CASE STUDY: Effects of Supplemental Linoleic and Linolenic Acids on Reproduction in Holstein Cows

INTRODUCTION

Linoleic acid is the major fatty acid in most seed lipids such as soybean and corn meal, whereas linolenic acid is the primary PUFA in forage lipids. Staples et al. (1998) noted that about 60 to 90% of PUFA in crushed rapeseed were biohydrogenated in the rumen before reaching the small intestine for absorption. One method of overcoming microbial biohydrogenation in the rumen is to transport EFA to the small intestine for delivery to the tissues for metabolism by feeding ruminally inert fats.

Essential fatty acids must be supplied to the cow through the diet. These EFA have several roles in the biological function of the cow. First, they are broadly utilized for normal cellular functions, such as the synthesis of phospholipid membranes. second, because of their high energy, they are thought to help alleviate the effects of negative energy balance (NEB; Sklan et al., 1989; Domecq et al., 1997). Decreasing the effects of NEB could lead to an earlier return to estrus postpartum, thus increasing fertility (Domecq et al., 1997; Moreira et al., 2001). Third, the EFA linoleic and linolenic acids are precursors to eicosanoids and steroid hormones, and also may aid in the stimulation or inhibition of these reproductive requirements (Mattos et al., 2000). Thus, EFA play an important role in reproduction.

Staples et al. (1998) stated that "11 studies reported an improvement (P < 0.10) either in first AI service conception rate or in the overall rate of conception or pregnancy," as a result of feeding supplemental EFA. Five of those studies fed calcium salts of long-chain fatty acids (Ca-LCFA), and in those 5 studies, cows fed CaLCFA had a 5.8-percentage point increase in standing estrus signs, 5.9-percentage point increase ovarian activity, and 12-percentage unit more cows responded to a 20 mg dose of prostaglandin F^sub 2a^ (PGF^sub 2a^; normal dose is 25 mg).

The purpose of this study was to obtain objective reproductive data in cows fed either Ca-LCFA or CaLCFA+EFA to ascertain if the extra linoleic and linolenic acid improved reproductive performance.

MATERIALS AND METHODS

Treatments and Cow Randomization 

This trial was approved by the Institutional Animal Care and Use Committee (University of Arizona, Tucson) before its commencement. There were 2 study groups, CaLCFA (Megalac, n = 1,312) and Ca-LCFA+EFA (Megalac-R, n = 708). Both supplements are rumen bypass fats produced by Church and Dwight Co. Inc. (Princeton, NJ). The Megalac is a high energy feed supplement of ruminally inert Ca-LCFA and has a NE^sub 1^ value of 6.5 Meal/kg. The Megalac-R maintains the same energetic and rumen-inert properties of regular Megalac; however, it is devised to transport higher concentrations of linoleic and linolenic acids to the small intestine for absorption and use in tissue metabolism. The Megalac-R contains approximately 4.5 times as much linoleic acid (31.75% of product) and more than 23 times as much linolenic acid (4.7% of product) as Megalac.

Holstein cows that were not receiving recombinant bovine somatotropin were used for this trial and were balanced by parity and milk production in their previous lactation. Cow data from the previous lactation were divided into groups by lactation number (1, 2, 3, and 4 or greater), and then by production based on the Arizona state average for the previous year (305-d ME < 9,545 kg or 305-d ME >9,545 kg). First lactation heifers were balanced for production using sire-predicted transmitting ability values for milk. Cows were then randomly sorted into one of the treatment groups. Cows were milked 4 times/d for 60 d postpartum and 2 times/d until dryoff. Cows were fed TMR twice daily, and the rations were balanced for caloric intakes between treatments by the herd nutritionist (CMW Nutrition, Corona, CA). Net energy for lactation for both treatment groups was 1.67 Meal/kg. Based on the recommendations of the herd nutritionist, all cows were fed 0.114 kg/d Ca-LCFA prepartum and 0.159 kg/d of either Ca-LCFA or CaLCFA + EFA from parturition until 150 DIM.

A subset of cows in each treatment was used for collection of milk so progesterone could be assayed twice weekly. Primiparous (Ca-LCFA, n = 5 and Ca-LCFA+EFA, n = 5) and multiparous (Ca-LCFA, n = 18 and Ca-LCFA+EFA, n = 17) cows were used for milk sample collection twice weekly from calving to 150 DIM. Cows in pens fed one of the unique rations had leg bands according to treatment. These bands were used for immediate identification of trial cows when they came into the parlor, and also to identify cows in the wrong pen. Milk samples from the banded cows were collected into 10-mL plastic vials with snap tops (Arizona DHIA, Tempe, AZ). Samples were obtained by forestripping 1 to 5 mL of milk into the vial and were frozen in an upright freezer at -20°C. Notes were kept on whether cows were in the wrong pen, if blood was in the sample, or if the milk appeared abnormal.

The voluntary waiting period (VWP) of the commercial herd was 60 DIM. A list of cows between 50 and 56 DIM was generated from DHI-Plus software (DHI, Provo, UT). This list was used to initiate the Ovsynch regimen (Pursley et al., 1997) on d 0 with a GnRH injection, and was kept for use during subsequent injections. On d 7 prostaglandin was administered. The second GnRH injection was given on d 9, and timed AI followed on d 10 by the herdsman. Timed AI 24 h after the last GnRH is a deviation from the traditional timed AI regimen and is not optimal for conception rates. However, this timing deviation fit best into the regular AI schedule of this large herd.

It was decided to compensate the cooperating dairy for the use of their cows, facilities, labor, and so on by feeding almost twice as many cows the Ca-LCFA diet, as it was being provided for free. This accounts for the population differences between treatments.

Ultrasound

A subset of cows in each treatment was subjected to real-time ultrasonography. Ultrasound was conducted by a trained technician using a SonoVet 2000 instrument (Medison, Cypress, CA). All ovarian structures (follicle numbers, class size, corpora lutea, and corpora hemorrhagicum) were recorded manually (Ginther, 1998) and were later entered into a log. Follicles were recorded as class 1 (1 to 5 mm), class 2 (6 to 10 mm), class 3 (11 to 15 mm), or class 4 (>16mm). Presence of corpus luteum, corpus hemorrhagicum, or both indicated an ovulation. Notes were made if the cow had any abnormalities such as metritis, fever, foot rot, or bad uterine position. Ultrasound was conducted every other week in May, June, December, and January. Initially, ultrasound was conducted on cows twice (about 14 and 30 DIM; Ca-LCFA, n = 9 and Ca-LCFA+EFA, n = 20). When no differences were detected between treatments in the first 2 wk postpartum, the focus became from 21 to 30 DIM. The subset population was 120 cows (Ca-LCFA, n = 63 and CaLCFA+EFA, n = 57).

Progesterone Assay

Progesterone concentrations were measured by enzyme immunoassay (EIA) of whole milk samples. Progesterone EIA were performed using solutions, plates, and directions supplied in a kit (Assay Designs, Ann Arbor, MI). An estrous cycle was counted as such if progesterone concentrations declined below 4 ng/ mL, followed by a rise above 4 ng/ mL. The 4 ng/mL was predetermined as the value at which a corpus Iuteum was fully functioning, because progesterone concentrations in milk are generally double those found in serum (Schams and Karg, 1986). The interassay CV was 9.7%, calculated between slopes of standard curves from 9 assay days, and the intra-assay CV was 4.9% between duplicate milk samples on 22 plates. Singlet samples were assayed to conserve costs of conducting the EIA. A doublet milk sample (n = 22) was analyzed on each plate to calculate the intraassay CV.

Two different extraction procedures were used to validate the assay by ensuring no components of the whole milk sample were competing with the assay antibodies.

Ether Extraction. Following the extraction method of Peralta (1982), 100 µL of standards or 75 µL of assay buffer combined with 25 µL of whole milk was placed in glass tubes and labeled. In a fume hood, 250 µL of diethyl ether (VWR, West Chester, PA) was added to the tubes, which were stoppered and shaken. The layers were allowed to separate, and the top ether layer was pipetted off into a clean glass tube. The ether step was repeated, and the second pipetted ether layer was combined with the first in the corresponding tube. The ether was evaporated to dryness by placing the tubes in a warm water bath (about 70°C). The extracted progesterone was then dissolved in 250 µL of assay buffer by vortexing thoroughly no fewer than 3 times over 5 min. The reconstituted samples were assayed immediately, using 100 µL of sample in each well. In a small subset of samples representing follicular luteal phases, progesterone concentrations were similar between whole milk and the ether extracts. This validated the antibody could be used with whole milk.

Fumed Silica Extraction. A 1:50 solution of fumed silica (Sigma-Aldrich, St. Louis, MO) was made and was added to tubes containing 75 µL of assay buffer and 25 µL of whole milk. These were swirled gently to mix and then centrifuged for 20 min at 2,000 × g. The delipidated sample was pipetted out by puncturing through the fat layer. This was assayed immediately using 100 µL of delipidated sample in each well. Removal of lipids led to assay values not being different than background readings on the plate reader.

Uterine Health Data

Uterine health data were collected through the event analysis function of DHI-Plus records (DHI-Provo) for prostaglandin treatment(s). The herd veterinarian (Dairy Veterinary Services, Chandler, AZ) was blind to treatment and diagnosed and treated cows based upon rectal examination of the reproductive tract. Abnormalities (uterine tone, endometritis) were recorded and treated with prostaglandin. At the conclusion of the experiment, the herd veterinarian discussed each individual cow treatment record and categorized abnormalities and treatments accordingly.

Statistical Analysis

Pen was the experimental unit, although the pens were dynamic because of the flow of cows. Ultrasound, synchronization, and uterine health data were analyzed using Pearson's chi-square (Moore and McCabe, 1999). Estrous cycle data obtained through progesterone profiles were analyzed using the PROG-MIXED procedure of SAS (SAS Inst. Inc., Gary, NC), with parity and milk production as covariates. The predetermined significance level was P = 0.05.

RESULTS AND DISCUSSION

A difference was detected (P < 0.005) for percentage of ovulations between treatments by 30 DIM (Table 1). No difference was found between treatments for numbers or sizes of follicles at either 14 or 30 DIM, as analyzed by chi-square, although a trend of Ca-LCFA+EFA supplemented cows having more follicles was noticed for number of class 1 follicles 15 to 30 DIM (P < 0.07) and number of class 2 follicles 15 to 30 DIM (P < 0.09; Table 2). Numbers of cows between Tables 1 and 2 differ because of a set of cows that were ultrasounded twice-once before 14 DIM and again between 15 and 30 DIM (Ca-LCFA, n = 9 and Ca-LCFA+EFA, n = 20). This subset of cows was ultrasounded twice to determine value of ovarian ultrasound less than 14 DIM.

A significant difference between treatments was detected between numbers of estrous cycles by 60 DIM in multiparous cows. The Ca-LCFA-fed cows (n = 18) had 2.0 estrous cycles/cow, as compared with 2.8 estrous cycles/cow for those fed CaLCFA+EFA (n = 17). However, no significant difference was detected between mean concentrations of progesterone, average lactation number, or average milk production in the covariate analyses. No significant differences were detected in primiparous cows in terms of number of cycles by 60 DIM. A significant effect of parity on progesterone concentrations regardless of treatment was detected between primiparous and multiparous cows.

Table 3 shows the differences between treatments for numbers of cows prescribed to be treated with prostaglandin by the herd veterinarian as a therapy for uterine abnormalities. Significant differences were detected between treatments for total number treated, as well as first and second case treatments. Additionally, a difference in the number of cysts was detected, although the overall frequency of cystic ovaries was less than 3% in both treatment groups (data not shown).

No significant differences were detected between treatments for first and second service conception rates, services per conception, or days open for the subset of cows subjected to the Ovsynch regimen. As anticipated, neither milk production nor components differed between treatments, because the trial was previously balanced for production. Milk production (3.5% FCM) averaged 37.0 and 37.1 kg/d during the trial period for Ca-LCFA and Ca-LCFA+EFA supplemented cows, respectively.

Excellent herd management is essential for profitability on a dairy. A cow's productivity depends upon proper management, particularly through the transition period. Several factors contribute to the success of a cow during this period, including body condition, energy balance, and stress (metabolic, environmental, or calving). All of these factors can be influenced by nutrition. It is vital for a cow to receive adequate energy, both for her subsequent lactation and for resumption of reproductive cyclicity (Domecq et al., 1997; Moreira et al, 2001). Thus, it has become a common practice on commercial dairies to supply energy in part through supplemental fat.

The results of this trial indicate cows supplemented with CaLCFA+EFA resumed cyclicity earlier postpartum, had more ovulatory cycles by the VWP, and were treated less often for uterine abnormalities and infections than cows supplemented with Ca-LCFA. A cow that ovulates sooner will generally have more cycles in a given period of time than a cohort that resumes cyclicity later. Additionally, a cow that has more estrual cycles has more opportunities to expel lochia and contaminants from her uterus, while also improving her rate of uterine involution, and therefore should have fewer uterine infections or abnormalities (Hafez, 1974; Senger, 2003). This improvement in reproductive health may also be due to the nutraceutical effects of group 2 eicosanoids (precursors of which are linoleic and linolenic acids). Thatcher et al. (2007) noted that group 2 prostaglandins are involved in proinflammatory immune responses, and PGF^sub 2a^ is involved in bacterial phagocytosis via neutrophil functions. Linoleic and linolenic acids are both known precursors to group 2 eicosanoids (Mattos et al., 2000), and feeding them likely benefits the transition health of the cow. This was confirmed by Cullens et al. (2004) in a study examining the effects of Megalac-R on periparturient health of Holstein cattle. Cows were fed one of 4 diets: control (no supplemental fat source), and Megalac-R fed starting at 28 d prepartum, at parturition, or 28 DIM. The study showed cows fed Megalac-R prepartum had significantly higher plasma concentrations of PGF metabolites than those fed at parturition, 28 DIM, or not at all. The conclusion is that feeding Megalac-R prepartum increased the lipid pool concentrations of linoleic and linolenic acids, which led to an increase in the substrate concentrations for arachidonic acid. Arachidonic acid is the primary precursor to PGF^sub 2a^ and other group 2 eicosanoids that would enhance the proinflammatory response in the cow postpartum. Fewer reproductive health problems means money saved for the producer in terms of medications, labor, and improved fertility (Erb et al., 1985).

No significant differences were found in terms of follicle population sizes and diameters between treatments on this trial, which is contrary to the studies summarized by Staples et al. (1998) in which ultrasonography was also used for follicular measurements. However, those studies (with the exception of Oldick et al., 1997) used zero dietary fat as a control, whereas this study used Ca-LCFA (Megalac) as a control. The linoleic and linolenic acid concentrations in this product, although in smaller concentrations than in the Megalac-R product, are still substantially higher than those found in the basal rations used in previous studies. This fact likely accounts for comparable follicle populations between treatments in this study.

Vasconcelos et al. (1999) reported improved pregnancy rates when Ovsynch was begun in cows midway through their estrous cycle. The Ca-LCFA cows had, on average, completed 2 ovulatory cycles and were in the beginning to midphase through their third cycle. The Ca-LCFA+EFA cows, however, had completed 2 ovulatory cycles and were in the late stages of their third cycle. This difference in phase of the estrous cycle at the initiation of Ovsynch could explain why no significant difference was detected between groups for conception rates, although all other points (earlier resumption of cyclicity, more cycles by VWP, and improved uterine health) would be expected to lead to increased fertility in the Ca-LCFA+EFA treated cows if they had not been hormonally programmed. Those results may have been observed if cows had not been scheduled for timed AI with Ovsynch. That was the design of this trial because of confounding results previously published in which fertility could not be directly compared because visual detection of estrus was the sole basis for determining when a cow should be inseminated (Scott et al., 1995; Sklan et al., 1989). Furthermore, the number of cycles a cow completes before breeding is positively correlated with conception rates (Morrow et al., 1969; Ax, 1991). Those findings were reported for cows not fed fat and not subjected to a hormonal management program.

The first 2 estrous cycles postpartum are usually accompanied by shorter luteal phases. Ideally, a cow should reach a normal luteal phase length before breeding, as discussed by Pursley et al. (1997). In a future trial, perhaps cows need a 75- or 80-d VWP in order to complete a normal-length cycle. Based on the findings in this trial, a recommendation could be made that herd managers feed Ca-LCFA+EFA and use an Ovsynch synchronization regimen initiating on 75 DIM.

In summary, cows supplemented with rumen-protected linoleic and linolenic acids (Megalac-R) resumed cyclicity earlier in the postpartum interval, had more estrous cycles by the VWP, and had fewer instances of metritis and other uterine abnormalities than cows fed lower concentrations of EFA (Megalac). Although no difference in fertility was apparent, all the positive reproductive outcomes attributed to higher EFA concentrations point to a biological scenario in which improved fertility would be expected in that population of cows.

IMPLICATIONS

The results indicated that the higher concentrations of linoleic and linolenic acids in Megalac-R contributed to improved uterine health and ovulation rates early postpartum compared with Megalac. Cows that resume cyclicity sooner are more fertile and have lower veterinary costs, both of which are financially beneficial to the producer.

ACKNOWLEDGMENTS

The authors would like to thank Nathan Goldman and the staff of Goldman Dairy (Coolidge, AZ) for the use of their dairy farm for this trial. Thanks also to Pfizer (New York) and Merial (Duluth, GA) for the hormone donations, and Arizona DHIA for the milk sample vials.

© 2008 American Registry of Professional Animal Scientists Provided by ProQuest LLC. All Rights Reserved.