Trait responses of Holstein-Friesian cows with different genetic merits for milk yield in relation to low-input feeding.

Contrary to conventional dairy cattle farming which focus on increased farm profitability due to increased cow productivity, low-input systems aim for efficient pasture-based milk production by reducing production costs. Low-input dairy cattle systems are also highly appreciated by consumers, because of welfare perceptions. From a breeding perspective, the targeted selection on increased productivity is accompanied by modifications to production systems, especially a feeding focus in barn systems on feeding high levels of concentrates. Even under these feeding conditions, the challenge remains to compensate for the negative energy balance of high-yielding cows during early lactation. The consequences of energy deficiency are accentuated in low-input feeding systems in genetically superior milk selection lines. Gaps in the energy balance due to reduced energy intake induce the mobilization of body fat depots to sustain milk production, implying an increased risk for metabolic disorders. Thus, in support, severe energy deficiency for high-yielding cows kept in grazing systems exhibited ongoing impaired health and fertility. Oppositely, in some other studies, high-yielding cows showed milk production superiority over local robust breeds under low-input grazing conditions, but with indications of increased disease susceptibility. From a quantitative genetics perspective, hampered phenotypic expressions of the genetic milk yield potential of high-yielding cows kept in low-input production systems are expected.

Different reactions of cows from the same breed in different production or feeding environments address possible genotype × feeding interaction (GFI), resulting in the need to select specific sires for specific systems. However, in a population-wide approach, feeding effects might be confounded with other disturbing parameters, for example, climate, milking technique, or grazing conditions, hampering effective genetic comparisons. Accordingly, most of the previously conducted breed comparisons neglected imperative cross-classified research conditions, and specific breeds represented only one specific system. In the study cited, the novelty was the comparison of Holstein-Friesian genetic lines in relation to high- and low-input feeding in a standardized and identical organic farm environment. The research design was unique to study trait responses and adaptation mechanisms in the period from high- to low-input feeding by keeping all other environmental factors identical.  The first objective was to study trait responses of the cows with differing genetic milk merit when performing under challenging low-input conditions. As a second objective, the genetic milk merit × feeding system interaction was analysed.

In the study, a variety of primary and functional trait responses during a recording period of 12 months for 125 lactating Holstein-Friesian cows was investigated. In the design, a cross-classified genetic merit × feeding group (FG) experiment was established in a barn. The barn building consisted of two identical compartments, A and B, mirrored across the feed alley regarding husbandry characteristics, technical equipment, and management systems, but with different feeding systems. Compartment A was used for organic high-input (HI) feeding (energy content of 7.03 MJ NEL/kg DM), and compartment B for organic low-input (LI) feeding (energy content of 6.25 MJ NEL/kg DM). For allocation of the same number of cows into the two feeding groups (FG) at the beginning of the experiment, a cross-classified experiment was designed; that is, identical genetic merits of lactating cows for milk yield in both groups, considering the 50^th percentile of estimated breeding value (EBV) for milk yield. Simultaneously, in the semi-randomized allocation procedure, equal distributions in both groups regarding parity, lactation stage, and the first two principal components from the pedigree relationship matrix were considered. Generalized linear mixed models were applied to test the main effects of the FG (two classes: HI and LI), genetic milk merit classes (three classes: EBV-low, EBV-medium, and EBV-high), and the respective genotype x feeding interaction (GFI) on a broad pattern of primary and functional traits.

The adaptation capability of high-yielding HF genetics to LI conditions during the challenging feeding transition period was manifested through stable and significantly higher least square means (LSM) for milk yield when compared with the other genetic milk merit groups. However, significant GFI for traits reflecting energy efficiency and metabolic physiology was observed, especially for MUN, the ketosis indicator β-hydroxy butyrate (BHB) concentration at lactation day five, and fatty acid profiles, displaying re-rankings of EBV-high, EBV-medium and EBV-low under HI and LI feeding. Hence, EBV-high cows displayed production superiority in milk yield, but partly by compromising metabolic stability, especially indicated through a substantial decline of LSM for body condition score (BCS) and back fat thickness under the LI feeding conditions.

Conclusions: The results from this genotype × feeding experiment indicated the general adaptability of high-yielding Holstein-Friesian genetics to LI conditions during the challenging feeding transition period. This manifested especially through a stable milk yield superiority of the EBV-high cows when compared with the other genetic milk merit groups. However, re-rankings of EBVs in both HI and LI and significant GFI for traits reflecting energy efficiency and metabolic physiology, including MUN, the ketosis indicator BHB, and certain fatty acid profiles, were observed. Hence, EBV-high cows maintain production superiority in milk yield, but with indications of compromised metabolic stability.