM was reported to alter ruminal fermenta- a linear decrease (P < 0.001) in the concentration of
tion leading to reduced acetate: propionate ratio, linoleic acid (C18:2). Conversely, concentrations of ar-
which has been associated with reduced milk fat per- achidonic acid (C20:4, n-6) increased with increasing
centage. In another study, FM reduced milk fat per- concentrations of dietary FM and fish oil (P < 0.003).
centage even when infused postruminally, suggesting Dietary FM and fish oil decreased the proportion of
that FM causes reduced milk fat percentage through oleic acid and total n-9 fatty acids in a linear fashion
mechanisms unrelated to ruminal fermentation (Cal- (P < 0.003). Dietary FM and fish oil increased linearly
samiglia et al., 1995). Proposed mechanisms of postru- the concentrations of the cis-9 trans-11 CLA, a fatty
minal inhibition of milk fat content include inhibition acid that has been shown to have potent anticarcino-
of de novo fatty acid synthesis, inhibition of lipoprotein genic effects. Similar effects of dietary fish oil on milk
lipase in the mammary gland, and alteration of postru- fat concentrations of arachidonic acid, oleic acid, and
minal metabolism. Eicosapentaenoic acid and DHA, CLA have been reported (Cant et al., 1997; Donovan
present at relatively high concentrations in FM and et al., 2000).
oil, have been reported to be more potent inhibitors of Approximately 40% of milk fatty acids originate
lipogenesis than other PUFA (Jump et al., 1996), and from intestinal very low-density lipoproteins, as re-
therefore may partly explain the effects of these feed ported by Byers and Schelling (1988). Therefore, the
ingredients on milk fat percentages. Calsamiglia et al. profile of milk fatty acids likely reflects dietary fatty
(1995) also hypothesized that FM may suppress milk acids absorbed. Absorption of greater amounts of n-3
fat percentage only when the diet contains large fatty acids increased the proportion of these fatty acids
amounts of fat. Diets in the present experiment con- in milk, and it is likely that similar changes occurred
tained ether extract fractions (4.5 to 5.3%) that were in fatty acid profiles of peripheral tissues, including
similar and within a range that is not expected to the uterus.
affect ruminal fermentation. Increasing concentra- Cows fed FM had reduced plasma PGFM concentra-
tions of FM did not affect milk urea nitrogen nor 4% tions during the peak response (15, 30, and 45 min)
FCM yield (Table 4). after the challenge with estradiol and oxytocin when
Fish meal in the diet decreased (P < 0.01) the propor- compared to cows not fed FM (Figure 2). This result
tion of total unsaturated fatty acids in milk fat (Table was confirmed when four treatment curves were com-
5), specifically at concentrations ≥ 5.2% of diet DM pared using the test of homogeneity of regression (Fig-
(cubic effect P < 0.07). Among the unsaturated fatty ure 3). Contrast analyses indicated that the regression
acids, increasing intake of FM increased linearly the curve of plasma PGFM concentrations from cows fed
proportion of n-3 fatty acids (P < 0.001) and reduced the 0FM diet was elevated during the early sampling
linearly the proportion of n-6 fatty acids in milk fat period compared with animals fed any of the diets
(P < 0.001). Within n-3 fatty acids, increases were sig- containing FM (P < 0.025). Also the regression curve
nificant for EPA and DHA (P < 0.001) but not for linole- for plasma PGFM concentration of animals fed the
Journal of Dairy Science Vol. 85, No. 4, 2002
REPRODUCTIVE AND PRODUCTION RESPONSES TO FISH MEAL 761
Table 5. Milk fatty acid profile (% of total fatty acid in milk).
Diet Contrast
Fatty acid1 Control 2.6%FM 5.2%FM 7.8%FM SE TRT Linear Quad. Cubic
(g/100 g of fatty acids) (P)
C4 3.8 3.7 4.0 3.4 0.15 0.04 0.26 0.09 0.05
C6 2.4 2.3 2.5 2.2 0.13 0.53 0.54 0.69 0.22
C8 1.36 1.36 1.36 1.27 0.07 0.78 0.43 0.56 0.78
C10 3.1 3.0 3.0 2.9 0.18 0.91 0.47 0.95 0.95
C12 3.5 3.5 3.4 3.4 0.20 0.97 0.63 0.97 0.93
C14 9.0 10.2 11.2 9.2 0.90 0.33 0.70 0.10 0.51
C14:1 0.87 0.94 0.90 0.88 0.06 0.84 0.97 0.42 0.66
C16:0 31.4 32.7 35.0 31.2 1.77 0.41 0.84 0.16 0.39
C16:1 1.22 1.45 1.55 1.79 0.02 0.19 0.03 0.98 0.74
C18:0 10.2 9.3 8.1 8.5 0.91 0.39 0.13 0.49 0.69
Total C18:12 21.2 20.8 17.6 19.3 0.95 0.04 0.04 0.28 0.08
C18:1, n9 16.3 15.6 12.4 13.0 0.77 < 0.01 < 0.01 0.44 0.09
C18:2, n6 7.1 6.1 5.1 5.0 0.32 < 0.01 < 0.01 0.22 0.50
CLA c9t11 0.71 0.78 0.92 1.1 0.09 0.055 < 0.01 0.71 0.90
CLA t10c12 0.012 0.013 0.006 0.030 0.03 < 0.01 < 0.01 < 0.01 < 0.01
C18:3, n3 0.33 0.25 0.28 0.28 0.32 0.37 0.36 0.24 0.33
C20:1 n9 0.17 0.20 0.21 0.27 0.03 0.08 < 0.01 0.70 0.59
C20:4, n6 0.007 0.019 0.038 0.042 0.007 < 0.01 < 0.01 0.51 0.49
C20:5, n3 0.023 0.059 0.094 0.173 0.002 < 0.01 < 0.01 < 0.01 0.05
C22:5, n3 0.046 0.095 0.122 0.164 0.01 < 0.01 < 0.01 0.67 0.41
C22:6, n3 0.05 0.12 0.22 0.26 0.024 < 0.01 < 0.01 0.55 0.38
Total n33 0.45 0.52 0.73 0.88 0.057 < 0.01 < 0.01 0.50 0.52
Total n64 7.12 6.20 5.19 5.10 0.32 < 0.01 < 0.01 0.21 0.49
Total n95 16.9 16.3 13.0 13.7 0.80 < 0.01 < 0.01 0.44 0.09
Saturated 64.8 66.0 68.6 62.1 1.85 0.12 0.51 0.05 0.23
Unsaturated 30.9 30.0 25.9 27.9 1.05 < 0.01 0.01 0.18 0.07
Other 4.3 4.0 5.5 10.0 1.23 < 0.01 < 0.01 0.07 0.81
1
Expressed as number of carbons:double bonds.
2
Total C18:1 = trans-14 + trans-15 + trans-6 + trans-8 + trans-9 + trans-10 + trans-11 + trans-12 + trans-
16 + cis-9 + cis-11 + cis-12 + cis-13 + cis-15.
3
n3 fatty acids: C18:3 + C20:5 + C22:5 + C22:6.
4
n6 fatty acids: C18:2 + C20:2 + C20:4.
5
n9 fatty acids: trans-9 C18:1 + cis-9 C18:1 + C20:1.
Figure 3. Regression curves of the prostaglandin F2α metabolite
(PGFM) responses of cows fed diets containing 0 ( ), 2.6 ( ), 5.2 ( )
Figure 2. Response of prostaglandin F2α metabolite (PGFM) of and 7.8% (◆) menhaden fish meal (FM) to injections of estradiol and
cows fed or not fed Menhaden fish meal (0% fish meal vs. [2.6%, 5.2%, oxytocin given on d 15 of a synchronized estrous cycle. Response of
plus 7.8% fish meal] dietary DM basis) to sequential injections of cows fed 0% FM was higher than that of cows fed fish meal (0% vs.
estradiol and oxytocin given on d 15 of a synchronized estrous cycle. 2.6%, 5.2%, 7.8%; SE = 3.6, P < 0.025). Response of cows fed a diet
The PGFM response was lower in cows fed fish meal at 15, 30, and of 5.2% FM was greater than that of cows fed a diet of 7.8% FM (5.2%
45 minutes after injection of oxytocin (*, P < 0.05). vs. 7.8% FM, P < 0.05).
Journal of Dairy Science Vol. 85, No. 4, 2002
762 MATTOS ET AL.
5.2FM diet was significantly higher than that of ani-
mals in the 7.8FM group (P < 0.05).
These results are in agreement with the findings of
Thatcher et al. (1997) in which cows fed diets con-
taining 5% FM (DM basis) had lower plasma PGFM
concentrations when induced to secrete PGF2α with
estradiol-17β and oxytocin injections than cows not
fed FM. However, responses observed in the present
study did not follow a typical dose response pattern,
as increasing intake of FM did not result in decreasing
responses of PGFM to estradiol and oxytocin injec-
tions. The PGFM response observed in the 5.2FM
group was greater than that observed in the other
FM groups. This effect could be explained by the high Figure 4. Profiles of plasma progesterone concentration of cows
fed diets containing 0 ( ), 2.6 ( ), 5.2 ( ), and 7.8% (◆) menhaden
variability of responses detected in two cows of the fish meal during d 0 to 15 of the synchronized estrous cycle (SE =
5.2FM group. The fact that EPA and DHA are strong 1.2, P > 0.05).
inhibitors of prostanoid synthesis and their relatively
high concentrations in FM suggest that they are the
active components reducing the secretion of PGF2α.
The protocol used to induce secretion of PGF2α from
The mechanism by which EPA and DHA inhibit se-
the uterus included sequential injections of estradiol-
cretion of uterine prostanoids is not understood fully,
17β and oxytocin. Estradiol-17β increases uterine
but requires incorporation of EPA and DHA into cellu- blood flow and induces acute secretion of PGF2α that
lar lipid pools and may involve competition with ara- peaks approximately 5.5 h after injection (Knicker-
chidonic acid for processing by the cyclooxygenase -1 bocker et al., 1986b). This effect is probably mediated
and -2 enzymes (COX-1 and COX-2), and inhibition of by induction of oxytocin receptors and increased re-
synthesis and activity of the cyclooxygenases. Feeding sponsiveness to oxytocin. Concentrations of COX-2 in
FM increased the proportion of EPA and DHA in endo- sheep uterus primed with progesterone were increased
metrial lipids of beef cows (Burns et al., 2000), indicat- by treatment with estradiol (Charpigny et al., 1997),
ing that dietary changes can alter the fatty acid compo- and expression of COX-2 in rat endometrium peaked
sition of the uterus. Recent evidence from studies us- during diestrus and coincided with peak plasma estra-
ing cultured bovine endometrial cells demonstrated diol concentrations, suggesting estradiol-dependent
that EPA and DHA inhibited synthesis of PGF2α in a induction of COX-2 (Shoda et al., 1995). Oxytocin stim-
potent, dose-dependent manner (Mattos et al., 2001). ulated secretion of PGF2α from uterine tissue when
Processing of EPA and arachidonic acid by the administered in vivo (Lafrance and Goff, 1985). Eicosa-
cyclooxygenase enzymes leads to the synthesis of pentaenoic acid reduced the efficiency of cyclooxygen-
prostanoids of the 3 and 2 series, respectively. Because ase catalysis (Kulmacz et al., 1994). It is possible that
both EPA and arachidonic acid are substrates for the induction of COX-2 expression by estradiol may have
cyclooxygenases, increased availability of dietary EPA antagonized the inhibitory effect of EPA and DHA and
could displace arachidonic acid, leading to increased made potential differences between the control and
synthesis of prostanoids of the 3 series at the expense FM diets less apparent.
of prostanoids of the 2 series, such as PGF2α. Prostan- No effect of diet was observed on the profiles of
oids of the 3 series are less bioactive (Needleman et plasma progesterone concentrations measured from
al., 1979), and there appears to be no evidence for their d1 to d15, as determined by comparison of the regres-
role in ruminant luteolysis. Besides this competitive sion curves (P > 0.05, Figure 4). The rate of increase
mechanism of EPA, EPA and DHA may reduce the in plasma progesterone concentrations from d1 to d15
expression of the cyclooxygenase genes (Achard et al., was 0.88 ng/ml per day. In previous studies, supple-
1997), which could make the cyclooxygenases less menting cattle with dietary fat resulted in increased
available and reduce prostanoid synthesis. Also, plasma concentrations of cholesterol (Grummer and
cyclooxygenase converts EPA into prostanoids of the Carroll, 1991) and progesterone (Staples et al., 1998).
3 series in a less efficient manner than it converts Replacement of distillers grains with FM and fish oil
arachidonic acid into prostanoids of the 2 series. A did not markedly increase the ether extract content of
lower efficiency of catalysis may result in reduced total the diet. This may partly explain why plasma proges-
prostanoid synthesis. terone concentrations were not affected.
Journal of Dairy Science Vol. 85, No. 4, 2002
REPRODUCTIVE AND PRODUCTION RESPONSES TO FISH MEAL 763
The length of the luteal phase (15.7 ± 1.9 d) of the Baguma-Nibasheka, M., J. T. Brenna, and P. W. Nathanielsz. 1999.
Delay of pre-term delivery in sheep by n-3 long chain polyunsatu-
induced estrous cycle was not affected (P > 0.05) by rates. Biol. Reprod. 60:698–701.
treatment (Table 4). Because secretion of PGF2α was Burke, J. M., C. R. Staples, C. A. Risco, R. L. De la Sota, and
induced with estradiol and oxytocin, these results may W. W. Thatcher. 1997. Effect of ruminant grade menhaden fish
meal on reproductive and productive performance of lactating
not reflect what would occur if potential PGF2α secre- dairy cows. J. Dairy Sci. 80:3386–3398.
tion was spontaneous. The findings indicate that the Burns, P. D., D. B. Abbey, T. R. Bonnette, M. A. Harris, and J. C.
ability of the CL to undergo regression after exposure Whittier. 2000. Effects of fishmeal supplementation on bovine
endometrial concentrations of n-3 fatty acids in Proceedings,
to high concentrations of PGF2α was not affected. The
Western Section, American Society of Animal Science.
rate of decline of plasma progesterone concentrations Byers, F. M., and G. T. Schelling. 1988. Lipids in Ruminant Nutri-
following onset of CL regression was not different tion. Pages 298–312 in The Ruminant Animal Digestive Physiol-
among treatments (P > 0.05). Feeding high concentra- ogy and Nutrition. D.C. Church, ed. Prentice-Hall, Inc., New
Jersey, USA.
tions of fat has been reported to reduce the rate of Calsamiglia, S., G. Caja, M. D. Stern, B. A. Crooker. 1995. Effects
clearance of progesterone (Hawkins et al., 1995). How- of ruminal versus duodenal dosing of fish meal on ruminal fer-
ever, fat content among the experimental diets was mentation and milk composition. J. Dairy Sci. 78:1999–2007.
Cant, J. P. A., A. H. Freeden, T. MacIntyre, J. Gunn, and N. Crowe.
similar and likely insufficient to reduce the rate of 1997. Effect of fish oil and monensin on milk fat composition in
progesterone clearance. dairy cows. Can. J. Anim. Sci. 77:125–131.
Charpigny, G., P. Reinaud, J. P. Tamby, C. Creminon, J. Martal,
J. Maclouf, and M. Guillomot. 1997. Expression of cyclooxygen-
CONCLUSIONS ase-1 and −2 in ovine endometrium during the estrous cycle and
early pregnancy. Endocrinology 138:2163–2171.
Feeding fish meal and fish oil decreased the uterine Christie, W. W. 1982. A simple procedure for rapid transmethylation
secretion of PGF2α induced by injection of estradiol of glycerolipids and cholesteryl esters. J. Lipid Res. 23:2344–
2349.
and oxytocin on d 15 of an induced estrous cycle. This Donovan, D. C., D. J. Schingoethe, R. J. Baer, J. Ryall, A. R. Hippen,
effect likely was due to the inhibitory effects of eicosa- and S. T. Franklin. 2000. Influence of dietary fish oil on conju-
pentaenoic acid and docosahexaenoic acid on synthesis gated linoleic acid and other fatty acids in milk fat from lactating
dairy cows. J. Dairy Sci. 83:2620–2628.
of PGF2α. Plasma concentrations of progesterone from Eley, D. S., W. W. Thatcher, H. H. Head, R. J. Collier, C. J. Wilcox,
d 0 to 15 of an induced estrous cycle were unaffected and E. P. Call. 1981. Periparturient and postpartum endocrine
by diet. Milk fat from cows fed fish meal and fish oil changes of conceptus and maternal units in Jersey cows bred
for milk yield. J. Dairy Sci. 64:312–320.
had increased proportions of n-3 fatty acids and of cis- Grummer, R. R., and D. J. Carroll. 1991. Effects of dietary fat on
9, trans-11 C18:2 and trans-10, cis-12 C18:2 isomers. metabolic disorders and reproductive performance of dairy cat-
tle. J. Anim. Sci. 69:3838–3852.
Guilbault, L. A., W. W. Thatcher, M. Drost, and S. M. Hopkins.
ACKNOWLEDGMENTS 1984. Source of F series prostaglandins during the early post-
partum period in cattle. Biol. Reprod. 31:879–887.
Fish meal and oil used in this study were kindly Hawkins, D. E., K. D. Niswender, G. M. Oss, C. L. Moeller, K. G.
donated by Omega Protein (Reedville, VA). Authors Odde, H. R. Sawyer, and G. D. Niswender. 1995. An increase
in serum lipids increases luteal lipid content and alters the
would like to thank the staff of the University of Flor- disappearance rate of progesterone in cows J. Anim. Sci.
ida Dairy Research Unit for preparing the concen- 73:541–545.
trates and managing the experimental cows, and Joce- Hussein, H. S., and R. M. Jordan. 1991. Fishmeal as a protein
supplement in ruminant diets: A review. J. Anim. Sci.
lyn Jennings for her assistance in processing feed and 69:2147–2156.
milk samples. Authors also express appreciation to Jones, D. F., W. P. Weiss, D. L. Palmquist, and T. C. Jenkins. 1998.
Richard Miles for his advice about incorporation of fish Dietary fish oil effects on milk fatty acid composition. J. Dairy
oil into the diet. This study received financial support Sci. 81(Suppl. 1):232. (Abstr.)
Jump, D. B., S. D. Clarke, A. Thelen, M. Liimatta, B. Ren, and
from the NRI Competitive Grant Program/USDA, M. Badin. 1996. Dietary polyunsaturated fatty acid regulation
grant no. 98-35203-6367; Fundacao CAPES—Coorde-
¸˜ of gene transcription. Prog. Lipid Res. 35:227–241.
nacao de Aperfeicoamento de Pessoal de Nıvel Supe-
¸˜ ¸ ´ Knickerbocker, J. J., W. W. Thatcher, F. W. Bazer, M. Drost,
D. H. Barron, K. B. Fincher, and R. M. Roberts. 1986a. Proteins
rior, Brazil; and Omega Protein. This is Florida Ag- secreted by day 16 to 18 conceptuses extend corpus luteum func-
ricultural Experimental Station Journal Series No. tion in cows. J. Reprod. Fertil. 77:381–391.
R-08470. Knickerbocker, J. J., W. W. Thatcher, D. B. Foster, D. Wolfenson,
F. F. Bartol, and D. Caton. 1986b. Uterine prostaglandin and
blood flow responses to estradiol-17 beta in cyclic cattle. Prosta-
REFERENCES glandins 31:757–776.
Kulmacz, R. J., R. B. Pendleton, and W. E. Lands. 1994. Interaction
Achard, D., M. Gilbert, C. Benistant, S. B. Slama, D. L. DeWitt, between peroxidase and cyclooxygenase activities in prostaglan-
W. L. Smith, and M. Lagarde. 1997. Eicosapentaenoic and doco- din-endoperoxide synthase. J. Biol. Chem. 269:5527–5536.
sahexaenoic acids reduce PGH synthase 1 expression in bovine Lafrance, M., and A. K. Goff. 1985. Effect of pregnancy on oxytocin-
aortic endothelial cells. Biochem. Biophys. Res. Comm. induced release of prostaglandin F2α release in heifers. Biol.
241:513–518. Reprod. 33:1113–1119.
Journal of Dairy Science Vol. 85, No. 4, 2002
764 MATTOS ET AL.
Levine, L., and N. Worth. 1984. Eicosapentaenoic acid: Its effects controlled trial of effect of fish-oil supplementation on pregnancy
on arachidonic acid metabolism by cells in culture. J. Allergy duration. Lancet 25:1003–1007.
Clin. Immun. 74:430–436. SAS. SAS/STAT User’s Guide (Release 6.03). 1988. SAS Institute
Littel, R. C., G. A. Milliken, W. W. Stroup, and R. D. Wolfinger, SAS Inc., Cary, NC.
for Mixed Models. 1996. SAS Institute Inc., Cary, NC. 633 pp. Shoda, T., K. Hatanaka, M. Saito, M. Majima, M. Ogino, Y. Harada,
Mattos, R., C. R. Staples, and W. W. Thatcher. 2000. Effects of M. Nishijima, M. Katori, and S. Yamamoto. 1995. Induction of
dietary fatty acids on reproduction in ruminants. Rev. Reprod. cyclooxygenase type-2 (COX-2) in rat endometrium at the peak
5:38–45. of serum estradiol during the estrus cycle. Jpn. J. Pharmacol.
Mattos, R., A. Guzeloglu, and W. W. Thatcher. 2001. Effect of polyun- 63:289–291.
saturated fatty acids on secretion of PGF2α from bovine endome- Spain, J. N., C. E. Polan and B. A. Watkins. 1995. Evaluating effects
trial (BEND) cells stimulated with phorbol 12, 13 dibutyrate of fish meal on milk fat yield of dairy cows. J. Dairy Sci.
(PDBu). Theriogenology 55(1):326. (Abstr.) 78:1142–1153.
Metcalf, J. A., D. E. Beever, J. D. Sutton, D. Wray-Cahen, R. T. Staples, C. R., J. M. Burke, and W. W. Thatcher. 1998. Influence
Evans, D. J. Humphries, F. R. C. Backwell, B. J. Bequette, and of supplemental fats on reproductive tissues and performance
of lactating cows J. Dairy Sci. 81:856–871.
J. C. MacRae. 1994. The effect of supplementary protein on in
Thatcher, W. W., C. R. Staples, G. Danet-Desnoyers, B. Oldick, and
vivo metabolism of the mammary gland in lactating dairy cows.
E. P. Schmitt. 1994. Embryo health and mortality in sheep
J. Dairy. Sci. 77:1816–1827. and cattle J. Anim. Sci. 72(Suppl. 3):16. (Abstr.)
Molkentin, J., and D. Precht. 1995. Optimized analysis of trans- Thatcher, W. W., M. Binelli, J. M. Burke, C. R. Staples, J. D. Am-
octadecenoic acids in edible fats. Chromatographia 41:267–272. brose, and S. Coelho. 1997. Antiluteolytic signals between con-
National Research Council. 1989. Nutrient Requirements of Dairy ceptus and endometrium. Theriogenology 47:131–140.
Cattle. 6th rev. ed. Natl. Acad. Sci., Washington, DC. Wilcox, C. J., W. W. Thatcher, and F. G. Martin. 1990. Statistical
Needleman, P., A. Raz, M. S. Minkes, J. A. Ferrendeli, and H. analysis of repeated measurements in physiological experiments
Sprecher. 1979. Triene prostaglandins: prostacyclin and throm- pp 141–145 in Proc. Final Research Co-ordination Meeting of
boxane biosynthesis and unique biological properties Proc. Natl. the FAO/IAEA/ARACAL III Regional Network for Improving
Acad. Sci. 76:944–948. the Reproductive Management of Meat- and Milk-Producing
Olsen, S. F., J. D. Sorensen, N. J. Secher, M. Hedegaard, T. B. Livestock in Latin America with the Aid of Radioimmunoassay.
Henriksen, H. S. Hansen, and A. Grant A. 1992. Randomised International Atomic Agency, Vienna.
Journal of Dairy Science Vol. 85, No. 4, 2002
