If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The European livestock sector has a significant deficit of high-quality protein feed ingredients. Consequently there is interest in using locally grown protein grain crops to partially or completely replace imported protein feeds in dairy cow rations. Field bean (FB; Vicia faba) has been identified as a locally grown crop with significant potential. The current study was designed to examine the effects of FB on cow performance and nutrient utilization in the diet of early-lactation dairy cows, including high levels of FB (up to 8.4 kg/cow per day). The experiment used 72 dairy cows in a 3-treatment continuous design (from calving until wk 20 of lactation). All cows were given ad libitum access to a mixed ration comprising grass silage and concentrates [45:55 on a dry matter (DM) basis]. Concentrates offered contained either 0, 349, or 698 g of FB/kg of concentrate (treatments FB0, FB-Low, and FB-High, respectively), with FB completely replacing soybean meal, rapeseed meal, maize gluten, and wheat in the concentrate for the FB-High treatment. Following completion of the 20-wk experiment, ration digestibility, nutrient utilization, and methane (CH4) production were measured using 4 cows from each treatment. Neither silage DM intake, total DM intake, nor milk yield were affected by treatment. Cows on FB0 had a higher milk fat content than those on FB-High, and cows on FB0 and FB-Low had higher milk protein contents than did those on FB-High. Field bean inclusion increased the degree of saturation of milk fat produced. Milk fat yield, milk protein yield, and milk fat plus protein yield were higher with FB0 than with either FB-Low or FB-High. Treatment had no effect on the digestibility of DM, organic matter, nitrogen (N), gross energy, or neutral detergent fiber, whereas digestibility of acid detergent fiber was higher with FB0 than with FB-High. Neither the efficiency of gross energy or N utilization, nor any of the CH4 production parameters examined, were affected by treatment. Similarly, none of the fertility or health parameters examined were affected by treatment. The reduction in milk fat observed may have been due to the higher starch content of the FB-High diet, and the reduction in milk protein may have been due to a deficit of methionine in the diet. It is likely that these issues could be overcome by changes in ration formulation, thus allowing FB to be included at the higher range without loss in performance.
The increasing milk yield potential of dairy cows in many European countries has led to a requirement for more nutrient-dense diets. Increased nutrient density has often been achieved through the adoption of higher-concentrate feed levels, and this in turn has increased the demand for high-quality protein ingredients. However, European agriculture has a significant deficit of high-quality protein feedstuffs (
) and, as such, might appear to be ideal in ruminant diets. However, the inclusion of FB in ruminant diets is normally limited due to the perceived risk associated with antinutritional substances, including trypsin inhibitors, tannins, lectins, and protease inhibitors, which can reduce DM intake and animal performance (
), which has also contributed to the limited inclusion of FB in dairy cow diets.
The use of FB in dairy cow diets has been examined in several studies, with the findings sometimes conflicting. For example, although neither DM intake nor milk yield were affected by FB inclusion in studies by
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
) with no adverse effects on DM intake and milk production, but higher inclusion levels do not appear to have been examined previously. Given the high starch content of FB, which suggests that it could replace both conventional protein and energy components of the diet, the effects of higher FB inclusion levels need to be examined. In addition, as highlighted by
, information on the effects of FB inclusion in the diet of dairy cows in early lactation, the period that encompasses the breeding season, is limited at present. This is important, given concerns about the presence of estrogenic compounds in FB, although there is no evidence that dairy cow fertility is reduced when FB are offered. Consequently, the current study was designed to examine the effects on cow performance, nutrient utilization efficiency, and fertility when early-lactation dairy cows were offered diets containing higher levels of FB than examined previously. We hypothesized that these parameters would be unaffected by these higher levels of FB inclusion.
MATERIALS AND METHODS
This experiment was conducted at the Agri-Food and Biosciences Institute, Hillsborough, Northern Ireland (NI). All experimental procedures were conducted under an experimental license granted by the Department of Health, Social Services and Public Safety for NI, in accordance with the Animals (Scientific Procedures) Act 1986.
Animals and Housing
This continuous-design 20-wk experiment involved 72 Holstein-Friesian dairy cows, 33 primiparous and 39 multiparous; mean parity 2.07 (SD 1.19). During the 3-wk period before calving, cows had ad libitum access to grass silage, while a dry-cow mineral and vitamin mix, along with calcined magnesite (Trouw Nutrition, Cheshire, UK), were mixed with the silage to achieve a target intake of 100 g/cow per day and 50 g/cow per day, respectively. Cows were moved to a straw-bedded maternity pen approximately 24 to 48 h before calving, based on behavioral observations. Following calving (normally within 24 h), cows were transferred to an experimental free-stall barn with solid concrete floors, where they had access to cubicles fitted with rubber mats, which were bedded with sawdust twice daily. The floor was scraped every 3 h, using an automated slurry scraper system.
Treatments
The variety of FB (Vicia faba var. Fuego) used in the study was described by
as having “colored flower, high standing ability, high yield potential, big seeds.” Colored-flower varieties of FB have long been known to have higher tannin contents than white-flowered varieties (
recently recording the tannin content of the Fuego variety as 6.1 mg/g of DM. This variety was chosen for use in this study because it has been successfully grown by local arable and livestock farmers in NI over several years. The study by
also demonstrated that this variety could be successfully grown over wide areas of Europe, with this variety having been evaluated previously in a dairy cow study in Lithuania (
). The beans offered within this study were grown in NI. The crop was sown during spring 2016, harvested on September 28, 2016, and then dried at 80°C for approximately 2 h to achieve a moisture content of approximately 14%.
Cows were randomly allocated to 1 of 3 treatments at calving (FB0, FB-Low, FB-High), with primiparous and multiparous cows allocated separately. A check was made to ensure that each treatment group remained balanced for parity, BW, and BCS at drying-off, and, in the case of multiparous cows, for previous lactation 305-d milk yield and milk fat and protein contents.
Treatments differed in concentrate type offered (ingredient composition in Table 1). The concentrate offered with FB0 contained no FB, whereas concentrates offered with FB-Low and FB-High contained 349 and 698 g of FB per kilogram (fresh basis), respectively. The experimental concentrates (in the form of a meal) were mixed with grass silage (forage:concentrate ratio of 45:55 on a DM basis), together with 320 g/cow per day of straw and 50 g/cow per day of a rumen buffer (Acid Buf, Celtic Sea Minerals, Cork, Ireland; buffering capacity approximately 2.5× that of sodium bicarbonate), and offered in the form of a mixed ration. The grass silage component of the diet was produced from a primary growth (approximately 75% of cow feeding days) and a primary regrowth (approximately 25% of cow feeding days) of perennial ryegrass (Lolium perenne)–based swards. The rations were prepared and offered daily at approximately 0900 h (at proportionally 1.07 of the previous day's intake), with uneaten ration removed the following day at approximately 0800 h. Rations were prepared using a feeder wagon (Vari-Cut 12, Redrock, Armagh, NI). The total silage requirement for all 3 treatments was initially mixed for approximately 4 min and then deposited on a clean covered concrete yard. Silage for each individual treatment was then removed from this “pile” and returned to the feeder wagon. The appropriate quantities of concentrate, straw, and buffer were added to the mix, and mixing continued for another 6 min. The rations were then transferred from the wagon to a series of feedboxes mounted on weigh scales, with cows accessing feed in these boxes via an electronic identification system, enabling individual cow intakes to be recorded daily (Controlling and Recording Feed Intake, BioControl, Rakkestad, Norway). Cows had free access to fresh water at all times. In addition, all cows were offered 1.0 kg per day of soy hulls via an in-parlor feeder during milking (0.5 kg at each milking). Cows remained on the experiment until d 140 of lactation.
Table 1Ingredient composition (g/kg) of concentrates offered during the experiment
During the experiment cows were milked twice daily (between 0500 and 0700 h, and between 1500 and 1700 h) using a 50-point rotary milking parlor (BouMatic, Madison, WI), with milk yields recorded automatically at each milking and a total daily milk yield for each cow for each 24-h period calculated. Milk samples were taken during 2 consecutive milkings each week, a preservative tablet was added (Broad Spectrum Microtabs II, D and F Control Systems, Advanced Instruments Inc., Norwood, MA), and samples were stored at 4°C until analyzed. Samples were analyzed for fat, protein, and lactose using an infrared milk analyzer (Milkoscan model FT+; Foss Electric, Hillerod, Denmark), and a weighted concentration of each constituent was determined for the 24-h sampling period. A single milk sample was collected for progesterone analysis from each cow on 2 occasions each week (Monday and Thursday p.m.) until 60 d postpartum. Samples were preserved and stored as above, for a maximum of 3 wk, with progesterone concentrations subsequently determined using a competitive ELISA kit (Ridgeway Science Ltd., Gloucestershire, UK). In addition, at 10 and 20 wk post-calving (±1 wk) a milk sample was taken in proportion to yield during 2 successive milkings (a.m. and p.m.), the 2 samples bulked, and the bulked sample frozen at −20°C. These samples were subsequently analyzed for milk FA, as follows: milk fat was extracted from 1.0 mL of homogenized milk using a chloroform methanol extraction method (
), and FAME prepared. The FA composition was determined using GLC, with an aliquot (1.0 µL) of the FAME extract injected onto a CP Sil88 capillary column (100 m × 0.25-mm internal diameter × 0.2-µm film thickness) in a, Agilent 7800 gas chromatograph (both from Agilent Technologies, Santa Clara, CA), equipped with a temperature programmable injector operated in the split mode and a flame ionization detector. The oven was initially held at 50°C for 4 min, then increased at 8°C/min to 110°C, then at 5°C/min to 170°C (hold time 10 min), and finally at 2°C/min to 225°C (hold time 30 min). Fatty acids were identified by their retention time with reference to commercially available FA standards (37 Supelco FAME mix) and individual standards for those not in the mix (Sigma-Aldrich Co. Ltd., Gillingham, UK), and were quantified using C13 FAME as an internal standard.
Body weight was recorded twice daily (immediately after each milking) using an automated weighbridge, and a mean weekly BW for each cow was determined. The BCS of each cow was estimated fortnightly according to
. Blood samples were collected from the tail of each cow before feeding, at 2, 4, 6, 8, 10, and 14 wk (±3 d) post-calving. Samples were collected into evacuated tubes (Becton Dickinson, Oxford, UK), which were coated with either a clot activator or fluoride oxalate. The blood samples were centrifuged (1,800 × g at 17°C for 30 min) to obtain serum (tubes with clot activator) or plasma (fluoride oxalate tubes), which were separated and stored at −20°C until analyzed. Plasma was analyzed for glucose concentrations, whereas serum was analyzed for BHB, nonesterified fatty acids (NEFA), and urea concentrations. Plasma glucose concentrations were determined using the hexokinase method (Roche Diagnostics Ltd., Burgess Hill, UK), and serum biochemistry analysis was carried out on a Randox Imola chemistry analyzer system, using Randox reagent kits (Randox Laboratories, Crumlin, UK).
A clinical examination of vaginal mucus was conducted at 2, 4, and 6 wk postpartum, as an indicator of immune status. The methodology and scoring system used was as described by
Concentrate supplementation of a diet based on medium-quality grass silage for 4 weeks prepartum: Effects on cow performance, health, metabolic status, and immune function.
, with the combined results expressed on a scale of 0 to 5. In the current study, scores were subsequently grouped into 3 categories for analysis (0, 1, ≥2).
Nutrient Utilization
On completion of the 20-wk feeding study, 4 cows from each treatment were selected for use in a nutrient-utilization study, with the cows selected from each treatment group balanced for daily milk yield and BW. Cows were transferred in pairs from the main experimental group into a digestibility unit, with pairs transferred at approximately 0900 h on Monday or Thursday of each week. The first pair transferred comprised cows from treatments FB0 and FB-Low; the second, cows from treatments FB-High and FB0; and the third, cows from FB-Low and FB-High. This pattern was repeated for the remaining 3 pairs. Cows were tied by the neck in individual stalls fitted with a rubber mat, and continued to be offered their experimental rations via a feedbox located at the front of each stall. Rations were offered ad libitum daily at 0900 h, at proportionally 1.07 of the previous d intake, and uneaten ration was removed the following day at 0800 h. Soy hulls were offered (0.5 kg at each milking) via a plastic feed bucket, which was placed within the feedboxes during milking (at 0630 and 1630 h). Cows had access to fresh water at all times via a drinker located within each stall.
Measurements of nutrient utilization commenced 24 h after cows were placed in the digestibility unit and comprised a 6-d feeding period, followed 48 h later by a 6-d total feces and urine collection period. Following collection of feces and urine on d 3 of the 6-d collection period, each pair of cows was transferred into 2 indirect open-circuit respiration chambers for measurement of CH4 production, with each chamber used twice with each treatment, to remove any possible chamber effects. Cows remained in the chambers for 72 h, with measurements of CH4 and CO2 production recorded during the final 48-h period used in the subsequent analysis. The milking routine, milk sampling, milk analysis, feces and urine collection methodologies, sample management throughout the 6-d collection period, specifications of the respiration chambers, details of measurements of CH4 production, laboratory analysis processes of all feed, milk, urine, and feces, have been described in our previous work (
). In the present study, a mean CH4 recovery rate of 98.5% was recorded.
Feed Analysis
Samples of grass silage were taken daily throughout the experiment and dried at 85°C for 24 h to determine oven DM (ODM) content. Subsamples of the dried milled silages were taken 3 times weekly and bulked for every 14 d, with the bulked sample analyzed for NDF, ADF, and ash concentrations. A fresh sample of grass silage was taken weekly and analyzed for concentrations of N, ammonia-N, lactic acid, acetic acid, ethanol, propanol, and gross energy (GE), and for pH, whereas ME content was predicted using near-infrared reflectance spectroscopy, as described by
The use of near infrared reflectance spectroscopy (NIRS) on undried samples of grass silage to predict chemical composition and digestibility parameters.
. A sample of each experimental concentrate offered (FB0, FB-Low, and FB-High) was taken weekly, bulked for each 2-wk period, and subsequently dried at 85°C for 24 h to determine ODM content. Additional samples of each experimental concentrate were taken weekly, bulked for each 2-wk period, dried at 60°C for 48 h, and analyzed for N, NDF, ADF, ash, GE, and starch concentrations. These latter samples were further bulked for each 4-wk period and analyzed using the procedure developed by
for ADIN and NDIN. Acid detergent insoluble protein (ADIP) and neutral detergent insoluble protein (NDIP) were calculated as ADIP = 6.25 × ADIN and NIDP = 6.25 × NDIN, respectively.
Straw and soy hulls offered were sampled weekly, the samples bulked every 4 wk, and the bulked samples dried at 85°C for 24 h to determine ODM content. Dried samples were subsequently analyzed for N, GE, ADF, NDF, and ash content. Laboratory analyses of all feed samples were undertaken as described previously by
Two primiparous cows from treatment FB-Low did not complete the study due to lameness and very low milk yield, respectively. Mean weekly data for DMI, milk production, milk composition, milk constituent yields, and BW over the 20-wk experimental period were analyzed using REML repeated measures analysis (autoregressive order 1). The mixed model used included the following terms as fixed effects: Parity (1, 2, 3, >3) + Wk (1–20, as the repeated-measures time factor) + FB inclusion level (FB0, FB-Low, FB-High) + Wk × FB inclusion level, and cow was fitted as a random effect. Locomotion score and BCS data were analyzed using the same model, except that fortnightly data were used as the repeated-measures time factor. Milk FA and blood metabolites were analyzed using a similar model, except that for the former, sampling week (10 and 20 post-calving) was used as the repeated-measure time factor, whereas with the latter, sampling week (2, 4, 6, 8, 10, and 14-post calving) was used as the repeated-measures time factor. Continuous BW, BCS, and fertility data were analyzed by ANOVA, with parity included as a covariate. For each continuous variable, polynomial contrasts of order 1 were fitted to the treatment effects to test the data for linear effects associated with increasing FB level. Binomial fertility and health data were analyzed via generalized linear model regression analysis using the binomial distribution with a logit link function. The model included treatment as a term, with parity included as a covariate in the case of mastitis incidence. Mucus scores of 2 and 3 were combined into a single category (2–3), yielding categories of 0, 1, and 2–3 for analysis. These categorical data were analyzed using a generalized linear mixed model (ordinal logistic regression) with random effects (proportional odds model). The factorial arrangement of treatment (FB0, FB-Low, or FB-High) and sampling week (wk 6 and 12 for feces scores; wk 2, 4, and 6 for mucus scores) were fitted as fixed effects, and cow was fitted as a random effect. Significance was identified using chi-squared. Mean nutrient utilization data over the 6-d measurement period, and mean CH4 production data over the 2-d measurement period, were analyzed using ANOVA. All data were analyzed using GenStat (Release 18.1; VSN International Ltd., Oxford, UK).
RESULTS
The silage offered had ODM, CP, and ME contents of 293 g/kg, 145 g/kg of DM, and 11.0 MJ/kg of DM, respectively (Table 2). The 3 experimental concentrates had similar CP contents (mean, 223 g/kg of DM), and the starch contents of FB0, FB-Low, and FB-High concentrates were 291, 313, and 338 g/kg of DM, respectively. However, the NDIP of the concentrates decreased, being 112, 98, and 90 g/kg of DM for FB0, FB-Low, and FB-High, respectively, and the respective values for ADIP were 28.5, 21.6, and 16.6. The straw offered had CP, ADF, NDF, and ash contents of 46 (SD ±11.2), 501 (±9.6), 866 (±20.7), and 57 (±13.2) g/kg of DM, and a GE content of 18.5 (±0.23) MJ/kg of DM. The respective values for the soy hulls offered were 111 (±4.3), 492 (±11.3), 685 (±29.4), 53 (±1.1) g/kg of DM, and 17.5 (±0.06) MJ/kg of DM, with a starch content of 13.9 (±7.84) g/kg of DM.
Table 2Chemical composition of the grass silage and experimental concentrates (FB0, FB-Low, and FB-High)
Neither total DMI nor milk yield were affected by treatment (P > 0.05; Table 3). Silage DMI was estimated as 9.7, 9.6, and 9.8 kg/d for FB0, FB-Low, and FB-High, respectively (based on ration formulation), and the respective values for concentrates were 11.8, 11.7, and 12.2 kg/d. Cows on FB0 had higher milk fat content than those on FB-High (P = 0.031), whereas cows on FB0 and FB-Low had higher milk protein content than those on FB-High (P < 0.001). Milk lactose content was not affected by treatment (P > 0.05). Milk fat yield (P < 0.001), milk protein yield (P = 0.035), and milk fat plus protein yield (P = 0.007) were higher with FB0 than with either FB-Low or FB-High. All of these DMI and milk production parameters varied with time post-calving (P < 0.001; see Figure 1 for DMI and Figure 2 for milk yield), but we found no treatment × time interaction for any of these parameters. Milk fat, protein, and lactose concentrations showed linear decrease with FB inclusion (P = 0.012, P < 0.001, and P = 0.046, respectively), as did milk fat yield (P = 0.002), milk protein yield (P = 0.020), and milk fat plus protein yield (P = 0.006).
Table 3Effects of field bean (FB) inclusion level in dairy cow concentrates on mean DMI and milk production parameters over the 20-wk experimental period, and on mean milk fatty acid (FA) concentrations (mean of 2 sampling occasions)
Figure 1Silage DMI (dashed lines) and total DMI (solid lines) responses of dairy cows over the first 20 wk of lactation, to concentrates containing a range of field bean (FB) inclusion levels (FB0, FB-Low, or FB-High). Silage DMI: SEM = 0.22; treatment P = 0.497; time P < 0.001; treatment × time P = 0.722. Total DMI: SEM = 0.50; treatment P = 0.497; time P < 0.001; treatment × time P = 0.722.
Figure 2Milk yield response of dairy cows to concentrates containing a range of field bean (FB) inclusion levels (FB0, FB-Low, or FB-High) over the first 20 wk of lactation. SEM = 0.89; treatment P = 0.272; time P < 0.001; treatment × time P = 0.723.
The total concentration of SFA in milk fat was lower with FB0 than with either FB-Low or FB-High (P = 0.028), but the reverse was true for the total concentration of MUFA (P = 0.016; Table 3). Treatment had no effect on the total PUFA content of milk fat (P > 0.05). Concentrations of C18:0 (P = 0.002), C18:1 cis-9 (P = 0.022), and C18:2 cis-9,trans-11 (P < 0.001) decreased with increasing FB inclusion; however, concentrations of C16:0 (P < 0.001) and C18:3n-3 (P < 0.006) increased. Concentrations of C14:0 and C18:2 were not affected by treatment (P > 0.05). Although total PUFA concentrations (P < 0.001) and concentrations of C16:0 (P = 0.002), C18:0 (P < 0.001), C18:2 (P < 0.001), and C18:3n-3 (P < 0.001) differed between the 2 sampling periods (time) with the exception of the latter (P < 0.001), we found no treatment × time interactions for any of the FA recorded (P > 0.05). Total MUFA concentrations (P = 0.016) and concentrations of C18:0 (P < 0.001), C18:1 cis-9 (P = 0.020), and CLA (P < 0.001) showed a linear decrease with FB inclusion, but concentrations of C16:0 (P < 0.001) and C18:3n-3 (P = 0.002) showed the reverse trend.
Although average BCS was unaffected by treatment (P < 0.05), end-of-study BCS tended to increase with increasing levels of FB inclusion (P = 0.054; Table 4), with this overall linear effect significant (P = 0.020). None of the BW parameters examined (average BW, end-of-study BW, nadir BW, BW loss to nadir, and days to nadir BW) were affected by treatment (P > 0.05). Both BCS (P = 0.046) and BW (P < 0.001) varied with time (Figure 3, Figure 4, respectively), but we found no treatment × time interactions for either (P < 0.05). Plasma glucose and BHB concentrations were unaffected by treatment (P > 0.05), although both increased with time post-calving (Figure 5a, P = 0.010; Figure 5b, P = 0.002, respectively). Plasma NEFA concentrations were significantly higher with FB0 than with FB-High (P = 0.022); however, plasma urea concentrations followed the reverse trend (P < 0.001). Plasma NEFA concentrations decreased with time post-calving (Figure 5c, P < 0.001), but plasma urea concentrations increased (Figure 5d, P < 0.001). We detected no treatment × time interactions for any of the blood metabolites examined (P > 0.05). With increasing FB inclusion levels, analysis revealed a tendency for linear decrease in plasma glucose concentrations (P = 0.067), whereas plasma NEFA concentrations showed a linear decrease (P = 0.005). Plasma urea levels increased linearly with FB inclusion (P < 0.001).
Table 4Effects of field bean (FB) inclusion level in dairy cow concentrates on BCS, BW, and plasma metabolite concentrations
Figure 3Body condition score (BCS) response of dairy cows to concentrates containing a range of field bean (FB) inclusion levels (FB0, FB-Low, or FB-High) over the first 20 wk of lactation. SEM = 0.024; treatment P = 0.259; time P < 0.046; treatment × time P = 0.248.
Figure 4Body weight response of dairy cows to concentrates containing a range of field bean (FB) inclusion levels (FB0, FB-Low, or FB-High) over the first 20 wk of lactation. SEM = 8.0; treatment P = 0.899; time P < 0.001; treatment × time P = 0.760.
Figure 5Effect of field bean (FB) inclusion level in the diet (FB0, FB-Low, or FB-High) on plasma chemical concentrations among dairy cows. (a) β-Hydroxybutyrate: SEM = 0.018; treatment P = 0.419; time P = 0.002; treatment × time P = 0123. (b) Glucose: SEM = 0.036: treatment P = 0.153; time P = 0.010; treatment × time P = 0.954. (c) Nonesterified fatty acids: SEM = 0.020; treatment P = 0.022; time P < 0.001; treatment × time P = 0.633. (d) Urea: SEM = 0.103; treatment P < 0.001; time P < 0.001; treatment × time P = 0.567.
Fecal scores differed between wk 6 and 12 post-calving (overall SEM = 0.048, P = 0.012), but treatment had no effect on mean fecal scores (probability of having a score of 1, 2, or ≥ 3: 0.08, 0.47, and 0.45 for FB0; 0.05, 0.42, and 0.53 for FB-Low; 0.07, 0.45, and 0.48 for FB-High: overall SEM = 0.061, P = 0.878). We found no interaction between treatment and measurement period (SEM = 0.084, P = 0.783). The digestibility of ADF decreased (P = 0.035: linear, P = 0.013) from FB0 to FB-High, but none of the other digestibility coefficients were affected by FB inclusion level (Table 5). Neither total N intake, digestible N intake, N output in feces, urine, manure, or milk, nor any of the N utilization coefficients examined were affected by treatment (P > 0.05; Table 6). Treatment had no effect (P > 0.05) on GE intake, digestible energy (DE) intake, ME intake, energy output in feces, urine, CH4, heat, or milk, or any of the energy utilization coefficients examined (Table 7). Total CH4 production, CH4 as a proportion of DMI, OM intake, and milk production were unaffected by treatment (P > 0.05; Table 7). Similarly, CH4-E, as a proportion of GE or ME intake, was unaffected by treatment (P > 0.05). There was a linear trend for ME/GE (P = 0.071) to decrease with increasing FB inclusion in the diet.
Table 5Effects of field bean (FB) inclusion level in dairy cow concentrates on DMI, milk yield, and total ration digestibility coefficients
Vaginal mucus scores decreased with week (2, 4, and 6) post-calving (overall SEM = 0.042, P < 0.001), but treatment had no effect on mean mucus scores (probability of having a score of 0, 1, or 2–3: 0.70, 0.13, and 0.17 for FB0; 0.71, 0.13, and 0.16 for FB-Low; 0.67, 0.16, and 0.17 for FB-High; overall SEM = 0.047, P = 0.97). Analysis revealed no interaction between treatment and measurement week (SEM = 0.072, P = 0.476). Treatment had no effect on the proportion of cows showing commencement of luteal activity before d 42 post-calving, days to commencement of luteal activity, or peak progesterone content at commencement of luteal activity (P > 0.05; Table 8). Treatment did not affect conception to first service, conception to first and second service, days to conception, or the proportion of cows pregnant at the end of the breeding season (P > 0.05). The proportion of cows with at least one incident of digestive upset, mastitis, or lameness, as well as mean locomotion score, were unaffected by treatment (Table 8; P > 0.05).
Table 8Effects of field bean (FB) inclusion level in dairy cow concentrates on fertility parameters and dairy cow health (95% CI in parentheses)
Item
Treatment
SEM
P-value
FB0
FB-Low
FB-High
Pre-d 42
Proportion of cows showing luteal activity
0.75 (0.54–0.88)
0.67 (0.44–0.83)
0.79 (0.58–0.91)
0.632
Days to commencement of luteal activity
22.1
21.6
22.1
2.25
0.980
Peak progesterone content at commencement of luteal activity (ng/mL)
31.3
26.9
32.5
2.74
0.353
Conception to first service (proportion)
0.45 (0.27–0.65)
0.33 (0.16–0.55)
0.30 (0.15–0.51)
0.511
Conception to first and second service (proportion)
0.62 (0.42–0.79)
0.57 (0.36–0.75)
0.50 (0.31–0.69)
0.681
Days to conception
84
102
90
9.5
0.412
Cows pregnant at end of breeding season (proportion)
0.91 (0.70–0.98)
0.80 (0.57–0.92)
0.74 (0.53–0.88)
0.307
Proportion of cows with at least one incidence of illness
Average daily FB intakes for FB-Low and FB-High were estimated as 4.1 and 8.4 kg/cow, respectively, with the latter substantially higher than FB intake levels used in any previous study. At the maximum inclusion level adopted (FB-High, 698 g of FB per kilogram of concentrate), FB replaced all of the soybean meal, rapeseed meal, maize gluten feed, and wheat in the concentrate, compared with the control treatment (FB0). Although it was not possible to achieve a common starch level with each of the concentrates (291, 309, and 338 g per kilogram of DM for FB0, FB-Low, and FB-High concentrates, respectively), all had a similar CP content, resulting in a total diet CP content of 180 g per kilogram of DM across the 3 diets. This total diet CP content was somewhat higher than planned, yet it remains typical of many grass silage–based diets offered in the United Kingdom.
Cow Performance
The potential negative effects of antinutritional factors within FB on DM intake are often cited by nutritionists as a reason for low inclusion levels in dairy cow diets. In addition, preference trials (
) have suggested that FB may be unpalatable. However, the current study clearly demonstrated that FB intakes of up to 8.4 kg/cow per day had no negative effects on DM intakes. This is in agreement with most published studies (
), albeit the maximum daily FB intake in these studies was 5.0 kg/cow per day, but the diets offered and concentrate feeding strategies adopted also differed considerably from those in the current study. In contrast,
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
found intakes to be reduced when FB replaced rapeseed meal in dairy cow diets. Those authors suggested that this may have been due to a reduction in NDF digestibility, a poorer amino acid profile with FB, and a “pull effect” caused by the higher milk yield with the rapeseed meal treatment. Despite reductions in ADF digestibility with FB inclusion in the current study, this had no effect on DMI.
In common with DMI, milk yield was unaffected by FB inclusion. However, the reduction in both milk fat and milk protein concentrations with FB inclusion meant that yields of fat, protein, and fat plus protein were lower with both FB-Low and FB-High, compared with the FB0 treatment. Literature evidence concerning the effects of FB inclusion on milk production and milk composition is inconsistent. For example, both
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
, in studies involving maximum FB intakes of 3.7 and 2.6 kg of DM per day, respectively, observed reductions in milk yield when rapeseed meal was replaced by FB. Milk protein content was reduced in the former study, but milk fat content was unaffected by FB inclusion in either study. In contrast, at an FB intake level of 4.7 kg/d (
), neither milk yield nor milk fat content were affected by FB inclusion level, and milk protein content tended to decrease (P = 0.08). In several on-farm studies conducted in Italy (
) milk yield was reduced but milk fat content was increased with FB inclusion. However, FB inclusion levels in these studies were low, normally less than 2.0 kg/cow per day.
Energy intake is a key driver of milk protein content, yet the reduction in milk protein content with the FB-Low treatment is unlikely to be driven by energy supply, given the similar intakes across all treatments and the absence of treatment effects on any of the BW or BCS score parameters. Indeed, plasma NEFA concentrations decreased with FB inclusion; however, there was a trend for end-of-study BCS to increase with FB inclusion, suggesting an improved energy balance with FB-High. Rather, it is more likely that this reduction was caused by a deficit of specific amino acids. For example, FB is lower in both lysine and methionine than are soybean meal or rapeseed meal (
), lysine supply relative to requirements were “adequate” for all 3 treatments, whereas for methionine the FB0 ration was classified as “borderline,” with both the FB-Low and FB-High rations classified as “poor.” When
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
replaced rapeseed meal with FB, total concentrations of essential amino acids and plasma concentrations of lysine and methionine were reduced; these authors suggested that this was likely partly responsible for the reduction in milk protein content and milk protein yield that they observed.
In the current study, FB replaced or partially replaced both maize starch (less degradable than FB starch) and wheat starch (more degradable than FB starch) in the concentrate (
), likely affecting rumen fermentation patterns and the associated reduction in milk fat content with the FB-High treatment. This reduction was in contrast to most previous studies; we propose that with the FB-High treatment, a reduction in rumen acetate production occurred due to the higher-concentrate starch and slightly lower fiber content, and this may have contributed to a reduction in de novo synthesis of milk fat. Possible adverse effects of diet on rumen function are supported in part by the lower ADF digestibility observed with this treatment. In addition, we postulate that altered rumen fermentation and biohydrogenation pathways, leading to inhibition of mammary lipid synthesis by specific fatty acid intermediates, as proposed by
), and, as such, the likely deficit of methionine in the FB diets in the present study may have contributed to the reduction in milk fat concentrations.
Effects of FB inclusion on the fatty acid profile of milk do not appear to have been examined previously; we found that FB inclusion in the current study increased the degree of saturation of the milk produced. This appears to have been largely driven by increasing concentrations of C16:0 in milk, with FB particularly high in C16:0 fatty acids (
estimating that 42% of C16:0 is derived directly from the diet. Similarly, decreasing concentrations of C18:1 with FB inclusion may reflect the higher concentration of C18:1 in soybean compared with FB (
It is possible that the reduction in cow performance observed with FB inclusion could be avoided by supplementing the diet with specific limiting amino acids. Nevertheless, cow performance data need to be considered within the context of the total diet CP in the study (180 g/kg of DM), a level not unusual with early-lactation dairy cows offered grass silage–based diets in the United Kingdom and Ireland due to the often variable composition of grass silage on offer. However, it is possible that greater differences between treatments might have been observed with lower-protein diets as, in general, responses to protein source and quality tend to be greater at lower diet protein levels. Nevertheless,
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
). For example, the FB variety Fuego is a colored-flower variety, which is known to be a tannin-containing type. However, fecal scores—a simple proxy for rumen function—were unaffected by treatment in this study. In addition, with the exception of ADF, no digestibility coefficients for the other parameters examined were affected by FB inclusion. In our previously published work, we found fiber digestibility to be unaffected with an FB intake similar to that used with the FB-Low treatment (
), but the reduction in ADF digestibility with the highest FB treatment in the current study is similar to the reduction in NDF digestibility observed by
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
attributed this to the readily fermentable FB carbohydrate and associated fermentation products, decreasing rumen pH and having a negative effect on cellulolytic bacteria activity, rather than to the presence of antinutritional factors.
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
also found the apparent digestibility of DM, OM, N, and starch to increase with FB inclusion, but they attributed this to the lower proportion of silage in the diets of cows offered FB, rather than to FB inclusion per se. Thus it appears that any effects of FB inclusion on digestibility are likely to be caused by other dietary factors, rather than by the presence of antinutritional factors. This is supported by the findings of
) indicated a reduction in the ratio of milk N to N intake with FB inclusion, a similar effect was not observed in the nutrient utilization portion of the current study, in agreement with other previous findings (
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
). Nevertheless, the nutrient utilization phase in the current study involved only 4 cows per treatment and was undertaken at the end of the 20-wk feeding study. When N use efficiency over the entire study period is examined using treatment mean intake and milk production data (with milk N calculated as milk protein/6.38), the ratio of milk N to N intake was 0.30, 0.28, and 0.27 for FB-0, FB-Low, and FB-High, respectively. The decreasing N use efficiency with increasing FB inclusion levels was reflected in an increase in plasma urea N levels, in agreement with
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
) have found both blood and milk urea N concentrations to be either reduced or unchanged with FB inclusion. Protein in FB has a higher degradability in the rumen compared with protein from soybean meal or rapeseed meal (
), except that these latter authors inexplicably found a quadratic effect of FB inclusion on urine E output. Similarly, none of the CH4 production parameters examined were affected by FB inclusion, in agreement with previous indications (
found that ratios of CH4 to DMI and CH4 to ECM were unaffected when rapeseed meal was replaced by FB, whereas total CH4 production tended to be reduced, which they attributed in part to the higher fat content of rapeseed meal. Despite evidence suggesting that CH4 production may be reduced when ruminants are offered legume forages (
), there appears to be little evidence that this is the case with grain legumes, in line with the findings of the current study. Our results demonstrate enteric CH4 emissions to be unaffected by FB inclusion, but the effects of replacing imported protein ingredients with locally grown FB, across the life cycle of milk, needs to be examined.
Fertility and Health
Negative energy balance can have a detrimental effect on reproductive performance (
), but BCS and BW data in the current study suggest that energy balance differed little between treatments. However, the presence of phytoestrogens in many legumes species (for example, daidzein and genistein;
), including some FB cultivars, is often highlighted as a cause for concern in relation to dairy cow fertility. Phytoestrogens, compounds that are produced naturally in many plant species, may have adverse effects on ovarian function, including possibly inhibited luteinizing hormone surge, altered development of or lack of dominant follicles, abnormal follicular waves, early embryonic death, and repeat breeding (
), neither the time to onset of cyclicity, the proportion of cows cycling within 42 d of calving, nor the peak progesterone concentrations during the first estrus cycle were affected by diet in the present study. While actual conception rates within the study must be considered within the context of the number of cows on the study, we found no evidence that reproductive outcomes were affected by FB inclusion level. Vaginal mucus scores, assessed as an indicator of immune status, did not differ across treatments, suggesting that FB treatment has no effect on immune status. In addition, none of the health parameters examined were affected by FB inclusion in the diet, which again suggests that immune function was unaffected by treatment.
CONCLUSIONS
Including FB (Vicia faba. var. Fuego) in dairy cow diets at up to 8.4 kg/cow per day had no detrimental effects on feed intake, milk production, body tissue reserves, fertility performance, cow health, or diet digestibility. The reduction in milk fat content with the highest FB treatment was likely due to the effect on rumen function of the high starch levels with this treatment, and the reduction in milk protein levels with FB inclusion was likely due to a deficit in specific limiting amino acids. Methane production was unaffected by FB inclusion.
ACKNOWLEDGMENTS
This study was funded by the Department of Agriculture, Environment, and Rural Affairs (DAERA) in Belfast, Northern Ireland, and by AgriSearch, Innovation Centre, Hillsborough, Northern Ireland. David Johnston was in receipt of a DAERA Postgraduate Studentship. The authors thank the staff of the Dairy Unit at the Agri-Food and Biosciences Institute (Hillsborough, Northern Ireland) for management of the experimental animals, and the laboratory staff for analysis of samples. The invaluable help provided by Jim Uprichard (Trouw Nutrition, Belfast, UK) in formulating the experimental concentrates was much appreciated.
REFERENCES
Abdulla J.M.
The use of field beans as a feed for broiler chickens. PhD Thesis.
Concentrate supplementation of a diet based on medium-quality grass silage for 4 weeks prepartum: Effects on cow performance, health, metabolic status, and immune function.
The use of near infrared reflectance spectroscopy (NIRS) on undried samples of grass silage to predict chemical composition and digestibility parameters.
Effects of replacing rapeseed meal with fava bean at 2 concentrate crude protein levels on feed intake, nutrient digestion, and milk production in cows fed grass silage–based diets.
30818-5&id=gr1.jpg){kind=link}
30818-5&id=gr2.jpg){kind=link}
30818-5&id=gr3.jpg){kind=link}
30818-5&id=gr4.jpg){kind=link}
30818-5&id=gr5.jpg){kind=link}