Residual feed intake of lactating Holstein-Friesian cows predicted from high-density genotypes and phenotyping of growing heifers

Impact of parity differences on residual feed intake estimation in Holstein cows

Residual feed intake (RFI) has been used as a measure of feed efficiency in farm animals. In lactating dairy cattle, RFI is typically obtained as the difference between dry matter intake observations and predictions from regression on known energy sinks, and effects of parity, days in milk, and cohort. The impact of parity (lactation number) on the estimation of RFI is not well understood, so the objectives of this study were to (1) evaluate alternative RFI models in which the energy sinks (metabolic body weight, body weight change, and secreted milk energy) were nested or not nested within parity, and (2) estimate variance components and genetic correlations for RFI across parities. Data consisted of 72,474 weekly RFI records of 5,813 lactating Holstein cows collected from 2007 to 2022 in 5 research stations across the United States. Estimates of heritability, repeatability, and genetic correlations between weekly RFI for parities 1, 2, and 3 were obtained using bivariate repeatability animal models. The nested RFI model showed better goodness of fit than the nonnested model, and some partial regression coefficients of dry matter intake on energy sinks were heterogeneous between parities. However, the Spearman's rank correlation between RFI values calculated from nested and nonnested models was equal to 0.99. Similarly, Spearman's rank correlation between the RFI breeding values from these 2 models was equal to 0.98. Heritability estimates for RFI were equal to 0.16 for parity 1, 0.19 for parity 2, and 0.22 for parity 3. Repeatability estimates for RFI across weeks within parities were high, ranging from 0.51 to 0.57. Spearman's rank correlations of sires' breeding values were 0.99 between parities 1 and 2, 0.91 between parities 1 and 3, and 0.92 between parities 2 and 3. We conclude that nesting energy sinks within parity when computing RFI improves model goodness of fit, but the impact on the estimated breading values appears to be minimal.

Dairy Cows Are Limited in Their Ability to Increase Glucose Availability for Immune Function during Disease

Shortages of energy and glucose have been hypothesized to play a key role in the development of and responses to production diseases in dairy cows during early lactation. Given the importance of glucose for immune functions, we used a recently established method for the estimation of glucose balance (GB) to evaluate glucose availability during disease phases. A dataset comprising ration analyses as well as individual daily milk yields (MY), dry matter intake (DMI), body weights, and health records of 417 lactations (298 cows) was used to calculate individual daily GB and energy balance (EB). The magnitude and dynamics of MY, DMI, GB, and EB were evaluated in the weeks before, at, and after diagnoses of inflammatory diseases in different stages of early lactation from week in milk 1 to 15. Diagnoses were categorized as mastitis, claw and leg diseases, and other inflammatory diseases. Mixed linear models with a random intercept and slope term for each lactation were used to evaluate the effect of diagnosis on MY, DMI, GB, and EB while accounting for the background effects of week in milk, parity, season, and year. When unaffected by disease, in general, the GB of cows was close to zero in the first weeks of lactation and increased as lactation progressed. Weekly means of EB were negative throughout all lactation stages investigated. Disease decreased both the input of glucose precursors due to a reduced DMI as well as the output of glucose via milk due to a reduced MY. On average, the decrease in DMI was −1.5 (−1.9 to −1.1) kg and was proportionally higher than the decrease in MY, which averaged −1.0 (−1.4 to −0.6) kg. Mastitis reduced yield less than claw and leg disease or other diseases. On average, GB and EB were reduced by −3.8 (−5.6 to −2.1) mol C and −7.5 (−10.2 to −4.9) MJ in the week of diagnosis. This indicates the need to investigate strategies to increase the availability of glucogenic carbon for immune function during disease in dairy cows.

Relationships among feed efficiency traits across production segments and production cycles in cattle

Understanding the relationships between feed efficiency traits measured in different stages of production is necessary to improve feed efficiency across the beef value chain. The objective of this study was to evaluate relationships among feed efficiency traits measured as growing heifers and breeding females and in their progeny in three full production cycles, and relationships of dam residual feed intake (RFI) with lifetime and lifecycle cow efficiency traits. Data were collected on 160 mixed-breed heifers from 240 d of age to weaning of their third progeny, and postweaning performance of progeny until harvest in experiments initiated in 1953, 1954, 1959, 1964, 1969, and 1974. Individual feed offered was recorded daily, and feed refusals measured every 28 d. Milk yield was measured at 14-d intervals throughout lactation by machine or hand milking. Females and progeny were weighed at 28-d intervals and progeny were harvested at a constant endpoint of live grade or age depending upon the experiment. Feed efficiency traits of RFI and residual BW gain (RG) were computed as the residual from linear regression for developing heifers, dams (RFI and residual energy-corrected milk [RECM]), and postweaning progeny. Feed:gain ratio (FCR) was computed for developing heifers and postweaning progeny, and feed:milk energy ratio (FME) was computed for dams. Various measures of cow efficiency were calculated on either a life cycle or lifetime basis using ratios of progeny and dam weight outputs to progeny and dam feed inputs. Pearson correlations were computed among traits adjusted for a random year-breed-diet group effect. Heifer RFI (0.74) and RG (-0.32) were correlated (P ≤ 0.05) with dam RFI in parity 1 only, but were not correlated (P > 0.05) with dam RECM in any parity. Heifer RFI was correlated (P ≤ 0.05) with progeny RFI (0.17) in parity 3 only. Heifer FCR was not correlated with dam FME or progeny FCR in any parity. Dam RFI was weakly correlated (r = 0.25 to 0.36; P ≤ 0.05) among parities, whereas dam FME and RECM were strongly correlated (r = 0.49 to 0.72; P ≤ 0.05) among parities. Dam RFI in parity 1 and 2 was weakly correlated (r = -0.20 to -0.33; P ≤ 0.05) with cow efficiency ratios that included dam weight as an output, whereas dam RFI in parity 3 was not correlated with any cow efficiency ratio. In conclusion, feed efficiency traits were poorly correlated across production segments, but moderately repeatable across production cycles.

The effects of energy metabolism variables on feed efficiency in respiration chamber studies with lactating dairy cows

The objective of the present study was to investigate factors related to variation in feed efficiency (FE) among cows. Data included 841 cow/period observations from 31 energy metabolism studies assembled across 3 research stations. The cows were categorized into low-, medium-, and high-FE groups according to residual feed intake (RFI), residual energy-corrected milk (RECM), and feed conversion efficiency (FCE). Mixed model regression was conducted to identify differences among the efficiency groups in animal and energy metabolism traits. Partial regression coefficients of both RFI and RECM agreed with published energy requirements more closely than cofficients derived from production experiments. Within RFI groups, efficient (Low-RFI) cows ate less, had a higher digestibility, produced less methane (CH₄) and heat, and had a higher efficiency of metabolizable energy (ME) utilization for milk production. High-RECM (most efficient) cows produced 6.0 kg/d more of energy-corrected milk (ECM) than their Low-RECM (least efficient) contemporaries at the same feed intake. They had a higher digestibility, produced less CH₄ and heat, and had a higher efficiency of ME utilization for milk production. The contributions of improved digestibility, reduced CH₄, and reduced urinary energy losses to increased ME intake at the same feed intake were 84, 12, and 4%, respectively. For both RFI and RECM analysis, increased metabolizability contributed to approximately 35% improved FE, with the remaining 65% attributed to the greater efficiency of utilization of ME. The analysis within RECM groups suggested that the difference in ME utilization was mainly due to the higher maintenance requirement of Low-RECM cows compared with Medium- and High-RECM cows, whereas the difference between Medium- and High-RECM cows resulted mainly from the higher efficiency of ME utilization for milk production in High-RECM cows. The main difference within FCE (ECM/DMI) categories was a greater (8.2 kg/d) ECM yield at the expense of mobilization in High-FCE cows compared with Low-FCE cows. Methane intensity (CH₄/ECM) was lower for efficient cows than for inefficient cows. The results indicated that RFI and RECM are different traits. We concluded that there is considerable variation in FE among cows that is not related to dilution of maintenance requirement or nutrient partitioning. Improving FE is a sustainable approach to reduce CH₄ production per unit of product, and at the same time improve the economics of milk production.

On the Search for Grazing Personalities: From Individual to Collective Behaviors

While grazing lands can offer a diverse range of forages, individuals within herds prefer to graze some habitats and not others. They can have consistent differences in grazing patterns and occupy specific spatial domains, whilst developing tactics and strategies for foraging that are specific to their grazing personalities. In this review, we explore the development of our understanding of grazing personalities, as we move away from the search for an "optimal animal" toward designing behavior-customized herds with an arrangement of individual grazing personalities that enhance ecosystem services and productivity. We present a "grazing personality model" that accounts for the personality of individual animals and for collective behaviors of herds. We argue that grazing personalities of grazing ruminants and other large herbivores are in part genetically determined, and that they can act at the individual and collective level. The social and biophysical environments as well as the emotional state of animals regulate the expression of "grazing genes" that are observed phenotypically as distinct grazing personalities. The reproductive and sexual successes of individuals and herds filter for allele variants of grazing genes and in turn determines their relative frequency. While the selection of one grazing personality may be adequate for homogeneous pastoral systems, the design of herds with a range of grazing personalities that are matched to the habitat diversity may be a better approach to improving the distribution of grazing animals, enhancing ecosystem services, and maximizing productivity.

FABP4 gene has a very large effect on feed efficiency in lactating Israeli Holstein cows

Improving feed efficiency (FE) is a major goal for the livestock industry. Previously, we have identified 48 SNP markers distributed over 32 genes significantly associated with residual feed intake (RFI) in Israeli Holstein male calves, the most significant of which are located in the bovine FABP4 gene. In the present study, we tested associations of eight of the FABP4 markers with RFI and feed conversion ratio (FCR), along with milk composition and feeding behavioral traits, in 114 lactating Israeli Holstein cows. Large allele effects were found, along with large contributions of FABP4 markers to the phenotypic variation [mean contribution of all significant markers (P < 0.05), 15.4 and 12.0% for RFI and FCR, respectively] and genotypic variation [means of all significant markers (P < 0.05), 75.7 and 32.4% in RFI and FCR, respectively]. However, the association of all significant FABP4 markers with FE and milk content traits was found in opposite directions, such that improved FE was accompanied by decreased milk content. Hence, before inclusion in breeding programs, the gain in FE must be economically balanced with the loss in milk contents. On the other hand, these findings imply that in any current improvement program concentrated on milk traits alone, without taking into account the effect on FE, the progress in milk composition is probably accompanied by deterioration of FE. These results, if confirmed in other populations and breeds, set FABP4 as a prime candidate in any marker-assisted selection program targeting FE as a whole and RFI in particular.

Hot topic: Selecting cattle for low residual feed intake did not affect daily methane production but increased methane yield

Reducing enteric methane (CH₄) production and improving feed conversion efficiency of dairy cows is of high importance. Residual feed intake (RFI) is one measure of feed efficiency, with low RFI animals being more efficient in feed conversion. Enteric CH₄ is an important source of digestible energy loss in ruminants and, because research in beef cattle has reported a positive relationship between RFI and daily CH₄ production, we hypothesized that low RFI dairy heifers, which are more feed efficient, would produce less CH₄/d. We measured the daily methane production (g of CH₄/d), methane yield [g of CH₄/kg of dry matter intake (DMI)], and CH₄ per kilogram of body weight (BW) gain for 56 heifers (20-22 mo old) in a 2 × 2 factorial arrangement: factors included 2 breeds (Holstein-Friesian and Jersey; n = 28/breed), with equal numbers of animals previously determined as being either high [+2.0 kg of dry matter (DM)/d] or low RFI (-2.1 kg of DM/d; n = 28/RFI category). All heifers were commingled and offered unrestricted access to the same diet of dried alfalfa cubes. Between RFI categories, heifers did not differ in BW or BW gain but low RFI heifers had 9.3 and 10.6% lower DMI and DMI/kg of BW, respectively, than high RFI heifers. Similarly, RFI category did not affect CH₄/d or CH₄/kg of BWg, but CH₄/kg of DMI was higher in low RFI heifers because of their lower DMI. These results might reflect more complete digestion of ingested feed in the more efficient, low RFI heifers, consistent with previous reports of greater apparent digestibility of organic matter. Holstein-Friesian heifers were heavier and consumed more total DM than Jersey heifers, but breed did not affect DMI/kg of BW or BWg. Jersey heifers produced less CH₄/d, but not CH₄/kg of DMI or CH₄/kg of BWg. We detected no interaction between breed and RFI category in any of the variables measured. In conclusion, differences in RFI in dairy heifers did not affect daily CH₄ production (g/d); however, low RFI heifers had a greater CH₄ yield (g/kg of DMI) on a high forage diet.

Use of genotype × environment interaction model to accommodate genetic heterogeneity for residual feed intake, dry matter intake, net energy in milk, and metabolic body weight in dairy cattle

Feed efficiency in dairy cattle has gained much attention recently. Due to the cost-prohibitive measurement of individual feed intakes, combining data from multiple countries is often necessary to ensure an adequate reference population. It may then be essential to model genetic heterogeneity when making inferences about feed efficiency or selecting efficient cattle using genomic information. In this study, we constructed a marker × environment interaction model that decomposed marker effects into main effects and interaction components that were specific to each environment. We compared environment-specific variance component estimates and prediction accuracies from the interaction model analyses, an across-environment analyses ignoring population stratification, and a within-environment analyses using an international feed efficiency data set. Phenotypes included residual feed intake, dry matter intake, net energy in milk, and metabolic body weight from 3,656 cows measured in 3 broadly defined environments: North America (NAM), the Netherlands (NLD), and Scotland (SAC). Genotypic data included 57,574 single nucleotide polymorphisms per animal. The interaction model gave the highest prediction accuracy for metabolic body weight, which had the largest estimated heritabilities ranging from 0.37 to 0.55. The within-environment model performed the best when predicting residual feed intake, which had the lowest estimated heritabilities ranging from 0.13 to 0.41. For traits (dry matter intake and net energy in milk) with intermediate estimated heritabilities (0.21 to 0.50 and 0.17 to 0.53, respectively), performance of the 3 models was comparable. Genomic correlations between environments also were computed using variance component estimates from the interaction model. Averaged across all traits, genomic correlations were highest between NAM and NLD, and lowest between NAM and SAC. In conclusion, the interaction model provided a novel way to evaluate traits measured in multiple environments in which genetic heterogeneity may exist. This model allowed estimation of environment-specific parameters and provided genomic predictions that approached or exceeded the accuracy of competing within- or across-environment models.

Divergence for residual feed intake of Holstein-Friesian cattle during growth did not affect production and reproduction during lactation

Residual feed intake (RFI) is the difference between actual and predicted dry matter intake (DMI) of individual animals. Recent studies with Holstein-Friesian calves have identified an ~20% difference in RFI during growth (calf RFI) and these groups remained divergent in RFI during lactation. The objective of the experiment described here was to determine if cows selected for divergent RFI as calves differed in milk production, reproduction or in the profiles of BW and body condition score (BCS) change during lactation, when grazing pasture. The cows used in the experiment (n=126) had an RFI of -0.88 and +0.75 kg DM intake/day for growth as calves (efficient and inefficient calf RFI groups, respectively) and were intensively grazed at four stocking rates (SR) of 2.2, 2.6, 3.1 and 3.6 cows/ha on self-contained farmlets, over 3 years. Each SR treatment had equal number of cows identified as low and high calf RFI, with 24, 28, 34 and 40/11 ha farmlet. The cows divergent for calf RFI were randomly allocated to each SR. Although SR affected production, calf RFI group (low or high) did not affect milk production, reproduction, BW, BCS or changes in these parameters throughout lactation. The most efficient animals (low calf RFI) lost similar BW and BCS as the least efficient (high calf RFI) immediately post-calving, and regained similar BW and BCS before their next calving. These results indicate that selection for RFI as calves to increase efficiency of feed utilisation did not negatively affect farm productivity variables (milk production, BCS, BW and reproduction) as adults when managed under an intensive pastoral grazing system.

FABP4 is a leading candidate gene associated with residual feed intake in growing Holstein calves

Ecological and economic concerns drive the need to improve feed utilization by domestic animals. Residual feed intake (RFI) is one of the most acceptable measures for feed efficiency (FE). However, phenotyping RFI-related traits is complex and expensive and requires special equipment. Advances in marker technology allow the development of various DNA-based selection tools. To assimilate these technologies for the benefit of RFI-based selection, reliable phenotypic measures are prerequisite. In the current study, we identified single nucleotide polymorphisms (SNPs) associated with RFI phenotypic consistency across different ages and diets (named RFI 1-3), using DNA samples of high or low RFI ranked Holstein calves. Using targeted sequencing of chromosomal regions associated with FE- and RFI-related traits, we identified 48 top SNPs significantly associated with at least one of three defined RFIs. Eleven of these SNPs were harbored by the fatty acid binding protein 4 (FABP4). While 10 significant SNPs found in FABP4 were common for RFI 1 and RFI 3, one SNP (FABP4_5; A<G substitution), in the promoter region of the gene, was significantly associated with all three RFIs. As the three RFI classes reflect changing diets and ages with concomitant RFI phenotypic consistency, the above polymorphisms and in particular FABP4_5, might be considered possible markers for RFI-based selection for FE in the Holstein breed, following a larger-scale validation.

Short communication: grazing pattern of dairy cows that were selected for divergent residual feed intake as calves

The aim of this study was to investigate and assess differences in the grazing pattern of 2 groups of mature dairy cows selected as calves for divergent residual feed intake (RFI). Sixteen Holstein-Friesian cows (471±31kg of body weight, 100 d in milk), comprising 8 cows selected as calves (6-8 mo old) for low (most efficient: CSCLowRFI) and 8 cows selected as calves for high (least efficient: CSCHighRFI) RFI, were used for the purpose of this study. Cows (n=16) were managed as a single group, and strip-grazed (24-h pasture allocation at 0800h) a perennial ryegrass sward for 31 d, with measurements taken during the last 21 d. All cows were equipped with motion sensors for the duration of the study, and jaw movements were measured for three 24-h periods during 3 random nonconsecutive days. Measurements included number of steps and jaw movements during grazing and rumination, plus fecal particle size distribution. Jaw movements were analyzed to identify bites, mastication (oral processing of ingesta) during grazing bouts, chewing during rumination, and to calculate grazing and rumination times for 24-h periods. Grazing and walking behavior were also analyzed in relation to the first meal of the day after the new pasture was allocated. Measured variables were subjected to multivariate analysis. Cows selected for low RFI as calves appeared to (a) prioritize grazing and rumination over idling; (b) take fewer steps, but with a higher proportion of grazing steps at the expense of nongrazing steps; and (c) increase the duration of the first meal and commenced their second meal earlier than CSCHighRFI. The CSCLowRFI had fewer jaw movements during eating (39,820 vs. 45,118 for CSCLowRFI and CSCHighRFI, respectively), more intense rumination (i.e., 5 more chews per bolus), and their feces had 30% less large particles than CSCHighRFI. These results suggest that CSCLowRFI concentrate their grazing activity to the time when fresh pasture is allocated, and graze more efficiently by walking and masticating less, hence they are more efficient grazers than CSCHighRFI.