Forage lucerne for grazing dairy cows: effects on milk yield, milk urea and fatty acid composition

Herbage measurements

Pre- and post-grazing herbage samples were collected on Days 9 and 10 of each run for botanical and chemical composition analyses. About 250 g fresh sample was collected by harvesting herbage at 4.5 cm grazing height from 8 to 10 random locations within each allocated grazing area by using a pair of clippers. The collected sample was homogenised and separated into the following three subsamples: for DM content determination; for botanical and nutrient composition; and for FA analysis. To determine DM content of each sward type, approximately 50 g fresh herbage was weighed and dried at 60°C for 48 h and reweighed. For botanical composition analyses, approximately 100 g fresh herbage was separated into sown species, weed, and dead material prior to oven drying at 60°C for at least 48 h to determine the DM content of each component. Oven-dried botanical components were then recombined within treatment × day and analysed for organic matter (OM), crude protein (CP) acid detergent fibre (ADF), neutral detergent fibre (NDF), water-soluble carbohydrates (WSC) and digestible OM in the DM (DOMD) by using near-infrared spectrophotometry (NIRS, Model: FOSS NIRSystems 5000, Maryland, USA). Digestible organic matter in the DM was used to estimate metabolisable energy (ME) by using the equation: ME (MJ/kg DM) = DOMD (g/kg DM) × 0.016 (AFRC 1993). The third sample of approximately 100 g fresh herbage, collected pre-grazing, was stored at −20° and later freeze dried for individual FA analysis. Freeze dried FA samples were prepared through transmethylation and analysed by gas chromatography (with AOC-20i auto-sampler, Shimadzu GC-2010, Japan), according to AOAC (2012) Method 2012.13 using a Varian CP742 silica capillary column (0.25 × 100 m × 0.2 μm).

Animal measurements

Statistical analyses

The mixed model ANOVA using lme4 package ver. 1.1-35.1 (Bates et al. 2015) in R software ver. 4.2.0 (R Core Team 2020) was used to compare means. For herbage chemical composition and nutrient-intake analyses, herbage type was fitted as fixed effect and date was nested in run fitted as random effect. For animal-performance parameters such as milk yield and composition, individual milk FAs, and MU, the herbage type was fitted as fixed effect, with date nested in run plus cow fitted as random effect. During the adaptation period of the first run, one cow on lucerne had to be excluded from the experiment because of lameness. Statistical significance was declared when P < 0.05 and a tendency was declared when P ≥ 0.05 but <0.1.

Herbage characteristics

The dry-weight proportions of perennial ryegrass and white clover within the PRW swards were 69.2% and 5.8% respectively. The remaining proportion consisted of weed (17.1%) and dead material (7.9%). The purity of the lucerne was high, forming more than 90% of the herbage dry weight. Over similar regrowth intervals, lucerne herbage accumulated a 1.3-fold greater biomass than did PRW (Table 1). Lucerne herbage had 31% higher CP (P = 0.037), 15% higher ADF concentration (P = 0.075), but 49% lower WSC concentration (P = 0.011) and 8% lower ME content (P = 0.063) than did PRW.

Overall, total FA accounted for 2.5% of the DM and did not differ between PRW and lucerne herbages. In both forages, alpha linolenic acid (ALA: C18:3 cis-9,12,15) was the predominant FA (Table 2), followed by palmitic (C16:0) and linoleic acids (LA: C18:2 cis-9,12). Alpha linolenic acid was 40% higher (P = 0.006) in PRW than in lucerne herbage, whereas linoleic and palmitic acids were 63% (P = 0.004) and 20% (P = 0.022) higher in lucerne than in PRW respectively.

Intake and performance

Apparent daily DM intake did not differ between treatments (19.7 ± 0.338 kg DM/cow; P = 0.334). Lucerne accounted for approximately 43% of apparent DM intake for cows grazing lucerne-based herbage. As with DM, the intake of ME was similar between groups (222 ± 0.043 ME: MJ/day; P = 0.334). Because of the 7.7% numerical difference in NDF concentration between PRW and lucerne, cows on control diet ingested more NDF (+1.14 kg/day) than did cows on lucerne (P < 0.001). The intake of WSC was also higher (+1.26 kg/day) in control cows than in lucerne cows (P < 0.001). The higher CP content in lucerne was reflected in the increased N intake (+107 g N/day) from cows grazing lucerne compared with control (P < 0.001). Because of the higher N intake from the cows grazing lucerne, the N:ME intake ratio was 1.3-fold higher in cows grazing lucerne than those on control (P < 0.001). ALA intake was 20% lower in cows grazing lucerne than in control cows, whereas LA intake was 1.2-fold higher in cows grazing lucerne than in those grazing the control diet.

There was no effect of pasture type on milk yield or milk fat, protein or lactose composition (Table 3). MU concentration was 43% higher in lucerne-fed cows than in those fed the control diet. However, total N in milk was similar for both treatments. Including lucerne in the diet decreased NUE (milk N/intake N) compared with feeding PRW only (0.27 vs 0.33; P = 0.001).

Milk FAs

The concentrations of individual milk FAs are displayed in Tables 4 and 5. Saturated FAs were the predominant FAs in the milk, forming nearly 73 g/100 g of FA in milk, followed by monounsaturated FAs at 20.6 ± 0.979 g/100 g of FA and polyunsaturated FAs at 3.14 ± 0.218 g/100 g of FA (Table 4). Saturated and monounsaturated FAs in milk were similar between cows grazing control and lucerne, whereas polyunsaturated FAs were 19% higher in the milk of cows grazing lucerne than in those grazing control. The differences in polyunsaturated FAs reflected the variation in the proportions of the long-chain FAs (i.e. omega-3; Table 5) such as ALA, LA and eicosapentaenoic acid (EPA; C20:5 cis-5,8,11,14,17), where they were 35% (P < 0.0001), 28% (P < 0.0001), and 13% (P = 042) respectively, higher in the milk of cows grazing lucerne than in those grazing control. The proportions of other functional FAs found in milk fat, such as conjugated linoleic acid (CLA; C18:2 cis-9 trans-11; 0.947 ± 0.16 g/100 g FA), vaccenic (C18:1 trans-11; 2.55 ± 0.55 g/100 g FA) and oleic acids (C18:1 cis-9; 13.8 ± 0.48 g/100 g FA), were similar in milk of cows grazing both herbages. The rate of transfer of LA from herbage to milk was similar between the two groups (0.146 ± 0.02 g intake/g in milk; P = 0.97). However, the rate of transfer of ALA was 1.8-fold higher in the group fed lucerne herbage than in cows fed control (0.042 vs 0.024 ± 0.005; P < 0.05).

Milk production and composition

Substituting 43% of the traditional PRW with lucerne in the diet of mid-lactation and spring-calving cows did not alter their milk production or composition. The findings corroborated those reported by Smith et al. (2013) for an early lactation (September–October) and mid-lactation (November–December) experiment that compared milk production and composition of 45 lactating dairy cattle cows fed either 100% lucerne or PRW in Canterbury, New Zealand. However, an indoor experiment involving 24 dairy cattle conducted in Waikato, New Zealand, showed improvements in milk yield and milk components in cows offered freshly cut lucerne compared with those fed the conventional PRW (Woodward et al. 2010). In the study reported by Woodward et al. (2010), improvements in milk production were congruent with improvement in DM intake from lucerne-fed cows. In contrast to this, Smith et al. (2013) reported that cows grazing PRW maintained DM and ME intakes compared with cows grazing lucerne, in line with the results from the current experiment. The use of irrigation to maintain herbage quality of the PRW control in the two Canterbury trials (Smith et al. 2013; current study) is likely to have contributed to the lack of treatment differences. Further, with indoor experiments, such as conducted in Waikato, cows were offered freshly cut herbage daily with less opportunity to exhibit herbage selection as they are under grazing conditions (Pembleton et al. 2016). The absence of variation in DM and ME intakes between control and lucerne-fed cows in the current experiment is likely to indicate that cows from both groups selected for similar nutrients during grazing, resulting in similar milk yield and composition.

Typically, diet N transformed to milk N is in the range of 0.15–0.35 for grazing dairy cattle (Gourley et al. 2012); the corresponding mean range of NUE obtained in the current experiment between 0.27 and 0.33 shows that NUE was within the published range for both feed types. Milk urea can be a useful indicator of dietary protein surplus, with most research suggesting excess protein (N) in the diet when MU is greater than 30 mg/dL (Spek et al. 2013; Mangwe et al. 2024). In the current study, the MU of cows fed lucerne remained within the recommended optimal range of 17–27 mg/dL suggested by Kohn et al. (2002). By contrast, the MU concentration among cows in the control was below the recommended lower threshold of 16–17 mg/dL recommended by those same authors (Kohn et al. 2002), at which a reduction in milk yield might be observed. Calculations to estimate metabolisable protein requirements and minimum dietary CP required for dairy cows to produce 25 kg milk and 2.24 kg milk solids in our study showed a minimum dietary CP of 168 g/kg DM in control and 166 g/kg DM in lucerne group, which was enough to meet the cows’ requirements for production and maintenance. The increased N intake in the lucerne group did not result in any milk production response, but increased their N:ME ratio and reduced NUE, as reflected by their higher MU concentration.

Milk FA composition

In most herbages, ALA and LA are the pre-dominant polyunsaturated FAs, contributing to over half of the FAs in the herbage, although this is dependent on herbage species. Generally, the concentration of ALA is higher in PRW herbage than in legumes (Kälber et al. 2011), which results in higher ALA intakes from cows grazing PRW-based diets than legume-based herbages, as was the case in the current study as well. In contrast, the concentration of LA was greater in lucerne herbage than in PRW.

Few studies have explored the impact of feeding fresh lucerne on individual milk FAs of dairy cattle in temperate regions. Our previous research has demonstrated that forages such as plantain and chicory can alter milk FA composition (Mangwe et al. 2020). In the current study, our findings showed that milk from cows grazing lucerne herbage contained greater LA and ALA concentrations than that from cows grazing grass-based herbages. This is consistent with previous research using legumes, which reported enriched levels of these two FAs in milk of cows grazing legume-based herbage (Dewhurst et al. 2006; Elgersma 2015). The higher LA concentration in the milk of cows grazing lucerne might reflect the higher intake of the acid from the herbage than for those grazing the control, whereas the higher concentration of ALA in the milk of cows grazing lucerne, obtained despite their lower intake of the FAs in the diet, reflects the greater transfer efficiency of the FAs from herbage to milk as previously reported for legumes (Elgersma 2015). This may be due to the higher concentrations of secondary compounds such as saponins in lucerne herbage (Kozłowska et al. 2021), which could have affected rumen biohydrogenation and modified the quality of FAs in milk of dairy cattle (Szczechowiak et al. 2016; Orlandi et al. 2020). Nonetheless, we did not analyse for secondary compounds in the current experiment; however, their expected effect on rumen biohydrogenation warrants further investigation.

Lucerne is a high-nutritive forage that has demonstrated potential to improve the quality of milk while maintaining animal performance in the current experiment. Although we did not observe any signs of bloat, the potential to cause bloat in dairy cows warrants further investigation. Consideration of the practicalities of managing bloat on lucerne diets would likely involve the use of bloat oils as a preventative, such actions may disguise or accentuate the impact of lucerne on milk FA composition. The current findings warrant further investigation to fully understand the mechanism underlying higher concentration of long-chain FAs in milk of cows grazing lucerne in situ.