Journal of Animal Science

Sanglard, Leticia P; Kuehn, Larry A; Snelling, Warren M; Spangler, Matthew L

doi: 10.1093/jas/skac059pmid: 35511236

Mitochondrial DNA copy number (mtDNA CN) has been shown to be highly heritable and associated with traits of interest in humans. However, studies are lacking in the literature for livestock species such as beef cattle. In this study, 2,371 individuals from a crossbred beef population comprising the Germplasm Evaluation program from the U.S. Meat Animal Research Center had samples of blood, leucocyte, or semen collected for low-pass sequencing (LPS) that resulted in both nuclear DNA (nuDNA) and mitochondrial DNA (mtDNA) sequence reads. Mitochondrial DNA CN was estimated based on the ratio of mtDNA to nuDNA coverages. Genetic parameters for mtDNA CN were estimated from an animal model based on a genomic relationship matrix (~87K SNP from the nuDNA). Different models were used to test the effects of tissue, sex, age at sample collection, heterosis, and breed composition. Maternal effects, assessed by fitting a maternal additive component and by fitting eleven SNP on the mtDNA, were also obtained. As previously reported, mtDNA haplotypes were used to classify individuals into Taurine haplogroups (T1, T2, T3/T4, and T5). Estimates of heritability when fitting fixed effects in addition to the intercept were moderate, ranging from 0.11 to 0.31 depending on the model. From a model ignoring contemporary group, semen samples had the lowest mtDNA CN, as expected, followed by blood and leucocyte samples (P ≤ 0.001). The effect of sex and the linear and quadratic effects of age were significant (P ≤ 0.02) depending on the model. When significant, females had greater mtDNA CN than males. The effects of heterosis and maternal heterosis were not significant (P ≥ 0.47). The estimates of maternal and mtDNA heritability were near zero (≤0.03). Most of the samples (98%) were classified as haplogroup T3. Variation was observed in the mtDNA within Taurine haplogroups, which enabled the identification of 24 haplotypes. These results suggest that mtDNA CN is under nuclear genetic control and would respond favorably to selection.

Stonecipher, Clinton A; Green, Ben T; Welch, Kevin D; Gardner, Dale R; Fritz, Scott A; Cook, Daniel; Pfister, James A

doi: 10.1093/jas/skac133pmid: 35419604

Abstract Larkspurs (Delphinium spp.) are native forbs that are poisonous to cattle and cost livestock producers millions of dollars in losses each year. Macro and micro minerals are required for normal functioning of essentially all metabolic processes in ruminants. The role that mineral status may play in larkspur poisoning in cattle is not clear. In this study, we seek to determine the effects a mineral-salt supplement, commonly used by cattle producers, to potentially reduce cattle losses to larkspur. The ability of mineral-salt supplementation to alter susceptibility to larkspur toxicosis was evaluated in a pen study. Animals supplemented with mineral-salt were found to be less susceptible to larkspur poisoning than the non-supplemented animals. A separate group of animals were then grazed on larkspur infested rangelands. One group was supplemented with a mineral-salt mix and the other group did not receive any mineral-salt. Supplementing cattle with the mineral-salt mix did not alter larkspur consumption (P > 0.05). However, overall larkspur consumption was low and averaged 3 ± 1.0% and 2 ± 1.1% for cattle supplemented with mineral and non-supplemented, respectively. Serum was collected from animals once a week during the grazing study. Average and maximum serum concentrations of toxic larkspur alkaloids were numerically higher in mineral-salt supplemented cattle compared with the non-supplemented animals. Results from the pen study suggest that a good mineral supplementation program will provide a protective effect for animals grazing in larkspur-infested ranges. The mineral-salt supplemented steers, in the grazing study, were not observed to consume less larkspur than the non-supplemented animals; however, the mineral-salt supplemented animals had higher concentrations of larkspur alkaloids in their serum indicating they may be able to tolerate higher larkspur consumption. The data also indicate that mineral-salt supplementation must be continuous throughout the time the animals are grazing these rangelands as the positive effects can be lost within 30 d post supplementation. Lay Summary Larkspurs (Delphinium spp.) are native forbs poisonous to cattle and cost livestock producers millions of dollars in losses each year. The role mineral status may play in larkspur poisoning in cattle is unclear. The ability of mineral-salt supplementation to alter susceptibility to larkspur toxicosis was evaluated in a pen and grazing study. In the pen study, animals supplemented with mineral-salt were found to be less susceptible to larkspur poisoning than non-supplemented animals. A separate group of animals grazed on larkspur infested rangelands. One group was supplemented with a mineral-salt mix and the other group did not receive any mineral-salt. Supplementing cattle with the mineral-salt mix did not alter larkspur consumption of grazing cattle. However, overall larkspur consumption was low. Results from the pen study suggest that a good mineral supplementation program will provide a protective effect for animals grazing in larkspur-infested ranges. The mineral-salt supplemented steers, in the grazing study, had higher concentrations of larkspur alkaloids in their blood serum indicating they may be able to tolerate higher larkspur consumption. The data also indicate that mineral-salt supplementation must be continuous throughout the time the animals are grazing as the positive effects can be lost within 30 days after supplementation. Introduction Tall forb plant communities are highly nutritious rangelands (Buchanan et al., 1972). Many tall forb plant communities in western North America contain larkspur (Delphinium spp.; Ralphs and Pfister, 1992). Larkspurs are native forbs that are poisonous to cattle and cost millions of dollars in losses each year due to animal deaths, increased management costs, and the underutilization of nutritious rangelands (Green et al., 2020). Alkaloids in larkspur cause neuromuscular paralysis from nicotinic acetylcholine receptor blockade at the postsynaptic neuromuscular junction (Dobelis et al., 1999; Green et al., 2013a). The net effect is muscular paralysis and often death. Macro and micro minerals are required for normal functioning of essentially all metabolic processes in ruminants. Dietary deficiencies, or excesses, of certain minerals can result in substantial economic losses in animal productivity (Spears and Weiss, 2014). Mineral deficiencies can result in reduced production potential, reduced immunity, and/or reduced reproductive efficiency. Minerals in excess can have negative effects, such as reduced feed intake and average daily gains (Engle and Spears, 2000). Minerals also play an important role in detoxification processes (Limaye et al., 2018; Rabbani et al., 2003). The role that an animal’s mineral status may play in larkspur poisoning in cattle is not clear. Some have postulated that cattle may eat larkspur because of higher mineral concentrations compared with other available forages (Logan, 1973; Knowles, 1974). Previous work has shown that mineral status does not affect the amount of larkspur ingested by grazing cattle under drought conditions, nor did dosing various concentrations of mineral intraruminally markedly alter ruminal fermentation patterns in free-grazing cattle (Pfister and Manners, 1991). There is no evidence that animals deficient in minerals will consume more or less larkspur. For example, Pfister and Manners (1995) reported that tall larkspur contained a high concentration of Na compared with other forages, but cattle, not supplemented with a mineral-salt, did not seek out and consume the larkspur. Ranchers managing cattle on rangelands with large populations of toxic larkspur are seeking solutions to the larkspur problem. There are anecdotal reports that mineral supplementation can reduce cattle losses to larkspur (Kay Dee Feed Company, 2013). Supplying grazing cattle with a mineral-salt supplement may help to meet mineral requirements of livestock (Sprinkle et al., 2006, 2018), and additionally help producers to reduce losses from larkspur. In this study, we seek to measure the effects of a mineral supplement commonly used by cattle producers to potentially reduce losses to larkspur with a pen-based laboratory study and a pasture-based grazing study. Materials and Methods All procedures were performed under veterinary supervision with the approval of the Utah State University Institutional Animal Care and Use Committee. Pen trial An exercise protocol developed by Green et al. (2019a) was used to measure larkspur intoxication in cattle. Briefly, cattle were dosed orally, one-time, with a bolus dose of larkspur with a known concentration of larkspur alkaloids. Larkspur alkaloids occur as 2 types: the N-(methylsuccinimido) anthranoyllycoctonine (MSAL) type, such as methyllycaconitine (MLA), and the non-MSAL-type alkaloids (Pfister et al., 1999; Panter et al., 2002). The MSAL-type alkaloids are highly toxic, whereas the non-MSAL alkaloids are much less toxic (Panter et al., 2002). Twenty-four hours after dosing, cattle were walked at 4 kph (kilometer per hour) for up to 40 min (walk time). As animals began to show signs of weakness (i.e., trembling and collapse), that animal would no longer be walked and time to collapse was recorded (Cook et al., 2011). As larkspur toxins affect muscle function (Green et al., 2019b), tolerance to exercise reveals the degree of intoxication, and provides a quantitative measurement of the poisoning. The exercise protocol was used to determine the response of animals supplemented with a mineral-salt or not supplemented to a larkspur challenge. Cattle were given an oral dose of a standardized amount of larkspur at 7.5 mg/kg BW (body weight) of MSAL-type alkaloids in the form of dried ground larkspur (D. barbeyi). At the start of the study, Angus steers (n = 12; 326 ± 13 kg BW) were orally dosed with the standardized amount of larkspur, prior to receiving the mineral supplement, to establish a baseline that represent walk times following larkspur treatment. The cattle were then split into 2 groups of 6 animals each. The groups either received a mineral-salt supplement or were not supplemented. One group of steers were taken through 4 ‘on-off’ cycles consisting of on-mineral/off-mineral and then back to on-mineral/off-mineral and the second group of steers were taken through 4 ‘off-on’ cycles consisting of off-mineral/on-mineral and then back to off-mineral/on-mineral, as a crossover design (Table 1). This first ‘on-off’ cycle consisted of 6 steers supplemented with the mineral-salt and 6 steers non-supplemented. Steers were switched for the second ‘on-off’ cycle with the 6 steers supplemented in the first cycle switched to non-supplemented and the non-supplemented steers in the first cycle switched to the mineral-salt supplement; animals continued to switch treatments as they went through the 4 ‘on-off’ cycles. The first 2 ‘on-off’ cycles represented period 1 so that all 12 steers were on mineral and all 12 steers were off mineral and the last 2 ‘on-off’ cycles represented period 2. The average ‘on-off’ cycle was 33 d; thus, an animal would be ‘on-mineral’ for 33 d, tested via larkspur dosing, then be ‘off-mineral’ for an additional 33 d before being tested again. Table 1. Mineral ‘on-off’ cycle of crossover design used in the pen trial. Steer group . Period 1 . Period 2 . Cycle . 1 . 2 . 3 . 4 . 1 (n = 6) On-mineral Off-mineral On-mineral Off-mineral 2 (n = 6) Off-mineral On-mineral Off-mineral On-mineral Steer group . Period 1 . Period 2 . Cycle . 1 . 2 . 3 . 4 . 1 (n = 6) On-mineral Off-mineral On-mineral Off-mineral 2 (n = 6) Off-mineral On-mineral Off-mineral On-mineral Open in new tab Table 1. Mineral ‘on-off’ cycle of crossover design used in the pen trial. Steer group . Period 1 . Period 2 . Cycle . 1 . 2 . 3 . 4 . 1 (n = 6) On-mineral Off-mineral On-mineral Off-mineral 2 (n = 6) Off-mineral On-mineral Off-mineral On-mineral Steer group . Period 1 . Period 2 . Cycle . 1 . 2 . 3 . 4 . 1 (n = 6) On-mineral Off-mineral On-mineral Off-mineral 2 (n = 6) Off-mineral On-mineral Off-mineral On-mineral Open in new tab This experiment used a 33-d period between larkspur doses to allow for the wash-out of any residual effects from larkspur. This length of time was based upon the results from previous toxicokinetic studies of larkspur in cattle (Green et al., 2009a, 2011, 2013b), and to allow animals to become adapted to being on- or off-mineral before receiving the next larkspur challenge. It was not the intent of the experiment to make animals deficient in mineral status but to deplete the ruminal mineral status over relatively short time spans (i.e., 33 d) as might happen under summer grazing conditions. Approximately 7 mL of blood was collected by venipuncture from steers immediately before oral dosing and then 24 h after dosing, immediately before walking. The blood was allowed to clot at room temperature, and the serum was separated from the blood cells by centrifugation at 2,000 × g for 20 min at 4 °C. The serum was then stored at −80 °C until chemical analysis. Daily mineral feeding protocol The mineral-salt supplement (Kaydets Western Range Special SM; Kay Dee Feed, Sioux City, IA, USA) was a granulated trace mineral mix and the micro- and macro-mineral nutrients in the mixture are listed in Table 2. Mineral supplement was fed daily to mineral-salt supplemented cattle, according to package instructions at 4 ounces/head/d at approximately 0730 hours. Each animal in the mineral-salt supplemented group were restrained in a stanchion while it consumed its daily mineral dose which took a maximum of about 15 min. After consuming the 4 ounces of mineral, steers were released and fed alfalfa hay. The non-supplemented steers did not have access to any mineral supplement. The on-mineral, off-mineral feeding pens remained the same throughout the study. The cattle were rotated between the pens depending on their mineral supplementation status for that portion of the study. All animals were fed alfalfa hay (Medicago sativa; 17% crude protein (CP), 45% neutral detergent fiber (NDF), 73% in vitro dry matter digestibility (IVDMD)) at 2.5% of BW/d prior to and during the study and had ad libitum access to fresh water. Table 2. Mineral composition (ppm)a and ingredientsb of ‘Kaydets Western Range Special SM’ mineral-salt mixturec. Element . Amount . Ag <0.1 Al 8,538 B 16 Ba 354 Ca 121,305 Co 52 Cr 37 Cu 1,636 Fe 7,996 K 9,985 Mg 22,875 Mn 5,645 Mo 7 Na 12,454 Ni 18 P 96,296 Sb 1 Se 30 Si 387 Sr 27 V 68 Zn 6,386 Element . Amount . Ag <0.1 Al 8,538 B 16 Ba 354 Ca 121,305 Co 52 Cr 37 Cu 1,636 Fe 7,996 K 9,985 Mg 22,875 Mn 5,645 Mo 7 Na 12,454 Ni 18 P 96,296 Sb 1 Se 30 Si 387 Sr 27 V 68 Zn 6,386 Mineral salt mixture was analyzed by Kansas State Veterinary Diagnostic Laboratory (Manhattan, KS). Ingredients in the mineral salt mixture as reported on the label: monocalcium phosphate, dicalcium phosphate, calcium carbonate, aluminosilicate, torula yeast dehydrated, molasses products, magnesium oxide, monosodium phosphate, potassium chloride, zinc amino acid chelate, copper amino acid chelate, manganese amino acid chelate, zinc oxide, sodium selenite, manganous oxide, vitamin A supplement, vitamin D3 supplement, vitamin E supplement selenium yeast, copper sulfate, mineral oil, colored with synthetic red iron oxide, ethylenediamine dihydriodide, choline chloride, magnesium mica, riboflavin supplement, niacin supplement, calcium D-pantothenate, Vitamin B12 supplement, folic acid, thiamine hydrochloride, pyridoxine hydrochloride, artificial flavors, and cobalt carbonate. Manufactured by Kay Dee Feed, Sioux City, IA, USA. Open in new tab Table 2. Mineral composition (ppm)a and ingredientsb of ‘Kaydets Western Range Special SM’ mineral-salt mixturec. Element . Amount . Ag <0.1 Al 8,538 B 16 Ba 354 Ca 121,305 Co 52 Cr 37 Cu 1,636 Fe 7,996 K 9,985 Mg 22,875 Mn 5,645 Mo 7 Na 12,454 Ni 18 P 96,296 Sb 1 Se 30 Si 387 Sr 27 V 68 Zn 6,386 Element . Amount . Ag <0.1 Al 8,538 B 16 Ba 354 Ca 121,305 Co 52 Cr 37 Cu 1,636 Fe 7,996 K 9,985 Mg 22,875 Mn 5,645 Mo 7 Na 12,454 Ni 18 P 96,296 Sb 1 Se 30 Si 387 Sr 27 V 68 Zn 6,386 Mineral salt mixture was analyzed by Kansas State Veterinary Diagnostic Laboratory (Manhattan, KS). Ingredients in the mineral salt mixture as reported on the label: monocalcium phosphate, dicalcium phosphate, calcium carbonate, aluminosilicate, torula yeast dehydrated, molasses products, magnesium oxide, monosodium phosphate, potassium chloride, zinc amino acid chelate, copper amino acid chelate, manganese amino acid chelate, zinc oxide, sodium selenite, manganous oxide, vitamin A supplement, vitamin D3 supplement, vitamin E supplement selenium yeast, copper sulfate, mineral oil, colored with synthetic red iron oxide, ethylenediamine dihydriodide, choline chloride, magnesium mica, riboflavin supplement, niacin supplement, calcium D-pantothenate, Vitamin B12 supplement, folic acid, thiamine hydrochloride, pyridoxine hydrochloride, artificial flavors, and cobalt carbonate. Manufactured by Kay Dee Feed, Sioux City, IA, USA. Open in new tab Grazing trial The study was conducted 20 km west of Island Park, Idaho (44.484°N, 111.666°W; 2,013-m elevation). Six pastures were established using electric fence encompassing an area with an abundant duncecap larkspur (D. occidentale) population. Each pasture was approximately 0.97 ha in size. The ecological site is classified as a loamy 16-22 Artemisia tridentata ssp. vaseyana/Festuca idahoensis-Pseudoroegneria spicata ssp. spicata site (USDA NRCS, 2021). Vegetation was dominated by Canada goldenrod (Solidago canadensis), duncecap larkspur, mountain big sagebrush, threetip sagebrush (A. tripartita), and silvery lupine (Lupinus argenteus). The major grasses present were thickspike wheatgrass (Elymus lanceolatus), bluebunch wheatgrass (Pseudoroegneria spicata), and plains reedgrass (Calamagrostis montanensis). Climatic data were extracted from a 4-km gridded PRISM dataset from 1937 to 2020 for a long-term average and also for the study period (PRISM, 2021). Twelve Angus steers were randomly assigned to one of two groups; mineral-salt supplemented (293 ± 13 kg BW; kg ± SE) and non-supplemented (312 ± 13 kg BW). Two animals from the same treatment group were then randomly allocated to each pasture (i.e., 3 replicate pastures per treatment). The mineral-salt supplemented group was started on the same mineral supplement, utilized in the pen trial, 29 d prior to grazing. For the first 7 d, the 6 steers were offered 3 ounces of mineral/steer/d, and animals were group fed, to acclimate them to the mineral without overeating according to the feeding directions supplied with the mineral. The remaining 22 d and throughout the grazing study the 6 steers were offered mineral free choice. Cattle grazed from July 8 to August 4, 2020. Larkspur was in the bud to pre-flower stage when grazing started, and pods were just starting to form as grazing ended. Prior to the grazing study animals were fed alfalfa hay (17% CP, 45% NDF, 73% IVDMD) at 2.5% of BW/d and had ad libitum access to fresh water. Forage availability and quality Forage availability was determined by clipping ten 0.5-m2 plots placed randomly throughout each pasture. Plots were clipped prior to grazing, every 7 d during the grazing study and at the end of grazing. Clipped material was separated into four vegetation categories: grasses, larkspur, other forbs, and lupine. Shrubs were ignored. Clipped material was dried (40 °C), weighed, and then composited to yield 5 representative samples from each pasture for nutritional analysis. Composite forage samples were ground to pass a 1-mm (millimeter) screen in a Wiley mill and analyzed for nitrogen content (AOAC, 1995) using a Leco FP-528 Nitrogen Analyzer (Leco Corp., St. Joseph, MI) and CP was determined by multiplying nitrogen content by 6.25. A 2-stage method was used to determine IVDMD, with the first stage consisting of a 48-h in vitro fermentation in an ANKOM Daisy II incubator (ANKOM Technology, Macedon, NY). Analyses of the second stage of IVDMD and of NDF were made using procedures modified for use in an ANKOM-200 Fiber Analyzer (ANKOM Technology). Animal diet selection Animals were acclimated to the presence of an observer for a 10-d period prior to starting the study which allowed close observation (1-2 m) without disturbing grazing. Cattle were placed in a corral at night around dusk and released to graze each day about 0700 hours. Daily bite counts were used to determine animal diet composition (Ortega et al., 1995). Each animal was focally observed (Altmann, 1974) in a predetermined order for a 5-min period before observing the next steer. Generally, about 25 min of daily observation time on each animal was obtained during active grazing periods throughout the day. Bites were recorded by forage class (larkspur, grasses, other forbs, and lupine). The percentage of each forage class in the diet was calculated based on the total number of bites. Animals were allowed ad libitum access to water. Blood serum collection The number of incidences in which animals were visibly intoxicated (i.e., muscle tremors, periodic collapse) was noted. In some instances, visibly intoxicated animals that became laterally recumbent had blood drawn via jugular venipuncture and were then given intravenous injections of the cholinesterase inhibitor neostigmine at 0.04 mg/kg BW (Green et al., 2009b); these animals typically became ambulatory within 15 to 20 min after injection. Animals that did not become laterally recumbent, but did show visible signs of intoxication, were left alone and blood was not drawn. Every 7 d, cattle were restrained in a chute and approximately 7 mL of blood was collected via jugular venipuncture. The blood was allowed to clot at ambient temperature, and the serum was separated from the red blood cells, the same day, by centrifugation at 2,000 × g (times gravity) for 20 min at 4 °C. The serum was then stored at −80 °C until chemical analysis. Larkspur and serum alkaloids Ten individual larkspur plants were collected outside the pastures, on the same day’s plots were clipped, to determine alkaloid concentrations. Whole plants were dried (40 °C) and ground to pass through a 1-mm screen. The concentration of MSAL alkaloids and total alkaloids were determined as previously described using Fourier transform infrared spectroscopy (Gardner et al., 1997). The non-MSAL alkaloid concentration was calculated by difference from the total alkaloid concentration and the concentration of MSAL-type alkaloids (Gardner and Pfister, 2009). Serum samples were analyzed for the major toxic MSAL-type alkaloid, MLA. Matrix-matched standards were prepared for MLA as follows: a stock solution of MLA was prepared at 1 mg/mL in ethanol and then 0.080 mL was diluted with 0.920 mL of ethanol to provide an 80 microgram (µg)/mL solution. A 0.050 mL aliquot was added to 1.950 mL of blank bovine sera and serially diluted with bovine sera to give matrix standards at 2,000, 1,000, 500, 250, 125, 62, 31, 16, 8, and 4 ng/mL MLA. Serum samples were thawed, vortexed, and then centrifuged for 5 min at 16,000 × g for 20 min at 20 °C. For the sera samples, and the matrix standards, a 0.5-mL aliquot was taken and placed in a 1.5-mL Eppendorf tube. An equal volume of acetonitrile (0.5 mL) containing 250 ng/mL reserpine (internal standard) was added to each sample. Samples were vortexed for 10 to 15 s and then centrifuged at 16,000 × g for 20 min at 20 °C. A 0.20-mL aliquot was then removed to a 0.3-mL autosample vial for analysis. Concentrations of MLA in sera were measured as described by Welch et al. (2015) with the following modifications using a Velos PRO LTQ (Thermo Scientific, Waltham, MA) MS (mass spectrometer) coupled with an Agilent 1260 autosampler, and MS binary pump (Agilent Technologies, Santa Clara, CA) used in-line with a Hypersil Gold C18 column (100 × 2.1 mm) with a guard column of equivalent phase. The column was eluted with a binary solvent gradient using 0.1 % formic acid (solvent A) and acetonitrile (solvent B) at a flow rate of 0.400 mL/min and the following gradient mixture with time: 5% to 50% B (5 min); 50% to 95% B (5 to 6 min); 95% B (6 to 8 min); 95% to 5% B (8 to 9 min); 5% B (9 to 13 min). The flow from the column was connected to a heated electrospray ion source. The MS was set to scan selected MS/MS experiments for MLA parent ion mass-to-charge ratio 683.3 with collision-induced dissociation fragmentation power of 33. Reconstructed ion chromatograms used the following selected ions for methyllycaconitine (619.3, 651.3, and 665.3) and for reserpine (397.1 and 448.2). Quantitation was completed using peak areas from reconstructed ion chromatograms for MLA versus reserpine. Mineral analysis The mineral-salt supplement and serum samples from both the pen and grazing trials were sent to Kansas State Veterinary Diagnostic Laboratory (Manhattan, KS) for an extended trace mineral panel analysis using Inductively Coupled Plasma-Mass Spectrometry. Statistical analysis For the pen-trail, walk times, serum MLA concentrations, and serum mineral concentrations were analyzed using a two-treatment (mineral supplemented versus non-supplemented), 2-period crossover design with a repeated measures ANOVA in Prism version 9.1.1 (GraphPad Software, San Diego, CA) as described (Graphpad Software, 2021). A crossover design was used because it is a repeated measures experiment such that each experimental unit (steer) received different treatments (i.e., mineral or no mineral) during the different measurement cycles. In summary, each ‘on-off’ cycle is a measurement cycle, and each animal serves as its own control. Four mineral ‘on-off cycles’, representing an on- or off-mineral cycle of 33 d, were completed with the first two ‘on-off’ cycles representing period 1 and the last two ‘on-off’ cycles representing period 2. For example, six animals were on-mineral and six animals off-mineral for the first cycle and then each group of 6 animals were switched for the second ‘on-off’ cycle, representing period 1. Previous research with Angus steers orally dosed the same plant collection showed that 7 d was required to clear 99% of the toxic alkaloids in larkspur (Green et al., 2011). The 33 d between each oral dosing was enough time for the larkspur alkaloids to clear from the bodies of the steers, and statistical sphericity was therefore assumed for the analysis. For the grazing study, bite count data were analyzed using the GLIMMIX (generalized linear mixed model) procedure in SAS (v.9.4, SAS Institute Inc., Cary, NC). The number of bites of each forage class (larkspur, grass, and other forbs) was calculated as a percentage of the diet and daily means were calculated for each animal’s diet selection. The model for diet variables included treatment (mineral supplemented versus non-supplemented), date, treatment × date interaction, pasture nested within treatment, animals nested within treatment and pasture, and date × pasture within treatment. Pasture, animal, and date were random factors in the model. Serum MLA and serum mineral concentrations were analyzed using the GLIMMIX procedure in SAS. The model for serum concentrations included treatment (mineral supplemented versus non-supplemented), date, and treatment × date interaction. Date was the random factor in the model. Least squares means were used for all comparisons and adjusted for type I error inflation using the Tukey method. Only means ± SE were calculated for data on forage availability, nutritional analysis of forage, and the alkaloid concentration in duncecap larkspur plant material and serum collected from animals that were intoxicated from larkspur poisoning and became lateral recumbent. The significance level for all statistical comparisons was set at P < 0.05. Results and Discussion Pen trial Angus steers were administered an oral dose of dried ground larkspur (D. barbeyi), at the start of the study to establish a baseline and again following each on- or off- mineral cycle of 33 d. Four mineral ‘on-off’ cycles were completed with the first two ‘on-off’ cycles representing period 1 and the last two ‘on-off’ cycles representing period 2 (Table 1). The walk times of steers supplemented with mineral-salt or non-supplemented are shown in Figure 1. Baseline walk times, that represent walk times following larkspur treatment prior to any animals being supplemented with the mineral-salt supplement, did not differ between groups (14 ± 7 and 16 ± 6 min, respectively; P = 0.8176). There was a difference in treatment walk times between steers supplemented with mineral-salt and non-supplemented (P = 0.0409). The supplemented steers (n = 24 doses) walked an average of 24 ± 3 min. The non-supplemented steers walk times were shorter and averaged 15 ± 3 min of walking (n = 24 doses). Larkspur alkaloids cause muscle weakness, so a longer walk time suggests a greater resistance to the toxic effects of larkspur alkaloids (Green et al., 2019a). Figure 1. Open in new tabDownload slide Walk times of Angus cattle 24 h after they received an oral dose of 7.5 mg/kg BW MSAL-type alkaloids in the form of dried ground larkspur (Delphinium barbeyi). As animals began to show signs of weakness (i.e.., trembling and collapse), that animal would no longer be walked and time to collapse was recorded. The mean ± standard error of walk times of cattle split into 2 groups of 6 each and taken through 4 ‘on-off’ cycles of mineral/no mineral with an average ‘on-off’ cycle of 33 ± 1 d as a repeated measure, crossover design. Period 1 is represented by the first 2 ‘on-off’ cycles (n = 12) and Period 2 is the last 2 ‘on-off’ cycles (n = 12). The on-mineral cycle steers received 4 ounce/head/d of Western Range Special SM (Kay Dee Feed LLC, Sioux City, IA, USA) and then were cycled to off-mineral for 33 ± 1 d. There was a significant difference between the on-mineral and off-mineral walk times (P = 0.0409). Figure 1. Open in new tabDownload slide Walk times of Angus cattle 24 h after they received an oral dose of 7.5 mg/kg BW MSAL-type alkaloids in the form of dried ground larkspur (Delphinium barbeyi). As animals began to show signs of weakness (i.e.., trembling and collapse), that animal would no longer be walked and time to collapse was recorded. The mean ± standard error of walk times of cattle split into 2 groups of 6 each and taken through 4 ‘on-off’ cycles of mineral/no mineral with an average ‘on-off’ cycle of 33 ± 1 d as a repeated measure, crossover design. Period 1 is represented by the first 2 ‘on-off’ cycles (n = 12) and Period 2 is the last 2 ‘on-off’ cycles (n = 12). The on-mineral cycle steers received 4 ounce/head/d of Western Range Special SM (Kay Dee Feed LLC, Sioux City, IA, USA) and then were cycled to off-mineral for 33 ± 1 d. There was a significant difference between the on-mineral and off-mineral walk times (P = 0.0409). Serum was collected 24 h after cattle were orally dosed with larkspur. Serum MLA concentrations differed between cattle supplemented with mineral-salt and non-supplemented (P = 0.0396; Figure 2). Baseline serum MLA concentrations, prior to animals receiving any mineral-salt, were 310 ± 49 ng/mL for all animals (n = 12). Serum MLA concentrations of steers supplemented with mineral-salt for the first and second period were 407 ± 32 and 416 ± 44 ng/mL, respectively. Serum MLA concentrations of non-supplemented steers during the first and second periods were 317 ± 28 and 525 ± 76 ng/mL, respectively. Figure 2. Open in new tabDownload slide Steer serum MLA concentrations (ng/mL) at 24 h after they received an oral dose of 7.5 mg/kg BW MSAL-type alkaloids in the form of dried ground larkspur (Delphinium barbeyi). The mean ± standard error of serum MLA concentrations in ng/mL of cattle split into 2 groups of 6 each and taken through 4 ‘on-off’ cycles of mineral/no mineral with an average on-off cycle of 33 ± 1 d as a repeated measures, crossover design. Period 1 is represented by the first 2 ‘on-off’ cycles (n = 12) and Period 2 is the last 2 ‘on-off’ cycles (n = 12). The on-mineral cycle steers received 4 ounce/head/d of Western Range Special SM (Kay Dee Feed LLC, Sioux City, IA, USA) and then were cycled to off-mineral for 33 ± 1 d. There was a significant difference between the on-mineral and off-mineral serum MLA concentrations (P = 0.0396). Figure 2. Open in new tabDownload slide Steer serum MLA concentrations (ng/mL) at 24 h after they received an oral dose of 7.5 mg/kg BW MSAL-type alkaloids in the form of dried ground larkspur (Delphinium barbeyi). The mean ± standard error of serum MLA concentrations in ng/mL of cattle split into 2 groups of 6 each and taken through 4 ‘on-off’ cycles of mineral/no mineral with an average on-off cycle of 33 ± 1 d as a repeated measures, crossover design. Period 1 is represented by the first 2 ‘on-off’ cycles (n = 12) and Period 2 is the last 2 ‘on-off’ cycles (n = 12). The on-mineral cycle steers received 4 ounce/head/d of Western Range Special SM (Kay Dee Feed LLC, Sioux City, IA, USA) and then were cycled to off-mineral for 33 ± 1 d. There was a significant difference between the on-mineral and off-mineral serum MLA concentrations (P = 0.0396). In the pen study, serum mineral concentrations of mineral-salt supplemented and non-supplemented treatment groups for each period are listed in Table 3. There was no period × treatment interaction (P > 0.05) for serum macro-mineral concentrations (Ca, K, Mg, Na, and P) in the pen study. There was a period × treatment interaction for serum micro-minerals Co, Mn, Mo, and Zn (P < 0.01; Table 3). Serum concentrations of Co increased from Period 1 to Period 2 in the mineral-salt supplemented group. Serum concentrations of Co were similar between periods in the non-supplemented group and concentrations were lower than the mineral-salt supplemented group at both periods. Serum concentrations of Mn and Mo followed similar patterns with concentrations the greatest in the non-supplemented group in Period 1. Serum concentrations in Period 2 for the non-supplemented group were similar to both periods in the mineral-salt supplemented group. Serum concentrations of Zn were greater in Period 2 for the mineral-salt supplemented group and Period 1 of the non-supplemented group. Period 1 of the non-supplemented group was similar to both periods of the mineral-salt supplemented group. Liver storage of minerals from animals previously fed mineral in the cross over design could have influenced the results of serum mineral concentrations. Table 3. Serum mineral concentration of steers supplemented with a mineral-salt and non-supplemented steersin the pen feeding trial Mineral . . Mineral . No Mineral . Baseline1 . Period 1 . Period 2 . Period 1 . Period 2 . Al 0.04 ± 0.012 0.06 ± 0.008 0.05 ± 0.009 0.09 ± 0.028 0.04 ± 0.005 B 0.79 ± 0.043 0.91 ± 0.043 0.88 ± 0.033 0.84 ± 0.033 0.93 ± 0.051 Ba 0.18 ± 0.005 0.18 ± 0.007 0.18 ± 0.008 0.21 ± 0.007 0.18 ± 0.007 Ca 59.88 ± 0.74 70.41 ± 1.80 75.3 ± 1.88 68.75 ± 1.524 77.1 ± 2.93 Co 0.25 ± 0.13 1.0 ± 0.12b 1.8 ± 0.22a 0.2 ± 0.11c 0.08 ± 0.08c Cr 0.015 ± 0.0028 0.017 ± 0.0039 0.006 ± 0.001 0.02 ± 0.003 0.005 ± 0 Cu 0.92 ± 0.027 0.77 ± 0.045 0.82 ± 0.024 0.72 ± 0.028 0.89 ± 0.063 Fe 2.85 ± 0.17 3.23 ± 0.395 2.63 ± 0.226 3.25 ± 0.237 2.8 ± 0.25 K 152.6 ± 3.03 208.7 ± 16.13 280.7 ± 10.602 191.1 ± 6.65 305.4 ± 14.25 Mg 33.19 ± 0.82 31.8 ± 1.18 36.92 ± 1.926 34.9 ± 1.01 34.1 ± 1.89 Mn 4.5 ± 0.47 5.6 ± 0.69b 4.6 ± 0.26b 8.9 ± 0.69a 4.3 ± 0.33b Mo 75.4 ± 11.37 29.7 ± 4.11b 11.0 ± 0.62b 105.3 ± 19.9a 21.0 ± 4.0b Na 2,820 ± 44.6 3,580 ± 137.1 3,788 ± 91.4 3,388 ± 84.4 3,794 ± 125.0 Ni 0.9 ± 0.08 1.3 ± 0.26 7.1 ± 5.36 2.3 ± 0.39 2.3 ± 0.28 P 105.8 ± 3.18 131.9 ± 5.10 131.0 ± 3.83 110.7 ± 3.52 108.8 ± 5.43 Se 0.06 ± 0.003 0.12 ± 0.007 0.16 ± 0.005 0.06 ± 0.005 0.08 ± 0.007 Si 5.76 ± 0.14 6.26 ± 0.164 7.1 ± 0.283 7.06 ± 0.168 7.61 ± 0.318 Sr 0.18 ± 0.005 0.16 ± 0.009 0.15 ± 0.006 0.17 ± 0.007 0.17 ± 0.008 Zn 1.08 ± 0.032 1.08 ± 0.051bc 1.27 ± 0.080a 1.2 ± 0.064ab 0.94 ± 0.062c Mineral . . Mineral . No Mineral . Baseline1 . Period 1 . Period 2 . Period 1 . Period 2 . Al 0.04 ± 0.012 0.06 ± 0.008 0.05 ± 0.009 0.09 ± 0.028 0.04 ± 0.005 B 0.79 ± 0.043 0.91 ± 0.043 0.88 ± 0.033 0.84 ± 0.033 0.93 ± 0.051 Ba 0.18 ± 0.005 0.18 ± 0.007 0.18 ± 0.008 0.21 ± 0.007 0.18 ± 0.007 Ca 59.88 ± 0.74 70.41 ± 1.80 75.3 ± 1.88 68.75 ± 1.524 77.1 ± 2.93 Co 0.25 ± 0.13 1.0 ± 0.12b 1.8 ± 0.22a 0.2 ± 0.11c 0.08 ± 0.08c Cr 0.015 ± 0.0028 0.017 ± 0.0039 0.006 ± 0.001 0.02 ± 0.003 0.005 ± 0 Cu 0.92 ± 0.027 0.77 ± 0.045 0.82 ± 0.024 0.72 ± 0.028 0.89 ± 0.063 Fe 2.85 ± 0.17 3.23 ± 0.395 2.63 ± 0.226 3.25 ± 0.237 2.8 ± 0.25 K 152.6 ± 3.03 208.7 ± 16.13 280.7 ± 10.602 191.1 ± 6.65 305.4 ± 14.25 Mg 33.19 ± 0.82 31.8 ± 1.18 36.92 ± 1.926 34.9 ± 1.01 34.1 ± 1.89 Mn 4.5 ± 0.47 5.6 ± 0.69b 4.6 ± 0.26b 8.9 ± 0.69a 4.3 ± 0.33b Mo 75.4 ± 11.37 29.7 ± 4.11b 11.0 ± 0.62b 105.3 ± 19.9a 21.0 ± 4.0b Na 2,820 ± 44.6 3,580 ± 137.1 3,788 ± 91.4 3,388 ± 84.4 3,794 ± 125.0 Ni 0.9 ± 0.08 1.3 ± 0.26 7.1 ± 5.36 2.3 ± 0.39 2.3 ± 0.28 P 105.8 ± 3.18 131.9 ± 5.10 131.0 ± 3.83 110.7 ± 3.52 108.8 ± 5.43 Se 0.06 ± 0.003 0.12 ± 0.007 0.16 ± 0.005 0.06 ± 0.005 0.08 ± 0.007 Si 5.76 ± 0.14 6.26 ± 0.164 7.1 ± 0.283 7.06 ± 0.168 7.61 ± 0.318 Sr 0.18 ± 0.005 0.16 ± 0.009 0.15 ± 0.006 0.17 ± 0.007 0.17 ± 0.008 Zn 1.08 ± 0.032 1.08 ± 0.051bc 1.27 ± 0.080a 1.2 ± 0.064ab 0.94 ± 0.062c Baseline values are from the start of the study before any animals were given the salt-mineral supplement. Means with different subscripts within a row differ (period × treatment interaction; P < 0.05). All mineral concentration values are ppm except Co, Mn, Mo, and Ni which are ppb. Useful conversions: 1 µg/mL = mg/L = 1 ppm; 1 ng/mL = µg/L = ppb. Open in new tab Table 3. Serum mineral concentration of steers supplemented with a mineral-salt and non-supplemented steersin the pen feeding trial Mineral . . Mineral . No Mineral . Baseline1 . Period 1 . Period 2 . Period 1 . Period 2 . Al 0.04 ± 0.012 0.06 ± 0.008 0.05 ± 0.009 0.09 ± 0.028 0.04 ± 0.005 B 0.79 ± 0.043 0.91 ± 0.043 0.88 ± 0.033 0.84 ± 0.033 0.93 ± 0.051 Ba 0.18 ± 0.005 0.18 ± 0.007 0.18 ± 0.008 0.21 ± 0.007 0.18 ± 0.007 Ca 59.88 ± 0.74 70.41 ± 1.80 75.3 ± 1.88 68.75 ± 1.524 77.1 ± 2.93 Co 0.25 ± 0.13 1.0 ± 0.12b 1.8 ± 0.22a 0.2 ± 0.11c 0.08 ± 0.08c Cr 0.015 ± 0.0028 0.017 ± 0.0039 0.006 ± 0.001 0.02 ± 0.003 0.005 ± 0 Cu 0.92 ± 0.027 0.77 ± 0.045 0.82 ± 0.024 0.72 ± 0.028 0.89 ± 0.063 Fe 2.85 ± 0.17 3.23 ± 0.395 2.63 ± 0.226 3.25 ± 0.237 2.8 ± 0.25 K 152.6 ± 3.03 208.7 ± 16.13 280.7 ± 10.602 191.1 ± 6.65 305.4 ± 14.25 Mg 33.19 ± 0.82 31.8 ± 1.18 36.92 ± 1.926 34.9 ± 1.01 34.1 ± 1.89 Mn 4.5 ± 0.47 5.6 ± 0.69b 4.6 ± 0.26b 8.9 ± 0.69a 4.3 ± 0.33b Mo 75.4 ± 11.37 29.7 ± 4.11b 11.0 ± 0.62b 105.3 ± 19.9a 21.0 ± 4.0b Na 2,820 ± 44.6 3,580 ± 137.1 3,788 ± 91.4 3,388 ± 84.4 3,794 ± 125.0 Ni 0.9 ± 0.08 1.3 ± 0.26 7.1 ± 5.36 2.3 ± 0.39 2.3 ± 0.28 P 105.8 ± 3.18 131.9 ± 5.10 131.0 ± 3.83 110.7 ± 3.52 108.8 ± 5.43 Se 0.06 ± 0.003 0.12 ± 0.007 0.16 ± 0.005 0.06 ± 0.005 0.08 ± 0.007 Si 5.76 ± 0.14 6.26 ± 0.164 7.1 ± 0.283 7.06 ± 0.168 7.61 ± 0.318 Sr 0.18 ± 0.005 0.16 ± 0.009 0.15 ± 0.006 0.17 ± 0.007 0.17 ± 0.008 Zn 1.08 ± 0.032 1.08 ± 0.051bc 1.27 ± 0.080a 1.2 ± 0.064ab 0.94 ± 0.062c Mineral . . Mineral . No Mineral . Baseline1 . Period 1 . Period 2 . Period 1 . Period 2 . Al 0.04 ± 0.012 0.06 ± 0.008 0.05 ± 0.009 0.09 ± 0.028 0.04 ± 0.005 B 0.79 ± 0.043 0.91 ± 0.043 0.88 ± 0.033 0.84 ± 0.033 0.93 ± 0.051 Ba 0.18 ± 0.005 0.18 ± 0.007 0.18 ± 0.008 0.21 ± 0.007 0.18 ± 0.007 Ca 59.88 ± 0.74 70.41 ± 1.80 75.3 ± 1.88 68.75 ± 1.524 77.1 ± 2.93 Co 0.25 ± 0.13 1.0 ± 0.12b 1.8 ± 0.22a 0.2 ± 0.11c 0.08 ± 0.08c Cr 0.015 ± 0.0028 0.017 ± 0.0039 0.006 ± 0.001 0.02 ± 0.003 0.005 ± 0 Cu 0.92 ± 0.027 0.77 ± 0.045 0.82 ± 0.024 0.72 ± 0.028 0.89 ± 0.063 Fe 2.85 ± 0.17 3.23 ± 0.395 2.63 ± 0.226 3.25 ± 0.237 2.8 ± 0.25 K 152.6 ± 3.03 208.7 ± 16.13 280.7 ± 10.602 191.1 ± 6.65 305.4 ± 14.25 Mg 33.19 ± 0.82 31.8 ± 1.18 36.92 ± 1.926 34.9 ± 1.01 34.1 ± 1.89 Mn 4.5 ± 0.47 5.6 ± 0.69b 4.6 ± 0.26b 8.9 ± 0.69a 4.3 ± 0.33b Mo 75.4 ± 11.37 29.7 ± 4.11b 11.0 ± 0.62b 105.3 ± 19.9a 21.0 ± 4.0b Na 2,820 ± 44.6 3,580 ± 137.1 3,788 ± 91.4 3,388 ± 84.4 3,794 ± 125.0 Ni 0.9 ± 0.08 1.3 ± 0.26 7.1 ± 5.36 2.3 ± 0.39 2.3 ± 0.28 P 105.8 ± 3.18 131.9 ± 5.10 131.0 ± 3.83 110.7 ± 3.52 108.8 ± 5.43 Se 0.06 ± 0.003 0.12 ± 0.007 0.16 ± 0.005 0.06 ± 0.005 0.08 ± 0.007 Si 5.76 ± 0.14 6.26 ± 0.164 7.1 ± 0.283 7.06 ± 0.168 7.61 ± 0.318 Sr 0.18 ± 0.005 0.16 ± 0.009 0.15 ± 0.006 0.17 ± 0.007 0.17 ± 0.008 Zn 1.08 ± 0.032 1.08 ± 0.051bc 1.27 ± 0.080a 1.2 ± 0.064ab 0.94 ± 0.062c Baseline values are from the start of the study before any animals were given the salt-mineral supplement. Means with different subscripts within a row differ (period × treatment interaction; P < 0.05). All mineral concentration values are ppm except Co, Mn, Mo, and Ni which are ppb. Useful conversions: 1 µg/mL = mg/L = 1 ppm; 1 ng/mL = µg/L = ppb. Open in new tab Serum concentrations of Ba (P = 0.0002), Mn (P < 0.0001), Mo (P < 0.0001), Si (P = 0.0007), and Sr (P = 0.0002) displayed a treatment effect in which serum mineral concentrations were greater in the non-supplemented group. The mineral-salt supplemented group had greater serum concentrations of Co (P < 0.0001), P (P < 0.0001), and Se (P < 0.0001). The only macro-mineral that displayed differences in concentrations between treatment groups was P and serum concentrations were greatest in the mineral-salt supplemented group which is the group that also had the highest walk times. The results of the pen study indicate that daily consumption of 4 ounces of mineral can increase the walk time of Angus steers given a sublethal, but toxic dose of larkspur. This suggests that daily mineral consumption has the potential to reduce the toxic effects of larkspur in cattle grazing on larkspur containing rangelands. Grazing trial Weather A summary of the weather for the grazing study location is given in Table 4. Precipitation from January to July, prior to the study, was slightly higher than the long-term average for the study site (Table 4). Precipitation during the grazing study in July was below the long-term average for July. The majority of the July precipitation came at the end of the month (July 23-31); however, if precipitation from the early days in July, before the grazing study started, were combined for the month then the precipitation values are similar to the long-term average. Table 4. Weather variables during the grazing study (July 8 to August 4, 2020) in southeastern Idaho1 Timeline . Precipitation (mm)2 . Temperature (°C)3 . Study year4 . Long-term5 . Study year4 . Long-term5 . January to July 449 443 1.1 −1.9 July 38.5 38 14.8 15.6 August 0.33 33 17.1 15.1 Timeline . Precipitation (mm)2 . Temperature (°C)3 . Study year4 . Long-term5 . Study year4 . Long-term5 . January to July 449 443 1.1 −1.9 July 38.5 38 14.8 15.6 August 0.33 33 17.1 15.1 Weather measurements were extracted from a 4-km gridded PRISM dataset (PRISM, 2021). Total precipitation (mm) for that period. Average temperature (°C) for that period. Study year is 2020. Long-term timeline is from 1937 to 2020. Open in new tab Table 4. Weather variables during the grazing study (July 8 to August 4, 2020) in southeastern Idaho1 Timeline . Precipitation (mm)2 . Temperature (°C)3 . Study year4 . Long-term5 . Study year4 . Long-term5 . January to July 449 443 1.1 −1.9 July 38.5 38 14.8 15.6 August 0.33 33 17.1 15.1 Timeline . Precipitation (mm)2 . Temperature (°C)3 . Study year4 . Long-term5 . Study year4 . Long-term5 . January to July 449 443 1.1 −1.9 July 38.5 38 14.8 15.6 August 0.33 33 17.1 15.1 Weather measurements were extracted from a 4-km gridded PRISM dataset (PRISM, 2021). Total precipitation (mm) for that period. Average temperature (°C) for that period. Study year is 2020. Long-term timeline is from 1937 to 2020. Open in new tab Forage availability and quality Forage availability was collected at the start of the grazing study and every seven d during the study (Table 5). Perennial grasses, other forbs and larkspur were abundant throughout the study with more than 500 kg/ha of each forage class available at each collection time with the exception of larkspur on July 28. Forage availability in this study is similar to what was reported by Pfister et al. (2018) over three years of grazing in July 2015 to 2017. There was ample forage available so that cattle were not forced to graze larkspur. The stocking rate (2 animals per pasture) was similar to that reported by Pfister et al. (2018). They reported abundant perennial grasses, other forbs, and duncecap larkspur available during the grazing periods which is similar to the current study. Table 5. Forage availability (kg/ha ± SE) during grazing study conducted in southeastern Idaho during 2020 Date . Grass . Other forbs1 . Larkspur . July 72 797 ± 86 740 ± 54 512 ± 74 July 14 1082 ± 103 839 ± 66 511 ± 55 July 21 628 ± 76 957 ± 76 507 ± 105 July 28 855 ± 84 891 ± 57 418 ± 45 August 4 497 ± 68 774 ± 71 596 ± 111 Date . Grass . Other forbs1 . Larkspur . July 72 797 ± 86 740 ± 54 512 ± 74 July 14 1082 ± 103 839 ± 66 511 ± 55 July 21 628 ± 76 957 ± 76 507 ± 105 July 28 855 ± 84 891 ± 57 418 ± 45 August 4 497 ± 68 774 ± 71 596 ± 111 All other forbs except for larkspur. Shrubs were not included in forage availability measurements. Average of 6 pastures on each date. Open in new tab Table 5. Forage availability (kg/ha ± SE) during grazing study conducted in southeastern Idaho during 2020 Date . Grass . Other forbs1 . Larkspur . July 72 797 ± 86 740 ± 54 512 ± 74 July 14 1082 ± 103 839 ± 66 511 ± 55 July 21 628 ± 76 957 ± 76 507 ± 105 July 28 855 ± 84 891 ± 57 418 ± 45 August 4 497 ± 68 774 ± 71 596 ± 111 Date . Grass . Other forbs1 . Larkspur . July 72 797 ± 86 740 ± 54 512 ± 74 July 14 1082 ± 103 839 ± 66 511 ± 55 July 21 628 ± 76 957 ± 76 507 ± 105 July 28 855 ± 84 891 ± 57 418 ± 45 August 4 497 ± 68 774 ± 71 596 ± 111 All other forbs except for larkspur. Shrubs were not included in forage availability measurements. Average of 6 pastures on each date. Open in new tab Forage quality was high at the start of the study and decreased over the season as plants matured (Table 6). Crude protein content of perennial grasses was 10% at the start of the study and remained above 9% throughout July and then dropped to 7% at the end of the study in August. The NDF content of perennial grasses remained below 66% throughout July and increased to 71% in August. In vitro dry matter digestibility content was above 74% throughout July and decreased to 66 % in August. Other forbs at the study site were highly nutritious as CP content remained above 11% throughout the study, NDF content remained below 50%, and IVDMD remained above 70%. Unfortunately for cattle producers, tall larkspur species are highly nutritious and at this location larkspur CP content ranged from 12% at the start of the study to as low as 9% later in the study. The NDF content remained below 43% and IVDMD content was greater than 76% throughout the study. The nutritional quality of larkspur was similar to results reported at this same location (Pfister et al., 2018) and on other mountain rangelands (Pfister et al., 1988). Table 6. Nutrient content1 (% ±SE) of forage classes collected from the larkspur grazing study site insoutheasternIdaho Date and plant class . CP . NDF . IVDMD . July 7  Larkspur 12.6 ± 0.31 35.9 ± 0.38 84.6 ± 0.35  Grasses 10.5 ± 0.30 63.5 ± 0.93 78.4 ± 0.82  Other forbs 14.5 ± 0.36 42.5 ± 1.10 77.6 ± 0.90 July 14  Larkspur 11.6 ± 0.21 37.3 ± 0.70 84.1 ± 0.43  Grasses 9.9 ± 0.16 65.2 ± 0.59 79.3 ± 0.38  Other forbs 14.6 ± 0.37 41.7 ± 0.82 79.7 ± 0.63 July 21  Larkspur 13.0 ± 0.56 35.2 ± 1.19 85.0 ± 1.00  Grasses 10.3 ± 0.49 64.1 ± 3.05 74.5 ± 0.54  Other forbs 13.8 ± 0.36 37.9 ± 1.63 79.0 ± 0.83 July 28  Larkspur 9.2 ± 0.16 42.6 ± 0.43 76.4 ± 0.37  Grasses 9.3 ± 0.22 64.3 ± 0.88 74.2 ± 0.58  Other forbs 11.8 ± 0.15 43.9 ± 1.22 76.2 ± 0.82 August 4  Larkspur 9.5 ± 0.41 42.3 ± 1.32 80.7 ± 1.09  Grasses 7.6 ± 0.35 71.2 ± 0.67 66.1 ± 1.50  Other forbs 11.8 ± 0.38 49.1 ± 0.81 70.8 ± 0.76 Date and plant class . CP . NDF . IVDMD . July 7  Larkspur 12.6 ± 0.31 35.9 ± 0.38 84.6 ± 0.35  Grasses 10.5 ± 0.30 63.5 ± 0.93 78.4 ± 0.82  Other forbs 14.5 ± 0.36 42.5 ± 1.10 77.6 ± 0.90 July 14  Larkspur 11.6 ± 0.21 37.3 ± 0.70 84.1 ± 0.43  Grasses 9.9 ± 0.16 65.2 ± 0.59 79.3 ± 0.38  Other forbs 14.6 ± 0.37 41.7 ± 0.82 79.7 ± 0.63 July 21  Larkspur 13.0 ± 0.56 35.2 ± 1.19 85.0 ± 1.00  Grasses 10.3 ± 0.49 64.1 ± 3.05 74.5 ± 0.54  Other forbs 13.8 ± 0.36 37.9 ± 1.63 79.0 ± 0.83 July 28  Larkspur 9.2 ± 0.16 42.6 ± 0.43 76.4 ± 0.37  Grasses 9.3 ± 0.22 64.3 ± 0.88 74.2 ± 0.58  Other forbs 11.8 ± 0.15 43.9 ± 1.22 76.2 ± 0.82 August 4  Larkspur 9.5 ± 0.41 42.3 ± 1.32 80.7 ± 1.09  Grasses 7.6 ± 0.35 71.2 ± 0.67 66.1 ± 1.50  Other forbs 11.8 ± 0.38 49.1 ± 0.81 70.8 ± 0.76 All concentrations are on a DM basis. All analysis were done on clipped forage samples collected during the trial to determine forage availability. CP, crude protein; NDF, neutral detergent fiber; IVDMD, in vitro dry matter digestibility. Open in new tab Table 6. Nutrient content1 (% ±SE) of forage classes collected from the larkspur grazing study site insoutheasternIdaho Date and plant class . CP . NDF . IVDMD . July 7  Larkspur 12.6 ± 0.31 35.9 ± 0.38 84.6 ± 0.35  Grasses 10.5 ± 0.30 63.5 ± 0.93 78.4 ± 0.82  Other forbs 14.5 ± 0.36 42.5 ± 1.10 77.6 ± 0.90 July 14  Larkspur 11.6 ± 0.21 37.3 ± 0.70 84.1 ± 0.43  Grasses 9.9 ± 0.16 65.2 ± 0.59 79.3 ± 0.38  Other forbs 14.6 ± 0.37 41.7 ± 0.82 79.7 ± 0.63 July 21  Larkspur 13.0 ± 0.56 35.2 ± 1.19 85.0 ± 1.00  Grasses 10.3 ± 0.49 64.1 ± 3.05 74.5 ± 0.54  Other forbs 13.8 ± 0.36 37.9 ± 1.63 79.0 ± 0.83 July 28  Larkspur 9.2 ± 0.16 42.6 ± 0.43 76.4 ± 0.37  Grasses 9.3 ± 0.22 64.3 ± 0.88 74.2 ± 0.58  Other forbs 11.8 ± 0.15 43.9 ± 1.22 76.2 ± 0.82 August 4  Larkspur 9.5 ± 0.41 42.3 ± 1.32 80.7 ± 1.09  Grasses 7.6 ± 0.35 71.2 ± 0.67 66.1 ± 1.50  Other forbs 11.8 ± 0.38 49.1 ± 0.81 70.8 ± 0.76 Date and plant class . CP . NDF . IVDMD . July 7  Larkspur 12.6 ± 0.31 35.9 ± 0.38 84.6 ± 0.35  Grasses 10.5 ± 0.30 63.5 ± 0.93 78.4 ± 0.82  Other forbs 14.5 ± 0.36 42.5 ± 1.10 77.6 ± 0.90 July 14  Larkspur 11.6 ± 0.21 37.3 ± 0.70 84.1 ± 0.43  Grasses 9.9 ± 0.16 65.2 ± 0.59 79.3 ± 0.38  Other forbs 14.6 ± 0.37 41.7 ± 0.82 79.7 ± 0.63 July 21  Larkspur 13.0 ± 0.56 35.2 ± 1.19 85.0 ± 1.00  Grasses 10.3 ± 0.49 64.1 ± 3.05 74.5 ± 0.54  Other forbs 13.8 ± 0.36 37.9 ± 1.63 79.0 ± 0.83 July 28  Larkspur 9.2 ± 0.16 42.6 ± 0.43 76.4 ± 0.37  Grasses 9.3 ± 0.22 64.3 ± 0.88 74.2 ± 0.58  Other forbs 11.8 ± 0.15 43.9 ± 1.22 76.2 ± 0.82 August 4  Larkspur 9.5 ± 0.41 42.3 ± 1.32 80.7 ± 1.09  Grasses 7.6 ± 0.35 71.2 ± 0.67 66.1 ± 1.50  Other forbs 11.8 ± 0.38 49.1 ± 0.81 70.8 ± 0.76 All concentrations are on a DM basis. All analysis were done on clipped forage samples collected during the trial to determine forage availability. CP, crude protein; NDF, neutral detergent fiber; IVDMD, in vitro dry matter digestibility. Open in new tab Alkaloids in larkspur plants Larkspur alkaloids occur as 2 types: the N-(methylsuccinimido) anthranoyllycoctonine (MSAL)-type, such as methyllycaconitine (MLA), and the non-MSAL-type alkaloids (Pfister et al., 1999; Panter et al., 2002). The MSAL-type alkaloids are highly toxic, whereas the non-MSAL alkaloids are much less toxic (Panter et al., 2002). At the start of the study, MSAL alkaloids in larkspur plants were 4.6 ± 0.59 mg/g of plant material and declined over the grazing season to 1.3 ± 0.13 mg/g (Figure 3). Concentration of MSAL alkaloids in D. occidentale varies from 4 to 2 mg/g of plant material from early growth to the pod stage (Gardner and Pfister, 2000; Ralphs et al., 1997; Pfister et al., 2018). Toxic (i.e., MSAL) alkaloid concentrations of D. occidentale may or may not vary with plant maturation (Ralphs et al., 1997). In the current study, MSAL alkaloids decreased over the season. At the same location as the current study, Pfister et al. (2018) reported MSAL levels decreasing over the season in 1 of 3 years and MSAL levels increasing over the season in 2 of the years. Concentrations of alkaloids later in the season are likely due to environmental conditions (i.e., moisture availability; Ralphs et al., 1997) and these conditions change over years. Figure 3. Open in new tabDownload slide Alkaloid concentrations (mg/g of plant material ± SE) of whole plants (stem, leaf, and inflorescence when present) of duncecap larkspur (Delphinium occidentale) during the grazing study in southeastern Idaho. The toxicity of larkspur is due to norditerpene alkaloids, including the highly toxic N-(methylsuccinimido) anthranollycoctonine (MSAL) type. The total alkaloid concentration is the sum of the MSAL-type alkaloids and the less-toxic non-MSAL-type alkaloids (n = 10 plants per collection date). Figure 3. Open in new tabDownload slide Alkaloid concentrations (mg/g of plant material ± SE) of whole plants (stem, leaf, and inflorescence when present) of duncecap larkspur (Delphinium occidentale) during the grazing study in southeastern Idaho. The toxicity of larkspur is due to norditerpene alkaloids, including the highly toxic N-(methylsuccinimido) anthranollycoctonine (MSAL) type. The total alkaloid concentration is the sum of the MSAL-type alkaloids and the less-toxic non-MSAL-type alkaloids (n = 10 plants per collection date). Total alkaloid concentration in larkspur plants was 16.7 ± 0.83 mg/g of plant material at the start of the study and decreased to 5.7 ± 0.26 mg/g at the end of the study (Figure 3). Total alkaloid concentrations decreased over the season, as the plants matured, in 1 year reported by Pfister et al. (2018). However, in the other 2 years, total alkaloid concentrations increased as the plants matured with one year reporting total alkaloid concentrations exceeding 20 mg/g for plants in the pod stage (late July) which exceeds the highest concentrations in the current study detected in plants in the bud stage (early July). Tall larkspur pods are typically higher in alkaloid concentration compared with other plant parts (Gardner and Pfister, 2000). Serum mineral concentration Serum mineral concentrations of mineral-salt supplemented and non-supplemented treatment groups at the start and end of the grazing study are listed in Table 7. There was a treatment × date interaction for serum mineral concentrations of B (P = 0.0038) and Mo (P = 0.0006). Serum concentrations of B were greatest, in both groups, at the start of the grazing study and were greater in the non-supplemented group than the mineral-salt supplemented group at the start of the study. Serum concentrations decreased during the study and were similar between treatment groups at the end of the grazing study (Table 7). Serum concentrations of Mo were lowest in the mineral-salt supplemented group at the start of the grazing study, but serum Mo concentrations increased over the study. Serum Mo concentrations were similar in the mineral-salt supplemented group at the end of the study to the non-supplemented group at both the start and end of the grazing study. Table 7. Serum mineral concentration of steers supplemented with mineral-salt and non-supplemented steers at the start and end of the grazing study Mineral1 . Mineral . No Mineral . Start . End . Start . End . Al 0.07 ± 0.008 0.09 ± 0.027 0.55 ± 0.5 0.12 ± 0.028 B 0.67 ± 0.037b 0.25 ± 0.015c 0.99 ± 0.062a 0.31 ± 0.033c Ba 0.15 ± 0.007 0.16 ± 0.013 0.17 ± 0.01 0.19 ± 0.008 Ca 72.56 ± 6.24 71.83 ± 1.87 68.02 ± 0.908 75.67 ± 1.605 Co 3.7 ± 0.21 3.3 ± 0.42 2.0 ± 0 2.0 ± 0 Cr 0.003 ± 0.0002 0.01 ± 0.009 0.004 ± 0.0007 0.02 ± 0.008 Cu 0.66 ± 0.041 0.87 ± 0.058 0.69 ± 0.049 0.89 ± 0.049 Fe 3.14 ± 0.286 3.67 ± 1.31 2.81 ± 0.187 4.45 ± 1.023 K 194.98 ± 6.911 201.49 ± 8.14 193.99 ± 12.47 211.31 ± 16.702 Mg 22.8 ± 0.632 24.61 ± 1.08 24.45 ± 0.901 25.06 ± 1.466 Mn 7.7 ± 1.3 6.0 ± 0.37 8.7 ± 2.67 6.0 ± 0.26 Mo 35.5 ± 5.14b 195.7 ± 20.69a 148.2 ± 15.27a 170.67 ± 21.04a Na 3,272.98 ± 76.79 3,647.01 ± 80.51 3,216.68 ± 39.14 3,768.45 ± 96.23 Ni 2.0 ± 0 2.5 ± 0.5 1.67 ± 0.33 2.17 ± 0.17 P 143.10 ± 2.423 143.42 ± 8.03 128.06 ± 6.895 142.07 ± 9.978 Se 0.09 ± 0.004 0.11 ± 0.007 0.06 ± 0.005 0.08 ± 0.007 Si 11.2 ± 1.466 9.89 ± 0.42 13.0 ± 2.974 9.81 ± 0.165 Sr 0.19 ± 0.013 0.15 ± 0.01 0.23 ± 0.008 0.15 ± 0.009 Zn 1.05 ± 0.238 1.09 ± 0.64 0.75 ± 0.039 1.00 ± 0.040 Mineral1 . Mineral . No Mineral . Start . End . Start . End . Al 0.07 ± 0.008 0.09 ± 0.027 0.55 ± 0.5 0.12 ± 0.028 B 0.67 ± 0.037b 0.25 ± 0.015c 0.99 ± 0.062a 0.31 ± 0.033c Ba 0.15 ± 0.007 0.16 ± 0.013 0.17 ± 0.01 0.19 ± 0.008 Ca 72.56 ± 6.24 71.83 ± 1.87 68.02 ± 0.908 75.67 ± 1.605 Co 3.7 ± 0.21 3.3 ± 0.42 2.0 ± 0 2.0 ± 0 Cr 0.003 ± 0.0002 0.01 ± 0.009 0.004 ± 0.0007 0.02 ± 0.008 Cu 0.66 ± 0.041 0.87 ± 0.058 0.69 ± 0.049 0.89 ± 0.049 Fe 3.14 ± 0.286 3.67 ± 1.31 2.81 ± 0.187 4.45 ± 1.023 K 194.98 ± 6.911 201.49 ± 8.14 193.99 ± 12.47 211.31 ± 16.702 Mg 22.8 ± 0.632 24.61 ± 1.08 24.45 ± 0.901 25.06 ± 1.466 Mn 7.7 ± 1.3 6.0 ± 0.37 8.7 ± 2.67 6.0 ± 0.26 Mo 35.5 ± 5.14b 195.7 ± 20.69a 148.2 ± 15.27a 170.67 ± 21.04a Na 3,272.98 ± 76.79 3,647.01 ± 80.51 3,216.68 ± 39.14 3,768.45 ± 96.23 Ni 2.0 ± 0 2.5 ± 0.5 1.67 ± 0.33 2.17 ± 0.17 P 143.10 ± 2.423 143.42 ± 8.03 128.06 ± 6.895 142.07 ± 9.978 Se 0.09 ± 0.004 0.11 ± 0.007 0.06 ± 0.005 0.08 ± 0.007 Si 11.2 ± 1.466 9.89 ± 0.42 13.0 ± 2.974 9.81 ± 0.165 Sr 0.19 ± 0.013 0.15 ± 0.01 0.23 ± 0.008 0.15 ± 0.009 Zn 1.05 ± 0.238 1.09 ± 0.64 0.75 ± 0.039 1.00 ± 0.040 All mineral concentration values are ppm except Co, Mn, Mo, and Ni which are ppb.. Useful conversions: 1 µg/mL = mg/L = 1 ppm; 1 ng/mL = µg/L = ppb. Means with different subscripts within a row differ (period × treatment interaction; P < 0.05). Open in new tab Table 7. Serum mineral concentration of steers supplemented with mineral-salt and non-supplemented steers at the start and end of the grazing study Mineral1 . Mineral . No Mineral . Start . End . Start . End . Al 0.07 ± 0.008 0.09 ± 0.027 0.55 ± 0.5 0.12 ± 0.028 B 0.67 ± 0.037b 0.25 ± 0.015c 0.99 ± 0.062a 0.31 ± 0.033c Ba 0.15 ± 0.007 0.16 ± 0.013 0.17 ± 0.01 0.19 ± 0.008 Ca 72.56 ± 6.24 71.83 ± 1.87 68.02 ± 0.908 75.67 ± 1.605 Co 3.7 ± 0.21 3.3 ± 0.42 2.0 ± 0 2.0 ± 0 Cr 0.003 ± 0.0002 0.01 ± 0.009 0.004 ± 0.0007 0.02 ± 0.008 Cu 0.66 ± 0.041 0.87 ± 0.058 0.69 ± 0.049 0.89 ± 0.049 Fe 3.14 ± 0.286 3.67 ± 1.31 2.81 ± 0.187 4.45 ± 1.023 K 194.98 ± 6.911 201.49 ± 8.14 193.99 ± 12.47 211.31 ± 16.702 Mg 22.8 ± 0.632 24.61 ± 1.08 24.45 ± 0.901 25.06 ± 1.466 Mn 7.7 ± 1.3 6.0 ± 0.37 8.7 ± 2.67 6.0 ± 0.26 Mo 35.5 ± 5.14b 195.7 ± 20.69a 148.2 ± 15.27a 170.67 ± 21.04a Na 3,272.98 ± 76.79 3,647.01 ± 80.51 3,216.68 ± 39.14 3,768.45 ± 96.23 Ni 2.0 ± 0 2.5 ± 0.5 1.67 ± 0.33 2.17 ± 0.17 P 143.10 ± 2.423 143.42 ± 8.03 128.06 ± 6.895 142.07 ± 9.978 Se 0.09 ± 0.004 0.11 ± 0.007 0.06 ± 0.005 0.08 ± 0.007 Si 11.2 ± 1.466 9.89 ± 0.42 13.0 ± 2.974 9.81 ± 0.165 Sr 0.19 ± 0.013 0.15 ± 0.01 0.23 ± 0.008 0.15 ± 0.009 Zn 1.05 ± 0.238 1.09 ± 0.64 0.75 ± 0.039 1.00 ± 0.040 Mineral1 . Mineral . No Mineral . Start . End . Start . End . Al 0.07 ± 0.008 0.09 ± 0.027 0.55 ± 0.5 0.12 ± 0.028 B 0.67 ± 0.037b 0.25 ± 0.015c 0.99 ± 0.062a 0.31 ± 0.033c Ba 0.15 ± 0.007 0.16 ± 0.013 0.17 ± 0.01 0.19 ± 0.008 Ca 72.56 ± 6.24 71.83 ± 1.87 68.02 ± 0.908 75.67 ± 1.605 Co 3.7 ± 0.21 3.3 ± 0.42 2.0 ± 0 2.0 ± 0 Cr 0.003 ± 0.0002 0.01 ± 0.009 0.004 ± 0.0007 0.02 ± 0.008 Cu 0.66 ± 0.041 0.87 ± 0.058 0.69 ± 0.049 0.89 ± 0.049 Fe 3.14 ± 0.286 3.67 ± 1.31 2.81 ± 0.187 4.45 ± 1.023 K 194.98 ± 6.911 201.49 ± 8.14 193.99 ± 12.47 211.31 ± 16.702 Mg 22.8 ± 0.632 24.61 ± 1.08 24.45 ± 0.901 25.06 ± 1.466 Mn 7.7 ± 1.3 6.0 ± 0.37 8.7 ± 2.67 6.0 ± 0.26 Mo 35.5 ± 5.14b 195.7 ± 20.69a 148.2 ± 15.27a 170.67 ± 21.04a Na 3,272.98 ± 76.79 3,647.01 ± 80.51 3,216.68 ± 39.14 3,768.45 ± 96.23 Ni 2.0 ± 0 2.5 ± 0.5 1.67 ± 0.33 2.17 ± 0.17 P 143.10 ± 2.423 143.42 ± 8.03 128.06 ± 6.895 142.07 ± 9.978 Se 0.09 ± 0.004 0.11 ± 0.007 0.06 ± 0.005 0.08 ± 0.007 Si 11.2 ± 1.466 9.89 ± 0.42 13.0 ± 2.974 9.81 ± 0.165 Sr 0.19 ± 0.013 0.15 ± 0.01 0.23 ± 0.008 0.15 ± 0.009 Zn 1.05 ± 0.238 1.09 ± 0.64 0.75 ± 0.039 1.00 ± 0.040 All mineral concentration values are ppm except Co, Mn, Mo, and Ni which are ppb.. Useful conversions: 1 µg/mL = mg/L = 1 ppm; 1 ng/mL = µg/L = ppb. Means with different subscripts within a row differ (period × treatment interaction; P < 0.05). Open in new tab Serum concentrations of Ba (P = 0.02) and Sr (P = 0.018) displayed a treatment effect in which serum mineral concentrations were greater in the non-supplemented group. Serum concentrations of Ba and Sr were also greater in the non-supplemented group in the pen trial. Serum concentrations of Co (P < 0.0001) and Se (P < 0.0001) were greater in the mineral-salt supplemented group. Serum concentrations of Co and Se were also greater in the mineral-salt supplemented group in the pen trial. Animal diet selection Cattle consumption of total larkspur (i.e., % of bites) accounted for 2 to 6% of cattle diets on a given day (Figure 4). Supplementing cattle with the mineral-salt mix did not alter larkspur consumption (P > 0.05) with consumption averaging 3 ± 1.0% and 2 ± 1.1% for cattle supplemented with mineral-salt and non-supplemented, respectively. There was no treatment × date interaction or day effect (P > 0.05) for larkspur consumption by cattle supplemented with mineral-salt (Figure 4A). However, there was a spike in larkspur consumption of animals supplemented with mineral-salt on July 9, 19, 23, and 31 which correlates with days animals become intoxicated and recumbent (Table 8). Table 8. Incidences of larkspur intoxication of animals grazing larkspur infested rangeland in southeastern Idaho during 2020 Date . Treatment group and animal number . MLA1 (ng/mL) . Mean larkspur2 (% of bites) . Maximum larkspur in grazing bout2 (% of bites) . Hours since last known larkspur consumption . Neostigmine administered . Remarks . July 9 Mineral no. 8 — —3 —3 Unknown No Animal was down at 6 a.m. and displayed weakness and continued to lay down throughout the day July 10 Mineral no. 5 — 6 37 21 No Collapsed about noon and continued to show signs of weakness and lay down throughout the day until about 9 p.m. July 19 Mineral no. 9 — 6 46 4 No Collapsedabout noon. Showed signs of weakness and was still down at 9 p.m. Animal was up grazing the next morning at 6 a.m. July 23 Mineral no. 8 1,322 1,030 14 34 1.5 Yes Yes Collapsed at noon. Treated with first dose at 1 p.m. Became ambulatory and began grazing. Collapsed the second time at 3 p.m. Treated with second dose at 6 p.m. Became ambulatory and was placed into the corral overnight. Started grazing at 6 a.m. the next morning July 31 No mineral no.3 — 4 19 1 No Collapsed at 8 a.m. Did not get up until 1 p.m. and was fine the remainder of the day Date . Treatment group and animal number . MLA1 (ng/mL) . Mean larkspur2 (% of bites) . Maximum larkspur in grazing bout2 (% of bites) . Hours since last known larkspur consumption . Neostigmine administered . Remarks . July 9 Mineral no. 8 — —3 —3 Unknown No Animal was down at 6 a.m. and displayed weakness and continued to lay down throughout the day July 10 Mineral no. 5 — 6 37 21 No Collapsed about noon and continued to show signs of weakness and lay down throughout the day until about 9 p.m. July 19 Mineral no. 9 — 6 46 4 No Collapsedabout noon. Showed signs of weakness and was still down at 9 p.m. Animal was up grazing the next morning at 6 a.m. July 23 Mineral no. 8 1,322 1,030 14 34 1.5 Yes Yes Collapsed at noon. Treated with first dose at 1 p.m. Became ambulatory and began grazing. Collapsed the second time at 3 p.m. Treated with second dose at 6 p.m. Became ambulatory and was placed into the corral overnight. Started grazing at 6 a.m. the next morning July 31 No mineral no.3 — 4 19 1 No Collapsed at 8 a.m. Did not get up until 1 p.m. and was fine the remainder of the day Sera was collected from animals that were laterally recumbent from larkspur intoxication. Sera was not collected from animals that were sternally recumbent because attempting to restrain such animals would overstress the animal and exacerbate the larkspur intoxication. The anticholinesterase drug neostigmine was administered intravenously at 0.04 mg/kg of BW to animals that were laterally recumbent. MLA = methyllycaconitine. Consumption of larkspur (% of bites). Mean consumption over the previous 24 h and the maximum consumption during any grazing bout during the previous 24 h. No bites on larkspur were recorded during data collection. The animal did not go into its corral overnight and must have consumed larkspur during the night. Open in new tab Table 8. Incidences of larkspur intoxication of animals grazing larkspur infested rangeland in southeastern Idaho during 2020 Date . Treatment group and animal number . MLA1 (ng/mL) . Mean larkspur2 (% of bites) . Maximum larkspur in grazing bout2 (% of bites) . Hours since last known larkspur consumption . Neostigmine administered . Remarks . July 9 Mineral no. 8 — —3 —3 Unknown No Animal was down at 6 a.m. and displayed weakness and continued to lay down throughout the day July 10 Mineral no. 5 — 6 37 21 No Collapsed about noon and continued to show signs of weakness and lay down throughout the day until about 9 p.m. July 19 Mineral no. 9 — 6 46 4 No Collapsedabout noon. Showed signs of weakness and was still down at 9 p.m. Animal was up grazing the next morning at 6 a.m. July 23 Mineral no. 8 1,322 1,030 14 34 1.5 Yes Yes Collapsed at noon. Treated with first dose at 1 p.m. Became ambulatory and began grazing. Collapsed the second time at 3 p.m. Treated with second dose at 6 p.m. Became ambulatory and was placed into the corral overnight. Started grazing at 6 a.m. the next morning July 31 No mineral no.3 — 4 19 1 No Collapsed at 8 a.m. Did not get up until 1 p.m. and was fine the remainder of the day Date . Treatment group and animal number . MLA1 (ng/mL) . Mean larkspur2 (% of bites) . Maximum larkspur in grazing bout2 (% of bites) . Hours since last known larkspur consumption . Neostigmine administered . Remarks . July 9 Mineral no. 8 — —3 —3 Unknown No Animal was down at 6 a.m. and displayed weakness and continued to lay down throughout the day July 10 Mineral no. 5 — 6 37 21 No Collapsed about noon and continued to show signs of weakness and lay down throughout the day until about 9 p.m. July 19 Mineral no. 9 — 6 46 4 No Collapsedabout noon. Showed signs of weakness and was still down at 9 p.m. Animal was up grazing the next morning at 6 a.m. July 23 Mineral no. 8 1,322 1,030 14 34 1.5 Yes Yes Collapsed at noon. Treated with first dose at 1 p.m. Became ambulatory and began grazing. Collapsed the second time at 3 p.m. Treated with second dose at 6 p.m. Became ambulatory and was placed into the corral overnight. Started grazing at 6 a.m. the next morning July 31 No mineral no.3 — 4 19 1 No Collapsed at 8 a.m. Did not get up until 1 p.m. and was fine the remainder of the day Sera was collected from animals that were laterally recumbent from larkspur intoxication. Sera was not collected from animals that were sternally recumbent because attempting to restrain such animals would overstress the animal and exacerbate the larkspur intoxication. The anticholinesterase drug neostigmine was administered intravenously at 0.04 mg/kg of BW to animals that were laterally recumbent. MLA = methyllycaconitine. Consumption of larkspur (% of bites). Mean consumption over the previous 24 h and the maximum consumption during any grazing bout during the previous 24 h. No bites on larkspur were recorded during data collection. The animal did not go into its corral overnight and must have consumed larkspur during the night. Open in new tab Figure 4. Open in new tabDownload slide Bites (% of diet ± SE) taken of larkspur, grasses, and forbs by mineral supplemented and non-supplemented cattle while grazing larkspur-infested rangeland in southeastern Idaho. There was no treatment interaction (P > 0.05) for total larkspur, grass, or forb consumption and no treatment × date interaction (P > 0.05) for total larkspur. Dates with an * represent a significant difference between mineral supplemented and non-supplemented cattle for grass and consumption (P < 0.05). Figure 4. Open in new tabDownload slide Bites (% of diet ± SE) taken of larkspur, grasses, and forbs by mineral supplemented and non-supplemented cattle while grazing larkspur-infested rangeland in southeastern Idaho. There was no treatment interaction (P > 0.05) for total larkspur, grass, or forb consumption and no treatment × date interaction (P > 0.05) for total larkspur. Dates with an * represent a significant difference between mineral supplemented and non-supplemented cattle for grass and consumption (P < 0.05). Perennial grasses made up the greatest proportion of cattle diets accounting for ≥62% of the diet on a given day (Figure 4). There was a treatment x day interaction for bites on grasses (P = 0.039) in which cattle supplemented with mineral-salt consumed greater amounts of grass than non-supplemented cattle on 6 d (July 9, 15, 23, 24, 26, and 29; Figure 4B). Other forbs made up the next greatest proportion of cattle diets accounting for 9% to 35% of the diet on a given day (Figure 4). There was a treatment × day interaction for bites on other forbs (P = 0.008) with non-supplemented cattle consuming more other forbs than cattle supplemented with mineral-salt on 4 d (July 9, 23, 24, and 26; Figure 4C). The mineral-salt supplement did not decrease consumption of larkspur by grazing cattle compared with cattle not offered the mineral-salt mixture. Consumption of tall larkspur (i.e., D. occidentale and D. barbeyi) has been reported in the range of 5% to 7% in other grazing studies (Pfister et al., 1988; Ralphs and Pfister, 1992) and 3% to 7% at this same location (Pfister et al., 2018). Larkspur consumption averaged 2% to 3% for both groups in the current study. Pfister et al. (2018) also report consumption of larkspur during any grazing bout by a single animal ranging from 24% to 73%. In our study, maximum consumption of larkspur during any grazing bout by a single animal was 57% and 52% for mineral-salt supplemented and non-supplemented cattle, respectively. Even though average consumption of larkspur was low (2% to 3%), heavy consumption of larkspur during a grazing bout will often result in animals becoming visibly intoxicated. Six animals were visibly poisoned by larkspur consumption over the course of the grazing study (Table 8). All six animals collapsed and displayed signs of weakness for a portion of the day. Larkspur poisoning in cattle is the result of norditerpene alkaloid blockade in nicotinic acetylcholine receptors in post-synaptic neuromuscular junctions (Aiyar et al., 1979; Green et al., 2013a; Welch et al., 2013). Clinical signs of poisoning include muscle weakness, trembling and lack of coordination, rapid heart rate, sternal recumbency (i.e., lying on brisket and unable to stand) followed by lateral recumbency (i.e., unable to maintain an upright posture even when lying down) which often leads to bloating followed by death. Only one animal became laterally recumbent and needed to be treated with an anticholinesterase drug to prevent bloating and death (Green et al., 2009b; Pfister et al., 1994). Five of the 6 animals that collapsed, including the animal that became laterally recumbent and needed to be treated, were in the mineral-salt supplement treatment. Blood serum was collected from the single animal that became laterally recumbent. The animal was able to stand within a few minutes after treatment. However, it collapsed for a second time later in the day and became laterally recumbent again at which time serum was collected for a second time. Serum MLA concentrations in those two instances were 1,322 and 1,030 ng/mL, respectively (Table 8). Pfister et al. (2018) provides the only other limited field data on serum concentration of MLA from recumbent animals grazing larkspur. They report concentrations ranging from 613 to 1,185 ng/mL from recumbent animals intoxicated from larkspur consumption. Previous research examining larkspur consumption of cattle supplemented with a mineral-salt reported low ingestion of larkspur (Pfister and Manners, 1991). The authors reported that mineral-salt supplementation had no influence on the amount of larkspur selected but cautioned that the results from the study may be influenced by drought. Larkspur consumption was low during the current grazing study and may have influenced the results of the current study. Serum was also collected from animals restrained in a portable chute once a week during the grazing study. Average and maximum serum concentrations of MLA were numerically higher in the mineral-salt supplemented group compared with the non-supplemented animals (Table 9). However, consumption of larkspur was similar between groups the previous 2 d before serum was collected (P > 0.05). The higher MLA concentrations in the mineral-salt supplemented group is an indication that animals on the mineral-salt may be able to tolerate higher concentrations of larkspur alkaloids. On the other hand, the instances of visible intoxication, and recumbency, in the mineral-salt supplemented cattle suggests that under grazing conditions their risk of fatal intoxication is still substantial. There were no animals that were visibly intoxicated in the previous 24 h before serum collection. Table 9. Mean concentrations (ng/mL) of methyllycaconitine (MLA) in sera1 of mineral supplemented or non-supplemented cattle grazing larkspur-infested rangelands in southeastern Idaho in 2020 Date2 . Treatment group . MLA . Maximum serum MLA concentration for any animal in the group . Mean larkspur3 (% of bites) . Maximum larkspur in grazing bouts3 (% of bites) . July 6 Mineral 0 0 — — No mineral 0 0 — — July 15 Mineral 267 ± 164 854 2 14 No mineral 27 ± 10 56 2 2 July 22 Mineral 301 ± 75 434 3 10 No mineral 90 ± 38 240 2 4 July 29 Mineral 338 ± 154 737 4 30 No mineral 99 ± 58 335 5 52 August 5 Mineral 16 ± 10 60 — — No Mineral 0 0 — — Date2 . Treatment group . MLA . Maximum serum MLA concentration for any animal in the group . Mean larkspur3 (% of bites) . Maximum larkspur in grazing bouts3 (% of bites) . July 6 Mineral 0 0 — — No mineral 0 0 — — July 15 Mineral 267 ± 164 854 2 14 No mineral 27 ± 10 56 2 2 July 22 Mineral 301 ± 75 434 3 10 No mineral 90 ± 38 240 2 4 July 29 Mineral 338 ± 154 737 4 30 No mineral 99 ± 58 335 5 52 August 5 Mineral 16 ± 10 60 — — No Mineral 0 0 — — Sera was collected from all animals (n = 5 or 6 per treatment group) at mid-day as animals were restrained in a chute. July 6 collection was prior to the start of the grazing trial. August 5 collection was collected 24 h after the animals were removed from the study pastures. Consumption of larkspur (% of bites). Mean consumption is the group average over the previous 24 h and the maximum consumption reflects the maximum by any individual animal during the previous 24 h. Open in new tab Table 9. Mean concentrations (ng/mL) of methyllycaconitine (MLA) in sera1 of mineral supplemented or non-supplemented cattle grazing larkspur-infested rangelands in southeastern Idaho in 2020 Date2 . Treatment group . MLA . Maximum serum MLA concentration for any animal in the group . Mean larkspur3 (% of bites) . Maximum larkspur in grazing bouts3 (% of bites) . July 6 Mineral 0 0 — — No mineral 0 0 — — July 15 Mineral 267 ± 164 854 2 14 No mineral 27 ± 10 56 2 2 July 22 Mineral 301 ± 75 434 3 10 No mineral 90 ± 38 240 2 4 July 29 Mineral 338 ± 154 737 4 30 No mineral 99 ± 58 335 5 52 August 5 Mineral 16 ± 10 60 — — No Mineral 0 0 — — Date2 . Treatment group . MLA . Maximum serum MLA concentration for any animal in the group . Mean larkspur3 (% of bites) . Maximum larkspur in grazing bouts3 (% of bites) . July 6 Mineral 0 0 — — No mineral 0 0 — — July 15 Mineral 267 ± 164 854 2 14 No mineral 27 ± 10 56 2 2 July 22 Mineral 301 ± 75 434 3 10 No mineral 90 ± 38 240 2 4 July 29 Mineral 338 ± 154 737 4 30 No mineral 99 ± 58 335 5 52 August 5 Mineral 16 ± 10 60 — — No Mineral 0 0 — — Sera was collected from all animals (n = 5 or 6 per treatment group) at mid-day as animals were restrained in a chute. July 6 collection was prior to the start of the grazing trial. August 5 collection was collected 24 h after the animals were removed from the study pastures. Consumption of larkspur (% of bites). Mean consumption is the group average over the previous 24 h and the maximum consumption reflects the maximum by any individual animal during the previous 24 h. Open in new tab Angus cattle supplemented with the mineral-salt, in the pen trial, were more resistant to larkspur induced muscle weakness than non-supplemented animals. In the grazing study, there was not a difference in larkspur intake between mineral-salt supplemented and non-supplemented animals. Also, there were more grazing animals visibly intoxicated in the mineral-salt supplemented treatment group. Inconsistencies in results between the pen trial and grazing trial can be attributed to fewer animals utilized in the grazing study compared with the pen trial. In the pen trial, animals rotated between ‘on-off’ mineral resulting in more animals supplemented with mineral-salt or non-supplemented resulting in greater statistical power in the pen trial. Also, it was easier to control the amount of larkspur given to each animal in the pen trial. Exact consumption of larkspur in the grazing trail is unknown as animals were observed for short periods throughout the day resulting in only a snapshot of the days larkspur consumption. Previous research at this laboratory has shown that approximately 10% of a sample of purebred Angus cattle are highly susceptible to larkspur intoxication and another 10% of a sample of purebred Angus cattle are much less susceptible (resistant) to larkspur intoxication (Green et al., 2019b). In the current pen trial, 17% of the animals were susceptible to larkspur toxicosis and 33% were resistant to larkspur as shown by the walk times after larkspur dosing (Figure 5). There was an intermediate group, that represented 50% of the steers that had the greatest changes in walk times between mineral cycles. Cattle that are highly susceptible, or resistant, to larkspur toxicosis are not likely to respond to a mineral-salt supplement. We speculate that susceptible animals may still be fatally poisoned after one or several consecutive bouts of larkspur consumption regardless of supplemental status, while the larkspur susceptibility of more resistant animals will likely be unaffected by a mineral-salt supplement. We suggest that cattle that fall within the intermediate group are the animals most likely to benefit from mineral-salt supplementation. A mineral-salt supplement may help reduce the incidence of larkspur poisoning as was shown by the exercise tolerance of supplemented steers in the pen feeding trial. Based on the results of our study (17% and 33% susceptible and resistant to larkspur, respectively) and genetic differences observed by Green et al. (2019b; 10% resistant and 10% susceptible, respectively), we suggest that approximately 50% to 80% of a given herd will benefit from a good mineral supplementation regimen with regards to increased protection for larkspur poisoning. Reducing the risk of larkspur losses in up to 80% of a cattle herd, in conjunction with improvements in heard health and reproduction from increases in mineral status (Sprinkle et al., 2011), will likely increase the profitability of grazing cattle on larkspur-containing rangelands. Figure 5. Open in new tabDownload slide Walk times of Angus steers 24 h after; a baseline oral dose (1; prior to animals supplemented with a mineral-salt) and 4 sequential oral doses (2 to 5; animals were supplemented with a mineral-salt for 33 ± 1 d before dosing) of 7.5 mg/kg BW of MSAL-type alkaloids in the form of dried ground larkspur. The steers were divided into 3 groups post hoc based-on susceptibility to larkspur intoxication: susceptible (4 ± 4 min), intermediate (15 ± 2 min), and resistant (32 ± 3 min) based upon their repeated average walk times over the course of the experiment. Figure 5. Open in new tabDownload slide Walk times of Angus steers 24 h after; a baseline oral dose (1; prior to animals supplemented with a mineral-salt) and 4 sequential oral doses (2 to 5; animals were supplemented with a mineral-salt for 33 ± 1 d before dosing) of 7.5 mg/kg BW of MSAL-type alkaloids in the form of dried ground larkspur. The steers were divided into 3 groups post hoc based-on susceptibility to larkspur intoxication: susceptible (4 ± 4 min), intermediate (15 ± 2 min), and resistant (32 ± 3 min) based upon their repeated average walk times over the course of the experiment. Conclusion Mineral supply is highly variable in rangeland-based beef operations and is dependent on a number of factors such as weather, forage type, stage of forage growth, and soil fertility (Greene, 2016). It is always good management practice to provide grazing animals with a mineral-salt tailored to specific regional forage conditions. Results from the pen study suggest that a mineral-salt supplementation program will provide a protective effect for animals grazing in larkspur-infested ranges. The Angus cattle used in the pen study, were more resistant to larkspur-induced muscle weakness when supplemented with mineral-salt. The data also suggest that the mineral-salt supplement must be provided continuously throughout the time the animals are grazing these rangelands as the positive effects can be lost within 30 d post supplementation. When cattle were allowed to graze rangelands infested with larkspur, the mineral-salt supplemented animals were not observed to consume any less larkspur than non-supplemented animals. The results may have been influenced by the relatively low overall intake of larkspur by both groups. There were more animals in the mineral-salt supplemented group that were visibly intoxicated than animals in the non-supplemented group while grazing. Serum concentrations of larkspur alkaloids were higher in the mineral-salt supplemented animals and may be an indicator that mineral-salt supplemented animals may be able to tolerate higher concentrations of larkspur alkaloids, thus consume more larkspur. Caution must be taken as there is an ever-present risk of animals on a mineral-salt supplementation program becoming fatally intoxicated after consuming larkspur. Mineral supplementation to reduce cattle losses to larkspur should be one portion of an integrated grazing management strategy that includes such aspects as local knowledge of larkspur species and their population density on rangelands (Gardner and Pfister, 2007), seasonal changes in the concentration of toxic alkaloids (Pfister et al., 2002), the growth stages of larkspur (Pfister et al., 1997; Gardner and Pfister, 2000), and the likelihood of cattle ingesting a lethal quantity of larkspur (Green et al., 2020). Abbreviations Abbreviations BW body weight CP crude protein GLIMMIX generalized linear mixed model IVDMD in vitro dry matter digestibility MLA methyllycaconitine MS mass spectrometer MSAL N-(methylsuccinimido) anthranoyllycoctonine NDF neutral detergent fiber Acknowledgments The authors thank Kermit Price, Anita McCollum, Jean Bennett, and Scott Larsen for technical assistance. We thank John Hagenbarth for allowing the use of his pasture, and we thank him and his crew for their assistance. The study was funded by the United States Department of Agriculture-Agricultural Research Service. Mention of a proprietary product does not constitute a guarantee or warranty of the product by the US Department of Agriculture (USDA) or the authors and does not imply its approval to the exclusion of other products. Conflict of Interest Statement The authors declare no real or perceived conflicts of interest. Literature Cited Aiyar , V. N. , M. H. Benn, T. Hanna, J. Jacyno, S. H. Roth, and J. L. Wilkens. 1979 . The principal toxin of Delphinium brownii Rydb., and its mode of action. Experientia 35 : 1367 – 1368 . doi:10.1007/BF01964013 Google Scholar Crossref Search ADS PubMed WorldCat Altmann , J. 1974 . Observational study of behavior: sampling methods. Behaviour 49 : 227 – 267 . doi:10.1163/156853974X00534 Google Scholar Crossref Search ADS PubMed WorldCat AOAC [Association of Official Analytical Chemists]. 1995 . Official methods of analysis . 16th ed. Arlington, VA : Assoc. Off. Anal. Chem . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Buchanan , H. , W. A. Laycock, and D. A. Price. 1972 . Botanical and nutritive content of the summer diet of sheep on a tall forb range in southwestern Montana. J. Anim. Sci . 35 : 423 – 430 . doi:10.2527/jas1972.352423x Google Scholar Crossref Search ADS PubMed WorldCat Cook , D. , B. T. Green, K. D. Welch, D. R. Gardner, J. A. Pfister, and K. E. Panter. 2011 . Comparison of the toxic effects of two duncecap larkspur (Delphinium occidentale) chemotypes in mice and cattle. Am. J. Vet. Res . 72 : 706 – 714 . doi:10.2460/ajvr.72.5.706 Google Scholar Crossref Search ADS PubMed WorldCat Dobelis , P. , J. E. Madl, J. A. Pfister, G. D. Manners, and J. P. Walrond. 1999 . Effects of Delphinium alkaloids on neuromuscular transmission. J. Pharmacol. Exp. Ther . 291 : 538 – 546 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Engle , T. E. , and J. W. Spears. 2000 . Effects of dietary copper concentration and source on performance and copper status of growing and finishing steers. J. Anim. Sci . 78 : 2446 – 2451 . doi:10.2527/2000.7892446x Google Scholar Crossref Search ADS PubMed WorldCat Gardner , D. R. , G. D. Manners, M. H. Ralphs, and J. A. Pfister. 1997 . Quantitative analysis of norditerpenoid alkaloids in larkspur (Delphinium spp.) by Fourier transform infrared spectroscopy. Phytochem. Anal . 8 : 55 – 62 . doi:10.1002/(SICI)1099-1565(199703)8:2<55::AID-PCA338>3.0.CO;2-Q Google Scholar Crossref Search ADS WorldCat Gardner , D. R. , and J. A. Pfister. 2000 . Late season toxic alkaloid concentrations in tall larkspur (Delphinium spp.). J. Range Manag . 53 : 329 – 334 . doi:10.2307/4003441. Google Scholar Crossref Search ADS WorldCat Gardner , D. R. , and J. A. Pfister. 2007 . Toxic alkaloid concentrations in Delphinium nuttallianum, Delphinium andersonii, and Delphinium geyeri in the intermountain region. Range. Ecol. Manag . 60 : 441 – 446 . doi:10.2111/1551-5028(2007)60[441:TACIDN]2.0.CO;2. Google Scholar Crossref Search ADS WorldCat Gardner , D. R. , and J. A. Pfister. 2009 . HPLC/MS analysis of toxic norditerpenoid alkaloids: refinement of toxicity assessment of low larkspurs (Delphinium spp.) . Phytochem. Anal . 20 : 104 – 113 . doi:10.1002/pca.1104. Google Scholar Crossref Search ADS PubMed WorldCat Graphpad Software. 2021 . How to analyze a 2x2 Crossover Study in Prism . – [accessed November 29 ]. https://www.graphpad.com/support/faq/how-to-analyze-crossover-study-in-prism/ Green , B. T. , D. R. Gardner, C. A. Stonecipher, S. T. Lee, J. A. Pfister, K. D. Welch, D. Cook, T. Z. Davis, and B. L. Stegelmeier. 2020 . Larkspur poisoning of cattle: plant and animal factors that influence plant toxicity. Rangelands 42 : 1 – 8 . doi:10.1016/j.rala.2020.01.004. Google Scholar Crossref Search ADS WorldCat Green , B. T. , J. W. Keele, G. L. Bennett, D. R. Gardner, C. A. Stonecipher, D. Cook, and J. A. Pfister. 2019b . Animal and plant factors which affect larkspur toxicosis in cattle: sex, age, breed, and plant chemotype. Toxicon 165 : 31 – 39 . doi:10.1016/j.toxicon.2019.04.013. Google Scholar Crossref Search ADS WorldCat Green , B. T. , J. W. Keele, D. R. Gardner, K. D. Welch, G. L. Bennett, D. Cook, J. A. Pfister, T. Z. Davis, C. A. Stonecipher, S. T. Lee, et al. . 2019a . Sex-dependent differences for larkspur (Delphinium barbeyi) toxicosis in yearling Angus cattle. J. Anim. Sci . 97 : 1424 – 1432 . doi:10.1093/jas/skz002. Google Scholar Crossref Search ADS WorldCat Green , B. T. , J. A. Pfister, D. Cook, K. D. Welch, B. L. Stegelmeier, S. T. Lee, D. R. Gardner, E. L. Knoppel, and K. E. Panter. 2009b . Effects of larkspur (Delphinium barbeyi) on heart rate and electrically evoked electromyographic response of the external anal sphincter in cattle. Am. J. Vet. Res . 70 : 539 – 546 . doi:10.2460/ajvr.70.4.539. Google Scholar Crossref Search ADS WorldCat Green , B. T. , K. D. Welch, D. R. Gardner, B. L. Stegelmeier, T. Z. Davis, D. Cook, S. T. Lee, J. A. Pfister, and K. E. Panter. 2009a . Serum elimination profiles of methyllycaconitine and deltaline in cattle following oral administration of larkspur (Delphinium barbeyi). Am. J. Vet. Res . 70 : 926 – 931 . doi:10.2460/ajvr.70.7.926. Google Scholar Crossref Search ADS WorldCat Green , B. T. , K. D. Welch, D. R. Gardner, B. L. Stegelmeier, and S. T. Lee. 2013b . A toxicokinetic comparison of two species of low larkspur (Delphinium spp.) in cattle. Res. Vet. Sci . 95 : 612 – 615 . doi:10.1016/j.rvsc.2013.04.018. Google Scholar Crossref Search ADS WorldCat Green , B. T. , K. D. Welch, D. R. Gardner, B. L. Stegelmeier, J. A. Pfister, D. Cook, and T. Z. Davis. 2011 . A toxicokinetic comparison of norditerpenoid alkaloids from Delphinium barbeyi and D. glaucescens in cattle. J. Appl. Toxicol . 31 : 20 – 26 . doi:10.1002/jat.1563. Google Scholar Crossref Search ADS PubMed WorldCat Green , B. T. , K. D. Welch, K. E. Panter, and S. T. Lee. 2013a . Plant toxins that affect nicotinic acetylcholine receptors: a review. Chem. Res. Toxicol . 26 : 1129 – 1138 . doi:10.1021/tx400166f. Google Scholar Crossref Search ADS WorldCat Greene , L. W. 2016 . Bill I. Kunkle Interdisciplinary Beef Symposium: assessing the mineral supplementation needs in pasture-based beef operations in the Southeastern United States. J. Anim. Sci . 94 : 5395 – 5400 . doi:10.2527/jas.2016-0871. Google Scholar Crossref Search ADS PubMed WorldCat Kay Dee Feed Company. 2013 . Kay Dee Ranger Talk . Retrieved from https://www.kaydeefeed.com/ranger_talk/June%202013.pdf Knowles , K. 1974 . An evaluation of larkspur poisoning in cattle and trampling damage that occurs during grazing on a summer range in eastern Idaho [MS thesis]. University of Idaho , Moscow . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Limaye , A. , R. -C. Yu, C. -C. Chou, J. -R. Liu, and K. -C. Cheng. 2018 . Protective and detoxifying effects conferred by dietary selenium and curcumin against AFB1-mediated toxicity in livestock: a review. Toxins 10 : 25 . doi:10.3390/toxins10010025. Google Scholar Crossref Search ADS WorldCat Logan , J. R. 1973 . Evaluation of a specifically-formulated supplement for the prevention of larkspur poisoning in cattle . [MS thesis]. Utah State University , Logan . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ortega , I. M. , F. C. Bryant, and D. L. Drawe. 1995 . Contrasts of esophageal-fistula versus bite-count techniques to determine cattle diets. J. Range Manag . 48 : 498 – 502 . doi:10.2307/4003059 Google Scholar Crossref Search ADS WorldCat Panter , K. E. , G. D. Manners, B. L. Stegelmeier, S. T. Lee, D. R. Gardner, M. H. Ralphs, J. A. Pfister, and L. F. James. 2002 . Larkspur poisoning: toxicology and alkaloid structure-activity relationships. Biochem. Syst. Ecol . 30 : 113 – 128 . doi:10.1016/S0305-1978(01)00123-5. Google Scholar Crossref Search ADS WorldCat Pfister , J. A. , D. R. Gardner, K. E. Panter, G. D. Manners, M. H. Ralphs, B. L. Stegelmeier, and T. K. Schoch. 1999 . Larkspur (Delphinium spp.) poisoning in livestock. J. Nat. Toxins 8 : 81 – 94 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Pfister , J. A. , B. T. Green, D. R. Gardner, K. D. Welch, D. Cook, and C. A. Stonecipher. 2018 . Grazing of Delphinium occidentale (duncecap larkspur) by susceptible and resistant cattle. Prof. Anim. Sci . 34 : 581 – 593 . doi:10.15232/pas.2018-01775. Google Scholar Crossref Search ADS WorldCat Pfister , J. A. , and G. D. Manners. 1991 . Mineral salt supplementation of cattle grazing tall larkspur-infested rangeland during drought. J. Range Manag . 44 : 105 – 111 . doi:10.2307/4002306 Google Scholar Crossref Search ADS WorldCat Pfister , J. A. , and G. D. Manners. 1995 . Effects of carbachol administration in cattle grazing tall larkspur-infested range. J. Range Manag . 48 : 343 – 349 . doi:10.2307/4002487 Google Scholar Crossref Search ADS WorldCat Pfister , J. A. , G. D. Manners, M. H. Ralphs, Z. X. Hong, and M. A. Lane. 1988 . Effects of phenology, site and rumen fill on tall larkspur consumption by cattle. J. Range Manag . 41 : 509 – 514 . doi:10.2307/3899528 Google Scholar Crossref Search ADS WorldCat Pfister , J. A. , K. E. Panter, G. D. Manners, and C. D. Cheney. 1994 . Reversal of tall larkspur (Delphinium barbeyi) poisoning in cattle with physostigmine. Vet. Hum. Toxicol . 36 : 511 – 514 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Pfister , J. A. , M. H. Ralphs, D. R. Gardner, B. L. Stegelmeier, G. D. Manners, K. E. Panter, and S. T. Lee. 2002 . Management of three toxic Delphinium species based on alkaloid concentrations. Biochem. Syst. Ecol . 30 : 129 – 138 . doi:10.1016/S0305-1978(01)00124-7. Google Scholar Crossref Search ADS WorldCat Pfister , J. A. , M. H. Ralphs, G. D. Manners, D. R. Gardner, K. W. Price, and L. F. James. 1997 . Early season grazing by cattle of tall larkspur (Delphinium spp.) infested rangeland. J. Range Manag . 50 : 391 – 398 . doi:10.2307/4003306 Google Scholar Crossref Search ADS WorldCat PRISM. 2021 . PRISM Climate Group. Available at: http://www.prism.oregonstate.edu Accessed 20 September 2021 . Rabbani , G. H. , S. K. Saha, M. Akhtar, F. Marni, A. K. Mitra, S. Ahmed, M. Alauddin, M. Bhattacharjee, S. Sultana, and A. K. Azad Chowdhury. 2003 . Antioxidants in detoxification of arsenic-induced oxidative injury in rabbits: preliminary results. J. Environ. Sci. Health A 38 : 273 – 287 . doi:10.1081/ESE-120016894. Google Scholar Crossref Search ADS WorldCat Ralphs , M. H. , G. D. Manners, J. A. Pfister, D. R. Gardner, and L. F. James. 1997 . Toxic alkaloid concentration in tall larkspur species in the western United States. J. Range Manag . 50 : 497 – 502 . doi:10.2307/4003704 Google Scholar Crossref Search ADS WorldCat Ralphs , M. H. , and J. A. Pfister. 1992 . Cattle diets in tall forb communities on mountain rangelands. J. Range Manag . 45 : 534 – 537 . doi:10.2307/4002566 Google Scholar Crossref Search ADS WorldCat Spears , J. W. , and W. P. Weiss. 2014 . Invited review: mineral and vitamin nutrition in ruminants. Prof. Anim. Sci . 30 : 180 – 191 . doi:10.15232/S1080-7446(15)30103-0. Google Scholar Crossref Search ADS WorldCat Sprinkle , J. E. , S. D. Baker, J. A. Church, J. R. Findlay, S. M. Graf, K. S. Jensen, S. K. Williams, C. M. Willmore, J. B. Lamb, and D. W. Hansen. 2018 . Case study: regional assessment of mineral element concentrations in Idaho forage and range grasses. Prof. Anim. Sci . 34 : 494 – 504 . doi:10.15232/pas.2017-01715. Google Scholar Crossref Search ADS WorldCat Sprinkle , J. E. , S. P. Cuneo, H. M. Frederick, R. M. Enns, D. W. Schafer, G. E. Carstens, S. B. Daugherty, T. H. Noon, B. M. Rickert, and C. Reggiardo. 2006 . Effects of a long-acting, trace mineral, reticulorumen bolus on range cow productivity and trace mineral profiles. J. Anim. Sci . 84 : 1439 – 1453 . doi:10.2527/2006.8461439x. Google Scholar Crossref Search ADS PubMed WorldCat Sprinkle , J. E. , D. W. Schafer, S. P. Cuneo, D. Tolleson, and R. M. Enns. 2011 . Effects of a long acting trace mineral rumen bolus upon range cow productivity. Proc. West. Sec. Am. Soc. Anim. Sci . 62 : 349 – 353 . Google Scholar OpenURL Placeholder Text WorldCat USDA NRCS [U.S. Department of Agriculture, Natural Resources Conservation Service]. 2021 . Loamy 16-22 ARTRV/FEID-PSSPS . – [accessed September 14, 2021 ]. https://edit.jornada.nmsu.edu/catalogs/esd/013X/R013XY005ID. Welch , K. D. , B. T. Green, D. R. Gardner, D. Cook, and J. A. Pfister. 2015 . The effect of administering multiple doses of tall larkspur (Delphinium barbeyi) to cattle. J. Anim. Sci . 93 : 4181 – 4188 . doi:10.2527/jas2015-9101. Google Scholar Crossref Search ADS PubMed WorldCat Welch , K. D. , J. A. Pfister, F. G. Lima, B. T. Green, and D. R. Gardner. 2013 . Effect of α7 nicotinic acetylcholine receptor agonists and antagonists on motor function in mice. Toxicol. Appl. Pharmacol . 266 : 366 – 374 . doi:10.1016/j.taap.2012.11.024. Google Scholar Crossref Search ADS PubMed WorldCat Published by Oxford University Press on behalf of the American Society of Animal Science 2022. This work is written by (a) US Government employee(s) and is in the public domain in the US. Published by Oxford University Press on behalf of the American Society of Animal Science 2022.

doi: 10.1093/jas/skac113pmid: 35537079

Junqueira, Vinícius Silva; Lourenco, Daniela; Masuda, Yutaka; Cardoso, Fernando Flores; Lopes, Paulo Sávio; Silva, Fabyano Fonseca e; Misztal, Ignacy

doi: 10.1093/jas/skac082pmid: 35289906

Abstract Efficient computing techniques allow the estimation of variance components for virtually any traditional dataset. When genomic information is available, variance components can be estimated using genomic REML (GREML). If only a portion of the animals have genotypes, single-step GREML (ssGREML) is the method of choice. The genomic relationship matrix (G) used in both cases is dense, limiting computations depending on the number of genotyped animals. The algorithm for proven and young (APY) can be used to create a sparse inverse of G (⁠ GAPY~-1 ⁠) with close to linear memory and computing requirements. In ssGREML, the inverse of the realized relationship matrix (H−1) also includes the inverse of the pedigree relationship matrix, which can be dense with a long pedigree, but sparser with short. The main purpose of this study was to investigate whether costs of ssGREML can be reduced using APY with truncated pedigree and phenotypes. We also investigated the impact of truncation on variance components estimation when different numbers of core animals are used in APY. Simulations included 150K animals from 10 generations, with selection. Phenotypes (h2 = 0.3) were available for all animals in generations 1–9. A total of 30K animals in generations 8 and 9, and 15K validation animals in generation 10 were genotyped for 52,890 SNP. Average information REML and ssGREML with G−1 and GAPY~-1 using 1K, 5K, 9K, and 14K core animals were compared. Variance components are impacted when the core group in APY represents the number of eigenvalues explaining a small fraction of the total variation in G. The most time-consuming operation was the inversion of G, with more than 50% of the total time. Next, numerical factorization consumed nearly 30% of the total computing time. On average, a 7% decrease in the computing time for ordering was observed by removing each generation of data. APY can be successfully applied to create the inverse of the genomic relationship matrix used in ssGREML for estimating variance components. To ensure reliable variance component estimation, it is important to use a core size that corresponds to the number of largest eigenvalues explaining around 98% of total variation in G. When APY is used, pedigrees can be truncated to increase the sparsity of H and slightly reduce computing time for ordering and symbolic factorization, with no impact on the estimates. Lay Summary The estimation of variance components is computationally expensive under large-scale genetic evaluations due to several inversions of the coefficient matrix. Variance components are used as parameters for estimating breeding values in mixed model equations (MME). However, resulting breeding values are not Best Linear Unbiased Predictions (BLUP) unless the variance components approach the true parameters. The increasing availability of genomic data requires the development of new methods for improving the efficiency of variance component estimations. Therefore, this study aimed to reduce the costs of single-step genomic REML (ssGREML) with the Algorithm for Proven and Young (APY) for estimating variance components with truncated pedigree and phenotypes using simulated data. In addition, we investigated the influence of truncation on variance components and genetic parameter estimates. Under APY, the size of the core group influences the similarity of breeding values and their reliability compared to the full genomic matrix. In this study, we found that to ensure reliable variance component estimation, it is required to consider a core size that corresponds to the number of largest eigenvalues explaining around 98% of the total variation in G to avoid biased parameters. In terms of costs, the use of APY slightly decreased the time for ordering and symbolic factorization with no impact on estimations. Introduction Restricted maximum likelihood (REML), described by Patterson and Thompson (1971), is a popular method for parameter estimation. Because it uses the mixed model equations (Henderson, 1975), it is resistant to selection bias, and efficient implementations are currently available. With the Average Information (AI) algorithm, convergence is often achieved in a few rounds. With traces obtained by sparse matrix factorization and inversion (Meyer, 1997), computing variance components is feasible even with large models. When genomic information is available, two versions of REML may be applicable. When only genotyped animals have phenotypes, genomic REML (GREML) can be applied with a genomic relationship matrix (G). In general, such a matrix is dense, and the cost of dense matrix operations would limit computations depending on the models. When only a fraction of animals are genotyped, a single-step genomic REML is applicable (ssGREML). In the latter, the realized relationship matrix (H) has dense blocks due to the genomic information, limiting the efficiency of sparse matrix operations. Lately, Masuda et al. (2015) developed a sparse matrix package YAMS that identifies dense blocks and computes them efficiently. For ssGREML, with genomic computation, such a package resulted in up to 100 times speedup, allowing four trait models with 20,000 genotyped animals (Masuda et al., 2015). In general, it is of interest to include many genotyped animals in parameter estimation and evaluations to account for genomic selection or pre-selection (Patry and Ducrocq, 2011). For instance, the greatest reliability in a single-step genomic BLUP was obtained using 50% of the heritability computed with a non-genomic REML (Misztal et al., 2017). The number of genotyped animals is increasing fast for some species. As an example, almost 3 million Holsteins have been genotyped in the United States (https://queries.uscdcb.com/Genotype/cur_freq.html). However, the cost of dense matrix operations with G in REML using YAMS is quadratic for memory and cubic for operations, which limits computations to around 50,000 animals. The genomic information has a limited dimensionality due to the limited effective population size (Stam, 1980; VanRaden, 2008; Misztal, 2016). Such dimensionality varied from 4,000 in pigs and chickens to 15,000 in Holsteins (Pocrnic et al., 2016c). Assuming limited dimensionality, the inverse of G (G-1)—as needed by REML—can be sparsely constructed using the APY algorithm, with close to linear memory and computing requirements. Subsequently, the inverses for over 2 million animals can be computed and stored (Tsuruta et al., 2021). However, the inverse of H also includes the inverse of a pedigree-based relationship matrix for genotyped animals (Aguilar et al., 2010). Such a matrix can be dense with a long pedigree, but it is sparser with a shorter pedigree. Thus, it could not be efficiently stored in large populations but had to be accommodated indirectly (Strandén and Mäntysaari, 2014; Masuda et al., 2017). The first purpose of this study was to find whether the costs of ssGREML can be reduced using the APY algorithm with truncated pedigree and phenotypes. We hypothesize the truncation could help us to preserve the system’s sparsity, given that APY G-1 is sparser than the inverse of the pedigree relationship matrices for deep pedigrees. The second purpose was to investigate to what extent such truncation influences variance components and heritability estimates when different numbers of core animals are used in APY. Material and Methods Animal care and use committee approval was not needed because data were simulated. Data simulation To evaluate the computational effectiveness of the proposed approach for estimating variance components using genomic information, we simulated data using the QMSim software (Sargolzaei et al., 2011). The simulator generated a historical population undergoing drift and mutation and a recent population undergoing selection. The historical population consisted of 1,000 generations with a constant size of 50,000 individuals. Then, 800 more generations were simulated where the number of individuals was reduced to 20,000, mimicking a bottleneck event. The recent population consisted of 20 males and 15,000 females randomly sampled from the last historical generation based on high phenotypic values. A total of 15,000 individuals were available for each non-overlapping generation. Parents were selected based on higher phenotypes and randomly mated along ten generations producing a litter size of 1 with an equal probability of being male or female. Moreover, we considered a sire replacement rate of 0.50 and a dam replacement rate of 0.20. Genomic information was available for 45,000 animals from generations 8 through 10 (three youngest generations). A total of 29 chromosomes of different lengths (ranging from 40 to 146 cM) were simulated. Biallelic markers (n = 52,890) were evenly spaced along the chromosomes with equal frequency in the first generation of the historical population. Potentially, a total of 1,242 quantitative trait loci (QTL) affected the trait were randomly sampled within chromosomes and explained all the additive genetic variation. The QTL allele effects were sampled from a Gamma distribution with a shape parameter of 0.4. The mutation rate for markers (recurrent mutation) and QTL was assumed to be equal to 2.5 × 10−5 per locus per generation (Solberg et al., 2008). The simulated trait had phenotypic variance and mean of 1.0, heritability and QTL heritability of 0.30, and residual variance of 0.70. The simulated phenotypes were composed of y=μ+u+e, where y is the vector of phenotypes, μ is the vector of overall mean, u is the vector of weighted sum of QTL effects (i.e., additive genetic effect or animal effect), and e is the vector of residuals. The standard error of estimates was small using 5 replicates during preliminary investigations of this study. Because of that, the results are based on one replicate. Genomic quality control A quality control procedure was implemented in the genomic data before the estimation of variance components. The adopted quality control criteria for SNP exclusion were the minor allele frequency (MAF) (<5%), genotype call rate (<90%), and monomorphic markers. The criteria to reject samples were call rate (<90%). After quality control, a total of 45,000 genotyped samples and 50,000 markers were retained for further analysis. Variance components Variance components were estimated using the average information (AI) REML algorithm as implemented in the AIREMLF90 software (Misztal et al., 2002), which was modified to incorporate the YAMS package (Masuda et al., 2014; Masuda et al., 2015). The incorporation of YAMS was essential for this kind of task when using genomic information. The package applies the supernodal method using multi-core optimized libraries (i.e., parallel computing). The most computationally expensive part of the variance component estimation is obtaining the inverse of the coefficient matrix used in traces. To that, efficient algorithms are used to invert large and sparse matrices, which are based on three steps: 1) ordering, 2) factorization (i.e., symbolic and numerical), and 3) sparse inversion. Ordering is not mandatory, but it saves a large amount of memory and time in the factorization step as it reduces the fill-in effect (zero elements in the original matrix could become nonzero elements in the factorized matrix). This effect can be minimized by ordering using appropriate techniques. In the next step, the coefficient matrix (LHS of the mixed model equations) is factorized into two triangular matrices by LU decomposition—L matrix. Finally, the Takahashi algorithm can be used for inversion. The supernodal method is expected to provide faster inversions because they find and process dense blocks in sparse matrices. Note that LHS inversion is only required to estimate variance components or compute prediction error variance (PEV, obtained from diagonal elements of an inverted LHS). If the objective is to solve the system of equations to obtain breeding values, iterative methods as the preconditioned conjugate gradient (Lidauer et al., 1999; Tsuruta et al., 2001) can be efficiently applied. The model used to estimate variance components was based on the single-step method, in which the inverse of the realized relationship matrix (⁠ H−1 ⁠) is used in the mixed model equations instead of A−1 ⁠. Single-step genomic BLUP (ssGBLUP) is used for breeding value estimation, whereas ssGREML is used for variance components estimation. The inversion of H is computed as follows (Aguilar et al., 2010): H−1=A−1+[000GAPY−1−A22−1], where A−1 is the inverse of the pedigree relationship matrix and A22−1 is the inverse of the pedigree relationship matrix for genotyped animals, computed by the algorithm described in Colleau (2002). The genomic relationship matrix (⁠ G ⁠) was computed as follows: G=ZZ′2∑pj(1−pj), where Z is the matrix of gene content centered by the allele frequencies of genotyped individuals, and pj is the allele frequency of SNP j. Inbreeding coefficients were considered when constructing the three relationship matrices. This provides a better equivalence between genomic and pedigree-based relationship matrices, leading to a more similar genetic base (Aguilar et al., 2020). The GAPY−1 is the inverse of the genomic relationship matrix obtained using the algorithm for proven and young (APY) (Misztal et al., 2014; Misztal, 2016). This algorithm considers that genotyped individuals are arbitrarily divided into core (c) and noncore (n). Breeding values for noncore (⁠ un ⁠) can be described as a linear function of breeding values of core (⁠ uc ⁠): un=Pnuc+φn, where Pn=Zn(Z′cZc+Iα)−1Z′c is a matrix that relates breeding values of noncore and core, and φn=εn+Znεa is the Mendelian sampling term with non-diagonal variance, which can be approximated to diagonal if the number of core individuals is greater or equal to the number of SNPs (⁠ εn≈0) ⁠. The term εn represents the fraction of breeding values not explained by SNPs with var(ε)=Iσε2 ⁠. In cases where the number of core is large enough, breeding values of noncore depend only on breeding values of core [see Misztal (2016) for additional details]. The inverse of GAPY is constructed as follows: GAPY−1=[I−Pcc′−Pcn0I][Mcc−100Mnn−1][I−Pcc0−PncI]. If Gcc−1=(I−P′cc)Mcc−1(I−Pcc) is known, the complete inverse can be simplified to GAPY−1=[Gcc−1000]+[−PcnI]Mnn−1[−PncI], where Pcc=GccGcc−1 ⁠, Mcc(nn)=diag{gi,i−pi,1:i−1gi,1:i−1′} for individual i in the core (noncore) group. Because GAPY−1 is conditioned only on the genotypic information of core animals, the matrix is sparser than the full G−1 regularly used in ssGBLUP (Misztal, 2016). Note that the covariance between two noncore individuals is null, but variances are stored in the matrix. The construction of the genomic matrix using APY in BLUPF90 software can be done in two possible implementations. The first construction builds a single matrix for all core and noncore. The second construction builds the genomic matrix in blocks and it aims to save computing memory as it require less operations than single matrix (Masuda et al., 2016). Currently, the single matrix construction is implemented for variance component estimation. Scenarios The scenarios below were built to evaluate the impact of the 1) size of the core group in APY, the 2) influence of skipping zero elements from the LHS under different amounts of pedigree and phenotypic data used in variance components estimation, and the 3) influence of zero elements in the Mixed Model Equations (MME). Core group of different sizes Pocrnic et al. (2016a) evaluated the prediction accuracy using APY in simulation tests. The authors suggested that the greatest accuracy was found by selecting the number of core individuals equal to the number of largest eigenvalues explaining 98% of G (a number from now on referred to as eigen98). This study tested core groups of different sizes to evaluate the impact on variance components and heritability estimates. A total of four scenarios were tested by allocating 1K (one thousand), 5K, 9K, and 14K randomly sampled out of 45,000 genotyped individuals. For each of those scenarios, the largest variation explained was 72.03% (eigen70), 91.09% (eigen90), 95.70% (eigen95), and 98.07% (eigen98), respectively. For computational reasons, the singular value decomposition of Z was calculated instead of the eigenvalue decomposition of G ⁠. Evaluating the influence of pedigrees and phenotypes Using GAPY−1 helps us to reduce computing time for genomic predictions because of its sparsity (Fragomeni et al., 2015; Masuda et al., 2016); however, in the single-step approach, the H−1 also contains A−1 and A22−1 ⁠, which are relatively dense. The APY method was earlier applied to the construction of A22−1 without success (Breno Fragomeni, personal communication). Although the sparsity of A22−1 may not be a requirement for genomic predictions, it becomes essential for reducing computing time for variance components estimation to follow the sparsity of GAPY−1 ⁠. A reduction in the number of generations was attempted to increase the sparsity in A−1 and A22−1 ⁠. A total of seven different scenarios were designed, differing on the number of pedigree generations used for variance components estimation. Reduction in the generations of phenotypes was also used to follow pedigree incompleteness and avoid bias. The scenarios were designed to mimic a real situation where the actual founder population is usually unknown. Only three genotyped generations (45,000 most recent animals) were kept in the genomic file for further analyses. Subsequent scenarios were constructed by removing one generation of phenotypes and pedigree at a time, from the oldest to the youngest animals. The influence of zero elements in the MME Lastly, a scenario aimed to evaluate the impact of discarding zero elements from the LHS of MME on computing performance and variance components estimation. For that, the OPTION skip_zero_in_dense_matrix was used in AIREMLF90 (Misztal et al., 2014) to store only non-zero elements of GAPY−1−A22−1 ⁠. When this option was used, the scenario was termed “Reduced,” and otherwise “Full.” RESULTS AND DISCUSSION Previous studies have investigated the properties of APY, including its implementation for large-scale genomic evaluations (Fragomeni et al., 2015; Lourenco et al., 2015; Masuda et al., 2016) and its efficiency in real and simulated populations with different effective population sizes (Pocrnic et al., 2016b; Pocrnic et al., 2016c). Bradford et al. (2017) studied the impact of different core definitions, and Misztal et al. (2020) evaluated the GEBV fluctuation when changing the core group in APY. Additionally, Vandenplas et al. (2018) investigated the impact of using APY on GEBV estimation in crossbreeding schemes; Hidalgo et al. (2021) compared the GEBV variation due to the inclusion of new data and changing the APY core animals. Finally, Lourenco et al. (2018) studied the impact of using GAPY−1 instead of G−1 on the estimation of SNP effects. Our study evaluated the feasibility of using APY for variance components estimation, the impact of removing generations of pedigree and phenotypic data on computing time, and the influence of using a different number of core animals to construct the genomic matrix. Variance components were estimated using AIREML modified to incorporate the YAMS package for sparse matrix calculations (Masuda et al., 2014). Heritability estimates and computing performance Heritability, residual variance, and additive variance estimated using a different number of generations in the pedigree and cores sizes in APY are shown in Figures 1–3. The standard deviation of variance components and heritability across generations is shown in Table 1. Because the simulation involved a certain level of selection, the expected heritability should slightly deviate from the simulated value of 0.3. Therefore, the scenario with 10 generations of data (i.e., full pedigree and full phenotypes) was used as a benchmark. Table 1. Standard deviation of variance components and heritability calculated across generations using complete (Full) mixed model equations (MME) and reduced MME after skipping zero elements (Reduced) Parameter1 . Core2 . Scenario . Full . Reduced . σa2 eigen70 0.037 0.037 eigen90 0.011 0.013 eigen95 0.008 0.008 eigen98 0.005 0.005 σe2 eigen70 0.028 0.028 eigen90 0.007 0.007 eigen95 0.005 0.005 eigen98 0.000 0.004 h2 eigen70 0.032 0.032 eigen90 0.011 0.011 eigen95 0.005 0.005 eigen98 0.005 0.005 Parameter1 . Core2 . Scenario . Full . Reduced . σa2 eigen70 0.037 0.037 eigen90 0.011 0.013 eigen95 0.008 0.008 eigen98 0.005 0.005 σe2 eigen70 0.028 0.028 eigen90 0.007 0.007 eigen95 0.005 0.005 eigen98 0.000 0.004 h2 eigen70 0.032 0.032 eigen90 0.011 0.011 eigen95 0.005 0.005 eigen98 0.005 0.005 σa2 ⁠, additive variance; σe2 ⁠, residual variance; h2 ⁠, heritability. Scenarios testing the allocation of 1K, 5K, 9K, and 14K individuals in core group to explain 72.03% (eigen70), 91.09% (eigen90), 95.70% (eigen95), and 98.07% (eigen98) of the variation in the genomic matrix (G), respectively, using the Algorithm of Proven and Young (APY). Open in new tab Table 1. Standard deviation of variance components and heritability calculated across generations using complete (Full) mixed model equations (MME) and reduced MME after skipping zero elements (Reduced) Parameter1 . Core2 . Scenario . Full . Reduced . σa2 eigen70 0.037 0.037 eigen90 0.011 0.013 eigen95 0.008 0.008 eigen98 0.005 0.005 σe2 eigen70 0.028 0.028 eigen90 0.007 0.007 eigen95 0.005 0.005 eigen98 0.000 0.004 h2 eigen70 0.032 0.032 eigen90 0.011 0.011 eigen95 0.005 0.005 eigen98 0.005 0.005 Parameter1 . Core2 . Scenario . Full . Reduced . σa2 eigen70 0.037 0.037 eigen90 0.011 0.013 eigen95 0.008 0.008 eigen98 0.005 0.005 σe2 eigen70 0.028 0.028 eigen90 0.007 0.007 eigen95 0.005 0.005 eigen98 0.000 0.004 h2 eigen70 0.032 0.032 eigen90 0.011 0.011 eigen95 0.005 0.005 eigen98 0.005 0.005 σa2 ⁠, additive variance; σe2 ⁠, residual variance; h2 ⁠, heritability. Scenarios testing the allocation of 1K, 5K, 9K, and 14K individuals in core group to explain 72.03% (eigen70), 91.09% (eigen90), 95.70% (eigen95), and 98.07% (eigen98) of the variation in the genomic matrix (G), respectively, using the Algorithm of Proven and Young (APY). Open in new tab Figure 1. Open in new tabDownload slide Heritability calculated from one replicate of simulation considering a different number of generations with pedigree and phenotypic data and a different number of core individuals in the Algorithm for Proven and Young (APY). Two scenarios were considered, where zeros were stored (Full) or not (Reduced). Error bars represent the standard error of prediction under restricted maximum likelihood (REML). Scenarios testing the allocation of 1K, 5K, 9K, and 14K individuals in the core group to explain 72.03% (eigen70), 91.09% (eigen90), 95.70% (eigen95), and 98.07% (eigen98) of the variation in the genomic matrix (G), respectively. Figure 2. Open in new tabDownload slide Residual variance calculated from one replicate of simulation considering a different number of generations with pedigree and phenotypic data and a different number of core individuals in the Algorithm for Proven and Young (APY). Two scenarios were considered, where zeros were stored (Full) or not (Reduced). Error bars represent the standard error of prediction under restricted maximum likelihood (REML). Scenarios testing the allocation of 1K, 5K, 9K, and 14K individuals in core group to explain 72.03% (eigen70), 91.09% (eigen90), 95.70% (eigen95), and 98.07% (eigen98) of the variation in the genomic matrix (G), respectively. Figure 3. Open in new tabDownload slide Additive variance calculated from one replicate of simulation considering a different number of generations with pedigree and phenotypic data and a different number of core individuals in the Algorithm for Proven and Young (APY). Two scenarios were considered, where zeros were stored (Full) or not (Reduced). Error bars represent the standard error of prediction under restricted maximum likelihood (REML). Scenarios testing the allocation of 1K, 5K, 9K, and 14K individuals in core group to explain 72.03% (eigen70), 91.09% (eigen90), 95.70% (eigen95), and 98.07% (eigen98) of the variation in the genomic matrix (G), respectively. In general, the variance components and heritability estimates approached the simulated values as the number of core approached eigen98. The scenario using 1K individuals (i.e., eigen70) in the core was the most sensitive to removing generations, suggesting that variance components are highly impacted when the core group in APY represents the number of eigenvalues explaining a smaller fraction of the total variation in G. From a prediction accuracy standpoint, a similar behavior was also observed in other studies (Pocrnic et al., 2016a; Pocrnic et al., 2016c); however, the impact on variance components had not been investigated before. Although pedigrees were more limited after removing a few generations of data, the combination of pedigree and genomic information and the use of a core size equal to eigen98 controlled the bias in variance components and heritability estimation. Small fluctuations on variance components were observed when retaining only 4 to 6 generations of pedigree and phenotypes with a core size equal to eigen98. In these scenarios, the difference in heritability was almost nonexistent; this was also true when comparing Full and Reduced models. The ratio σe2/σa2 is important when predicting breeding values using the mixed model equations as it is the shrinkage factor for additive effects. The variability of the ratio under different core sizes is shown in Figure 4. As the core size approached eigen98, the ratio became closer to the simulated value of 2.33. Additionally, the ratio became less influenced by the number of generations used to estimate the variance components as the core size approached eigen98. Reliable variance component estimates (or at least their ratio and heritability) are of great importance to ensure the accurate prediction of breeding values. The resulting breeding values are not BLUP unless the true variances are known or are approaching the true parameters (Kennedy, 1981). Figure 4. Open in new tabDownload slide Distribution of the variance ratio (σe2/σa2) across a different number of generations (i.e., 10, 9, 8, 7, 6, 5, and 4) with pedigree and phenotypic data using different sizes for the core group in the Algorithm for Proven and Young (APY). Two scenarios were considered, where zeros were stored (Full) or not (Reduced). Error bars extend from the hinge to the largest (smallest) no further than 1.5 times the distance between the first and third quartiles. Ratios were estimated under restricted maximum likelihood (REML). Scenarios testing the allocation of 1K, 5K, 9K, and 14K individuals in core group to explain 72.03% (eigen70), 91.09% (eigen90), 95.70% (eigen95), and 98.07% (eigen98) of the variation in the genomic matrix (G), respectively. The adoption of a core group that explains less than eigen98 affected the ability to represent all the independent chromosome segments segregating in the population, traceback gene frequencies, and consequently, accurately establish covariances between genotypic values. In this study, we might have three different sources of changes for genetic variances. The first source is related to the lack of relationships because generations were sequentially removed in different scenarios. Unknown relationships (i.e., incorrect base population definition) affect the estimation of Mendelian sampling variance in different intensities depending on the number of known parents. If both parents are unknown, Mendelian sampling is equal to 0.5 σa2 ⁠, and if only one parent is known, it equals [0.75 − 0.25×fp]σa2 ⁠, where fp is the inbreeding coefficient of a parent (Henderson, 1976). Under mixed models, offspring breeding values are estimated as a function of parent breeding values and Mendelian sampling. Thus, all individuals with unknown relationships are treated as samples from the base population with average breeding value of 0 and common variance σa2 ⁠. The second source of change in genetic variance is the presence of selection over generations, which affects the distribution of sire and dam breeding values. Unfortunately, it is impossible to identify the contribution of each factor separately because this study was not designed for that purpose. The third source of genetic variation, which is the aim of this study, is the intentional use of a sparse representation of G−1 ⁠, i.e., GAPY−1 ⁠. In APY, it is intrinsically assumed that the complete genome is divided into many independent chromosome segments (ICS) containing non-redundant genomic information. The number of ICS is a statistical concept that depends on the effective population size and the genome length (Stam, 1980). The consequence of this assumption is that a small error in variance components estimation can be observed by building the core group considering the dimensionality of G as a function of the number of eigenvalues explaining a certain proportion of variance. For example, if GAPY−1 is built based on the number of core animals equal to that of eigenvalues explaining 98% of the variance in G, the assumed error is 2% (Misztal et al., 2020). Results from the current study add a new dimension to the factors driving the estimation of reliable variance components in the genomic era. Thus, if the definition of the core group considers the genetic architecture of the population, G might contain all the genetic information necessary to estimate reliable variance components (Junqueira et al., 2017; Junqueira et al., 2020). In addition to the factors evaluated in this study, Cesarani et al. (2019) have found that the selection design and genotyping structure can influence bias in estimating variance components. Computing resources Nowadays, much effort has been placed on developing faster and computationally feasible methods for a virtually unlimited number of genotyped individuals. Using large-scale datasets becomes more problematic when the objective is to estimate variance components. This is because most algorithms require several rounds of inversion of the LHS of MME before the convergence is reached. During computations, factorization and inversion are the most demanding steps in the REML estimation. The possibility to combine APY to compute a sparse representation of G−1 ⁠, data reduction, and YAMS (i.e., dense blocks operation) (Masuda et al., 2014; Masuda et al., 2015) seems computationally beneficial. In this study, we evaluated the factors impacting the timing required for computational operations. Figure 5 shows the average computing time, relative to total (i.e., in percentage), required for ordering, factorization (symbolic and numerical), and sparse inversion with data reduction (pedigree and phenotypes). The most time-consuming operation was the inversion, which took more than 50% of the total time. This was expected because matrix inversion has a cubic computing cost. Next, numerical factorization consumed nearly 30% of the total computing time, whereas ordering and symbolic factorization took approximately 9% and 7.5%, respectively. Skipping zero elements in the MME did not improve the computing time of any of the inverse operations. Figure 5. Open in new tabDownload slide Average timing in percentage (ratio between total timing) relative to each operation used in the process of matrix inversion. The average timing and error bars (standard deviation) were calculated across scenarios using a different number of generations in the pedigree and phenotypic and core sizes. The x-axis represents the steps required to invert matrices: finding the ordering, symbolic factorization (Symbolic Fact., setting up the data structure), numerical factorization (Numerical Fact.), and sparse inversion. Two scenarios were considered, where zeros were stored (Full) or not (Reduced). A detailed description of the computing time required by each step after data removal is shown in Figure 6. The descriptive statistics of computing time savings across generations is shown in Table 2. Ordering showed the most prominent timing decrease due to data removal, followed by symbolic factorization among the four steps. On average, a 7% decrease in the computing time for ordering was observed by removing each generation of data. During MME computations, ordering and symbolic factorization are not mandatory. These operations are mainly implemented to reduce computing time for numerical factorization and inversion. As more genotypes and/or pedigree records are included in the model, the time required for numerical factorization and sparse inversion increases. Using a simulated dataset with GAPY−1 and YAMS, we observed an opposite behavior where a shorter pedigree sometimes caused an increase in computing time for the numerical factorization and sparse inversion operations. In these operations, there were no gains in computing performance due to data removal, as shown by the regression slope, which was close to 0 (Table 2). The greatest savings were around 10% when using six generations of pedigree and phenotypic data. It is known that numerical factorization and sparse inversion are the most demanding operations in REML computations. The fact that the required time for these operations was not reduced can be explained by the creation of nonzero elements not present in the coefficient matrix before the numerical factorization is done. Those elements are known as “fill-in elements.” Table 2. Descriptive statistics of computing time savings for the matrix operations and the slope of a regression of computing time on generations after removing pedigree and phenotypic data Core Size1 . Matrix Operation . Full . Reduced . Min (%) . Mean (%) . Max (%) . SD (%)2 . Slope3 . 4 . Min (%) . Mean (%) . Max (%) . SD (%) . Slope . . eigen70 Ordering 1.16 24.58 50.94 16.98 −0.07 ** 9.10 23.95 52.47 16.99 −0.08 ** Symbolic Factorization 0.61 4.77 16.22 6.04 −0.02 ** 0.08 4.57 18.44 6.92 −0.02 * Numerical Factorization 2.38 7.47 16.92 4.93 −0.02 ns 3.36 6.57 16.32 4.97 −0.02 * Sparse Inversion 4.67 8.08 18.25 5.08 −0.02 ** 0.59 5.80 16.47 5.71 −0.01 ns eigen90 Ordering 21.35 32.13 42.88 8.60 −0.06 ** 13.61 26.58 42.48 11.19 −0.07 ** Symbolic Factorization 4.98 9.24 13.06 3.08 −0.02 ** 3.10 5.50 9.97 2.89 −0.01 ** Numerical Factorization 2.13 6.21 11.34 3.47 −0.00 ns 3.22 6.17 8.41 2.22 −0.00 ** Sparse Inversion 4.39 6.81 8.52 1.48 −0.00 ns 2.52 5.33 8.51 2.51 −0.00 ns eigen95 Ordering 7.94 21.40 36.80 11.13 −0.06 ** 6.65 24.55 41.48 13.99 −0.07 ns Symbolic Factorization 2.76 6.39 10.33 2.97 −0.01 ** 2.48 5.19 7.43 2.20 −0.01 ns Numerical Factorization 4.96 7.26 9.98 1.82 −0.00 ns 3.45 7.63 10.33 2.93 −0.00 ns Sparse Inversion 0.07 5.52 9.08 3.38 −0.00 ns 3.77 6.59 10.14 2.82 −0.00 ns eigen98 Ordering 25.04 38.19 49.13 9.85 −0.07 ** 15.79 30.57 41.97 11.27 −0.07 ** Symbolic Factorization 8.28 16.33 19.63 4.61 −0.03 ** 1.26 5.79 9.71 3.72 −0.02 ** Numerical Factorization 5.92 10.15 13.99 2.89 −0.00 ns 4.32 7.91 11.39 2.35 −0.00 ns Sparse Inversion 5.34 8.09 10.32 2.14 −0.01 ns 2.85 5.88 9.32 2.25 −0.00 ns Core Size1 . Matrix Operation . Full . Reduced . Min (%) . Mean (%) . Max (%) . SD (%)2 . Slope3 . 4 . Min (%) . Mean (%) . Max (%) . SD (%) . Slope . . eigen70 Ordering 1.16 24.58 50.94 16.98 −0.07 ** 9.10 23.95 52.47 16.99 −0.08 ** Symbolic Factorization 0.61 4.77 16.22 6.04 −0.02 ** 0.08 4.57 18.44 6.92 −0.02 * Numerical Factorization 2.38 7.47 16.92 4.93 −0.02 ns 3.36 6.57 16.32 4.97 −0.02 * Sparse Inversion 4.67 8.08 18.25 5.08 −0.02 ** 0.59 5.80 16.47 5.71 −0.01 ns eigen90 Ordering 21.35 32.13 42.88 8.60 −0.06 ** 13.61 26.58 42.48 11.19 −0.07 ** Symbolic Factorization 4.98 9.24 13.06 3.08 −0.02 ** 3.10 5.50 9.97 2.89 −0.01 ** Numerical Factorization 2.13 6.21 11.34 3.47 −0.00 ns 3.22 6.17 8.41 2.22 −0.00 ** Sparse Inversion 4.39 6.81 8.52 1.48 −0.00 ns 2.52 5.33 8.51 2.51 −0.00 ns eigen95 Ordering 7.94 21.40 36.80 11.13 −0.06 ** 6.65 24.55 41.48 13.99 −0.07 ns Symbolic Factorization 2.76 6.39 10.33 2.97 −0.01 ** 2.48 5.19 7.43 2.20 −0.01 ns Numerical Factorization 4.96 7.26 9.98 1.82 −0.00 ns 3.45 7.63 10.33 2.93 −0.00 ns Sparse Inversion 0.07 5.52 9.08 3.38 −0.00 ns 3.77 6.59 10.14 2.82 −0.00 ns eigen98 Ordering 25.04 38.19 49.13 9.85 −0.07 ** 15.79 30.57 41.97 11.27 −0.07 ** Symbolic Factorization 8.28 16.33 19.63 4.61 −0.03 ** 1.26 5.79 9.71 3.72 −0.02 ** Numerical Factorization 5.92 10.15 13.99 2.89 −0.00 ns 4.32 7.91 11.39 2.35 −0.00 ns Sparse Inversion 5.34 8.09 10.32 2.14 −0.01 ns 2.85 5.88 9.32 2.25 −0.00 ns The benchmark is the model using full pedigree and phenotypic data. The comparison is based on using core group of different sizes in algorithm for proven and young (APY), and based on a full mixed model equations (Full) and a reduced mixed model equations after skipping zero elements (Reduced). Scenarios testing the allocation of 1K, 5K, 9K, and 14K individuals in core group to explain 72.03% (eigen70), 91.09% (eigen90), 95.70% (eigen95), and 98.07% (eigen98) of the variation in the genomic matrix (G), respectively. Standard deviation. Slope of a regression of computing time on generations. Slope statistical significance: *P < 0.05, **P < 0.10; ns, not significant. Open in new tab Table 2. Descriptive statistics of computing time savings for the matrix operations and the slope of a regression of computing time on generations after removing pedigree and phenotypic data Core Size1 . Matrix Operation . Full . Reduced . Min (%) . Mean (%) . Max (%) . SD (%)2 . Slope3 . 4 . Min (%) . Mean (%) . Max (%) . SD (%) . Slope . . eigen70 Ordering 1.16 24.58 50.94 16.98 −0.07 ** 9.10 23.95 52.47 16.99 −0.08 ** Symbolic Factorization 0.61 4.77 16.22 6.04 −0.02 ** 0.08 4.57 18.44 6.92 −0.02 * Numerical Factorization 2.38 7.47 16.92 4.93 −0.02 ns 3.36 6.57 16.32 4.97 −0.02 * Sparse Inversion 4.67 8.08 18.25 5.08 −0.02 ** 0.59 5.80 16.47 5.71 −0.01 ns eigen90 Ordering 21.35 32.13 42.88 8.60 −0.06 ** 13.61 26.58 42.48 11.19 −0.07 ** Symbolic Factorization 4.98 9.24 13.06 3.08 −0.02 ** 3.10 5.50 9.97 2.89 −0.01 ** Numerical Factorization 2.13 6.21 11.34 3.47 −0.00 ns 3.22 6.17 8.41 2.22 −0.00 ** Sparse Inversion 4.39 6.81 8.52 1.48 −0.00 ns 2.52 5.33 8.51 2.51 −0.00 ns eigen95 Ordering 7.94 21.40 36.80 11.13 −0.06 ** 6.65 24.55 41.48 13.99 −0.07 ns Symbolic Factorization 2.76 6.39 10.33 2.97 −0.01 ** 2.48 5.19 7.43 2.20 −0.01 ns Numerical Factorization 4.96 7.26 9.98 1.82 −0.00 ns 3.45 7.63 10.33 2.93 −0.00 ns Sparse Inversion 0.07 5.52 9.08 3.38 −0.00 ns 3.77 6.59 10.14 2.82 −0.00 ns eigen98 Ordering 25.04 38.19 49.13 9.85 −0.07 ** 15.79 30.57 41.97 11.27 −0.07 ** Symbolic Factorization 8.28 16.33 19.63 4.61 −0.03 ** 1.26 5.79 9.71 3.72 −0.02 ** Numerical Factorization 5.92 10.15 13.99 2.89 −0.00 ns 4.32 7.91 11.39 2.35 −0.00 ns Sparse Inversion 5.34 8.09 10.32 2.14 −0.01 ns 2.85 5.88 9.32 2.25 −0.00 ns Core Size1 . Matrix Operation . Full . Reduced . Min (%) . Mean (%) . Max (%) . SD (%)2 . Slope3 . 4 . Min (%) . Mean (%) . Max (%) . SD (%) . Slope . . eigen70 Ordering 1.16 24.58 50.94 16.98 −0.07 ** 9.10 23.95 52.47 16.99 −0.08 ** Symbolic Factorization 0.61 4.77 16.22 6.04 −0.02 ** 0.08 4.57 18.44 6.92 −0.02 * Numerical Factorization 2.38 7.47 16.92 4.93 −0.02 ns 3.36 6.57 16.32 4.97 −0.02 * Sparse Inversion 4.67 8.08 18.25 5.08 −0.02 ** 0.59 5.80 16.47 5.71 −0.01 ns eigen90 Ordering 21.35 32.13 42.88 8.60 −0.06 ** 13.61 26.58 42.48 11.19 −0.07 ** Symbolic Factorization 4.98 9.24 13.06 3.08 −0.02 ** 3.10 5.50 9.97 2.89 −0.01 ** Numerical Factorization 2.13 6.21 11.34 3.47 −0.00 ns 3.22 6.17 8.41 2.22 −0.00 ** Sparse Inversion 4.39 6.81 8.52 1.48 −0.00 ns 2.52 5.33 8.51 2.51 −0.00 ns eigen95 Ordering 7.94 21.40 36.80 11.13 −0.06 ** 6.65 24.55 41.48 13.99 −0.07 ns Symbolic Factorization 2.76 6.39 10.33 2.97 −0.01 ** 2.48 5.19 7.43 2.20 −0.01 ns Numerical Factorization 4.96 7.26 9.98 1.82 −0.00 ns 3.45 7.63 10.33 2.93 −0.00 ns Sparse Inversion 0.07 5.52 9.08 3.38 −0.00 ns 3.77 6.59 10.14 2.82 −0.00 ns eigen98 Ordering 25.04 38.19 49.13 9.85 −0.07 ** 15.79 30.57 41.97 11.27 −0.07 ** Symbolic Factorization 8.28 16.33 19.63 4.61 −0.03 ** 1.26 5.79 9.71 3.72 −0.02 ** Numerical Factorization 5.92 10.15 13.99 2.89 −0.00 ns 4.32 7.91 11.39 2.35 −0.00 ns Sparse Inversion 5.34 8.09 10.32 2.14 −0.01 ns 2.85 5.88 9.32 2.25 −0.00 ns The benchmark is the model using full pedigree and phenotypic data. The comparison is based on using core group of different sizes in algorithm for proven and young (APY), and based on a full mixed model equations (Full) and a reduced mixed model equations after skipping zero elements (Reduced). Scenarios testing the allocation of 1K, 5K, 9K, and 14K individuals in core group to explain 72.03% (eigen70), 91.09% (eigen90), 95.70% (eigen95), and 98.07% (eigen98) of the variation in the genomic matrix (G), respectively. Standard deviation. Slope of a regression of computing time on generations. Slope statistical significance: *P < 0.05, **P < 0.10; ns, not significant. Open in new tab Figure 6. Open in new tabDownload slide Timing (in seconds) relative to each operation to invert matrices using a different number of generations in the pedigree and phenotypes and a different number of core animals in the computation of G-1 with the Algorithm for Proven and Young (APY). Matrix inversion steps: finding the ordering (Ordering), symbolic factorization (Symbolic Fact.), numerical factorization (Numerical Fact.), and sparse inversion. Two scenarios were considered, where zeros were stored (Full) or not (Reduced). Scenarios testing the allocation of 1K, 5K, 9K, and 14K individuals in core group to explain 72.03% (eigen70), 91.09% (eigen90), 95.70% (eigen95), and 98.07% (eigen98) of the variation in the genomic matrix (G), respectively. Consequently, extra calculations are needed, obviously increasing the amount of time to complete the sparse inversion. There are several efforts in developing faster algorithms focused on typical nonzero structures in sparse matrices. The sparse matrix algorithm in YAMS uses supernodal techniques (i.e., common nonzero pattern between adjacent columns) to speed-up computations. Computing time might be significantly improved compared to other sparse matrix packages (e.g., FSPAK) because the memory hierarchy is more effectively exploited in dense operations, and multiple columns within a submatrix are simultaneously updated (Masuda et al., 2014). Conclusions The algorithm for proven and young (APY) can be successfully applied to create the inverse of the genomic relationship matrix used in single-step genomic restricted maximum likelihood for estimating variance components. To ensure reliable variance component estimation, it is important to use a core size that corresponds to the number of largest eigenvalues explaining around 98% of total variation in G. When APY is used, pedigrees can be truncated to increase the sparsity of H and slightly reduce computing time for ordering and symbolic factorization, with no impact on the estimates. A reduction in computing time for numerical factorization and sparse inversion is unlike because of the fill-in elements effect. The savings in computing time for estimating variance components is far less than the expected efficiency that APY has shown compared to the use of regular G−1 for breeding values estimation. This inefficiency is because the block implementation of APY is still not possible for variance components estimation. Abbreviations Abbreviations A pedigree relationship matrix AIREML average information; estricted maximum likelihood APY algorithm for proven and young BLUP best linear unbiased prediction EBV estimated breeding value G genomic matrix Gapy genomic matrix created using APY GEBV genomic enhanced breeding value GREML genomic restricted maximum likelihood IOD iteration on data LHS left hand side of mixed model equations MME mixed model equations QTL quantitative trait loci REML restricted maximum likelihood ssGBLUP single step genomic BLUP ssGREML single step genomic restricted maximum likelihood YAMS yet another MME solver Conflict of Interest Statement The authors declare that they do not have any conflict of interest. Literature Cited Aguilar , I. , E. N. Fernandez, A. Blasco, O. Ravagnolo, and A. Legarra. 2020 . Effects of ignoring inbreeding in model-based accuracy for BLUP and SSGBLUP . J. Anim. Breed. Genet . 137 : 356 – 364 . doi:10.1111/jbg.12470. Google Scholar Crossref Search ADS PubMed WorldCat Aguilar , I. , I. Misztal, D. L. Johnson, A. Legarra, S. Tsuruta, and T. J. Lawlor. 2010 . Hot topic: a unified approach to utilize phenotypic, full pedigree, and genomic information for genetic evaluation of holstein final score . J. Dairy Sci . 93 : 743 – 752 . doi:10.3168/jds.2009-2730. Google Scholar Crossref Search ADS PubMed WorldCat Bradford , H. , I. Pocrnić, B. Fragomeni, D. Lourenco, and I. Misztal. 2017 . Selection of core animals in the algorithm for proven and young using a simulation model . J. Anim. Breed. Genet . 134 : 545 – 552 . doi:10.1111/jbg.12276. Google Scholar Crossref Search ADS PubMed WorldCat Cesarani , A. , G. Gaspa, F. Correddu, M. Cellesi, C. Dimauro, and N. Macciotta. 2019 . Genomic selection of milk fatty acid composition in sarda dairy sheep: effect of different phenotypes and relationship matrices on heritability and breeding value accuracy . J. Dairy Sci . 102 : 3189 – 3203 . doi:10.3168/jds.2018-15333. Google Scholar Crossref Search ADS PubMed WorldCat Colleau , J. J. 2002 . An indirect approach to the extensive calculation of relationship coefficients . Genet. Sel. Evol . 34 : 409 . doi:10.1186/1297-9686-34-4-409. Google Scholar Crossref Search ADS PubMed WorldCat Fragomeni , B. O. , D. A. L. Lourenco, S. Tsuruta, Y. Masuda, I. Aguilar, A. Legarra, T. J. Lawlor, and I. Misztal. 2015 . Hot topic: use of genomic recursions in single-step genomic best linear unbiased predictor (BLUP) with a large number of genotypes . J. Dairy Sci . 98 : 4090 – 4094 . doi:10.3168/jds.2014-9125. Google Scholar Crossref Search ADS PubMed WorldCat Henderson , C. R. 1975 . Best linear unbiased estimation and prediction under a selection model . Biometrics 31 : 423 – 447 . doi:10.2307/2529430. Google Scholar Crossref Search ADS PubMed WorldCat Henderson , C. R. 1976 . A simple method for computing the inverse of a numerator relationship matrix used in prediction of breeding values . Biometrics . 32 : 69 – 83 . doi:10.2307/2529339. Google Scholar Crossref Search ADS WorldCat Hidalgo , J. , D. Lourenco, S. Tsuruta, Y. Masuda, S. Miller, M. Bermann, A. L. Garcia, and I. Misztal. 2021 . Changes in genomic predictions when new information is added . J. Anim. Sci . 99 : skab004 . doi:10.1093/jas/skab004. Google Scholar Crossref Search ADS PubMed WorldCat Junqueira , V. S. , F. F. Cardoso, M. M. Oliveira, B. P. Sollero, F. F. Silva, and P. S. Lopes. 2017 . Use of molecular markers to improve relationship information in the genetic evaluation of beef cattle tick resistance under pedigree-based models . J. Anim. Breed. Genet . 134 : 14 – 26 . doi:10.1111/jbg.12239. Google Scholar Crossref Search ADS PubMed WorldCat Junqueira , V. S. , P. S. Lopes, D. Lourenco, F. F. e Silva, and F. F. Cardoso. 2020 . Applying the metafounders approach for genomic evaluation in a multibreed beef cattle population . Front. Genet . 11 . doi:10.3389/fgene.2020.556399. Google Scholar OpenURL Placeholder Text WorldCat Kennedy , B. 1981 . Variance component estimation and prediction of breeding values . Can. J. Genet. Cytol . 23 : 565 – 578 . doi:10.1139/g81-062. Google Scholar Crossref Search ADS WorldCat Lidauer , M. , I. Strandén, E. A. Mäntysaari, J. Pösö, and A. Kettunen. 1999 . Solving large test-day models by iteration on data and preconditioned conjugate gradient . J. Dairy Sci . 82 : 2788 – 2796 . doi:10.3168/jds.S0022-0302(99)75536-0. Google Scholar Crossref Search ADS PubMed WorldCat Lourenco , D. A. , B. O. Fragomeni, S. Tsuruta, I. Aguilar, B. Zumbach, R. J. Hawken, A. Legarra, and I. Misztal. 2015 . Accuracy of estimated breeding values with genomic information on males, females, or both: an example on broiler chicken . Genet. Sel. Evol . 47 : 56 . doi:10.1186/s12711-015-0137-1. Google Scholar Crossref Search ADS PubMed WorldCat Lourenco , D. A. L. , A. Legarra, S. Tsuruta, D. Moser, S. Miller, and I. Misztal. 2018 . Tuning indirect predictions based on SNP effects from single-step GBLUP . Interbull Bulletin . 53 : 48 – 53 . Google Scholar OpenURL Placeholder Text WorldCat Masuda , Y. , I. Aguilar, S. Tsuruta, and I. Misztal. 2015 . Technical note: acceleration of sparse operations for average-information REML analyses with supernodal methods and sparse-storage refinements . J. Anim. Sci . 93 : 4670 – 4674 . doi:10.2527/jas.2015-9395. Google Scholar Crossref Search ADS PubMed WorldCat Masuda , Y. , T. Baba, and M. Suzuki. 2014 . Application of supernodal sparse factorization and inversion to the estimation of (co) variance components by residual maximum likelihood . J. Anim. Breed. Genet . 131 : 227 – 236 . doi:10.1111/jbg.12058. Google Scholar Crossref Search ADS PubMed WorldCat Masuda , Y. , I. Misztal, A. Legarra, S. Tsuruta, D. A. L. Lourenco, B. O. Fragomeni, and I. Aguilar. 2017 . Avoiding the direct inversion of the numerator relationship matrix for genotyped animals in single-step genomic best linear unbiased prediction solved with the preconditioned conjugate gradient . J. Anim. Sci . 95 : 49 – 52 . doi:10.2527/jas.2016.0699. Google Scholar PubMed OpenURL Placeholder Text WorldCat Masuda , Y. , I. Misztal, S. Tsuruta, A. Legarra, I. Aguilar, D. A. L. Lourenco, B. O. Fragomeni, and T. J. Lawlor. 2016 . Implementation of genomic recursions in single-step genomic best linear unbiased predictor for US holsteins with a large number of genotyped animals . J. Dairy Sci . 99 : 1968 – 1974 . doi:10.3168/jds.2015-10540. Google Scholar Crossref Search ADS PubMed WorldCat Meyer , K. 1997 . An average information restricted maximum likelihood algorithm for estimating reduced rank genetic covariance matrices or covariance functions for animal models with equal design matrices . Genet. Sel. Evol . 29 : 97 . Google Scholar Crossref Search ADS WorldCat Misztal , I. 2016 . Inexpensive computation of the inverse of the genomic relationship matrix in populations with small effective population size . Genetics . 202 : 401 – 409 . doi:10.1534/genetics.115.182089. Google Scholar Crossref Search ADS PubMed WorldCat Misztal , I. , H. L. Bradford, D. A. L. Lourenco, S. Tsuruta, Y. Masuda, A. Legarra, and T. J. Lawlor. 2017 . Studies on inflation of GEBV in single-step GBLUP for type . Interbull Bulletin . 51 : 38 – 42 . Google Scholar OpenURL Placeholder Text WorldCat Misztal , I. , A. Legarra, and I. Aguilar. 2014 . Using recursion to compute the inverse of the genomic relationship matrix . J. Dairy Sci . 97 : 3943 – 3952 . doi:10.3168/jds.2013-7752. Google Scholar Crossref Search ADS PubMed WorldCat Misztal , I. , S. Tsuruta, I. Pocrnic, and D. Lourenco. 2020 . Core-dependent changes in genomic predictions using the algorithm for proven and young in single-step genomic best linear unbiased prediction . J. Anim. Sci . 98 . doi: 10.1093/jas/skaa374. Google Scholar OpenURL Placeholder Text WorldCat Misztal , I. , S. Tsuruta, T. Strabel, B. Auvray, T. Druet, and D. H. Lee. 2002 . BLUPF90 and related programs. In: Proceedings of the 7th World Congress on Genetics Applied to Livestock Production. Patry , C. , and V. Ducrocq. 2011 . Evidence of biases in genetic evaluations due to genomic preselection in dairy cattle . J. Dairy Sci . 94 : 1011 – 1020 . doi:10.3168/jds.2010-3804. Google Scholar Crossref Search ADS PubMed WorldCat Patterson , H. D. , and R. Thompson. 1971 . Recovery of inter-block information when block sizes are unequal . Biometrika . 58 : 545 – 554 . doi:10.1093/biomet/58.3.545. Google Scholar Crossref Search ADS WorldCat Pocrnic , I. , D. A. Lourenco, Y. Masuda, A. Legarra, and I. Misztal. 2016a . The dimensionality of genomic information and its effect on genomic prediction . Genetics . 203 : 573 – 581 . doi:10.1534/genetics.116.187013. Google Scholar Crossref Search ADS WorldCat Pocrnic , I. , D. A. L. Lourenco, Y. Masuda, A. Legarra, and I. Misztal. 2016b . The dimensionality of genomic information and its effect on genomic prediction . Genetics . 203 : 573 – 581 . Google Scholar Crossref Search ADS WorldCat Pocrnic , I. , D. A. L. Lourenco, Y. Masuda, and I. Misztal. 2016c . Dimensionality of genomic information and performance of the algorithm for proven and Young for different livestock species . Genet. Sel. Evol . 48 : 82 . doi:10.1186/s12711-016-0261-6. Google Scholar Crossref Search ADS WorldCat Sargolzaei , M. , J. Chesnais, and F. Schenkel. 2011 . FImpute—an efficient imputation algorithm for dairy cattle populations . J. Dairy Sci . 94 : 421 . Google Scholar OpenURL Placeholder Text WorldCat Solberg , T. , A. Sonesson, J. Woolliams, and T. Meuwissen. 2008 . Genomic selection using different marker types and densities . J. Anim. Sci . 86 : 2447 – 2454 . doi:10.2527/jas.2007-0010. Google Scholar Crossref Search ADS PubMed WorldCat Stam , P. 1980 . The distribution of the fraction of the genome identical by descent in finite random mating populations . Genet. Res . 35 : 131 – 155 . Google Scholar Crossref Search ADS WorldCat Strandén , I. , and E. A. Mäntysaari. 2014 . Comparison of some equivalent equations to solve single-step GBLUP. In: Proceedings of the 10th World Congress on genetics applied to Livestock production. Vancouver . p. 22 . Tsuruta , S. , T. Lawlor, D. Lourenco, and I. Misztal. 2021 . Bias in genomic predictions by mating practices for linear type traits in a large-scale genomic evaluation . J. Dairy Sci . 104 : 662 – 677 . doi:10.3168/jds.2020-18668. Google Scholar Crossref Search ADS PubMed WorldCat Tsuruta , S. , I. Misztal, and I. Stranden. 2001 . Use of the preconditioned conjugate gradient algorithm as a generic solver for mixed-model equations in animal breeding applications . J. Anim. Sci . 79 : 1166 – 1172 . doi:10.2527/2001.7951166x. Google Scholar Crossref Search ADS PubMed WorldCat Vandenplas , J. , M. P. Calus, and J. Ten Napel. 2018 . Sparse single-step genomic BLUP in crossbreeding schemes . J. Anim. Sci . 96 : 2060 – 2073 . doi:10.1093/jas/sky136. Google Scholar Crossref Search ADS PubMed WorldCat VanRaden , P. M. 2008 . Efficient methods to compute genomic predictions . J. Dairy Sci . 91 : 4414 – 4423 . doi:10.3168/jds.2007-0980. Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2022. Published by Oxford University Press on behalf of the American Society of Animal Science. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected] © The Author(s) 2022. Published by Oxford University Press on behalf of the American Society of Animal Science.

doi: 10.1093/jas/skac103pmid: 35511235

Graphical Abstract

Cull, Charley A; Rezac, Darrel J; DeDonder, Keith D; Seagren, Jon E; Cull, Brooke J; Singu, Vijay K; Theurer, Miles E; Martin, Miriam; Amachawadi, Raghavendra G; Kleinhenz, Michael D; Lechtenberg, Kelly F

doi: 10.1093/jas/skac049pmid: 35176757

Abstract The objective of this study was to determine the effects of flunixin meglumine or meloxicam on behavioral response and performance characteristics associated with surgical castration in crossbred bulls. Intact male Bos taurus calves (n = 252; averaging 176 kg) were randomly allocated into one of three treatment groups within pen: control (CON), flunixin meglumine (FLU; 2.2 mg/kg intravenous injection), or meloxicam (MEL; 2.0 mg/kg per os). The individual animal was the experimental unit. Calves were individually weighed on days 0 and 14 of the trial to evaluate performance outcomes. On study day 0, treatments were administered, according to their random allocation, immediately prior to surgical castration using the Henderson tool method. Visual analog scale (VAS) assessment and categorical attitude score (CAS) were collected on days −1, 0 (6 h post-castration), 1, 2, 3, and 4 in the study. The VAS was assigned using a 100 mm horizontal line with “normal” labeled at one end of the line and “moribund” at the other end of the horizontal line. The masked observer assigned a mark on the horizontal line based upon the observed severity of pain exhibited by that individual animal. The CAS was assigned by the same observer using five different categories with a score of 0 being “normal”. Average daily gain tended (P = 0.09) to be associated with the treatment group, and MEL had a greater (P = 0.04) average daily gain through day 14 compared with CON. A significant (P < 0.01) treatment by day interaction was indicated for VAS score, and MEL had lower VAS scores on days 0, 1, 2, and 3 post-castration compared with CON; FLU had lower VAS scores on days 0 and 1 compared with CON. A significant treatment by day interaction was not present (P = 0.25) for CAS. The FLU had lesser percent CAS ≥1 (17.5%; P = 0.05) compared with CON (29.4%); MEL has lesser percent CAS ≥1 observations (14.9%; P = 0.01) compared with CON. The median VAS increased as CAS was more severe. Results indicated MEL and FLU calves temporally improved behavioral responses following surgical castration with positive numerical trends for a 14 d average daily gain (ADG). The VAS system appeared to be an effective method of subjective evaluation of pain in beef calves in this study. Route of administration, duration of therapy, and low relative cost make oral meloxicam a reasonable analgesic treatment in calves when administered at the time of surgical castration. Introduction Castration has been recognized as one of the most common surgical procedures performed on bull calves, with an estimated 16 million calves castrated every year in the United States (USDA, 2016). Calves are castrated in the beef industry to reduce aggression and behavioral responses, improve carcass quality, and prevent unwanted pregnancy (Stafford and Mellor, 2005; Coetzee, 2013). Castration of bull calves is considered a painful procedure for beef calves but routinely performed without analgesia (Coetzee et al., 2010; Fajt et al., 2011). As emphasis increases for the administration of an analgesic therapy with castration procedures, extra-label therapy is needed due to the lack of an approved food animal analgesic compound labeled for castration in the United States (U.S. Food and Drug Administration, 2006). Previous literature has evaluated the effects of two commercially available nonsteroidal anti-inflammatory drugs (NSAID), flunixin meglumine and meloxicam, prior to castration in calves (Coetzee et al., 2012; Stock and Coetzee, 2015). Flunixin meglumine is an NSAID approved for intravenous route of administration and the control of pyrexia associated with bovine respiratory disease and inflammation associated with endotoxemia in food animals in the United States (Smith et al., 2008; Smith, 2013). A transdermal flunixin formulation has recently been approved by the United States Food Drug Administration (US FDA) for the control of pain associated with foot rot in cattle (FDA, 2017). Meloxicam is a relatively selective NSAID, preferentially inhibiting the COX-2 isoenzyme, approved to alleviate pain and inflammation following surgical and band castration in cattle in Canada (Engelhardt et al., 1996; https://solvet.ca/wp-content/uploads/2019/11/SOL-Meloxicam-Oral-Suspension-USP-CVP.pdf). However, no approved food animal analgesic compounds are indicated for castration-related pain in the United States. The onset of therapeutic activity is similar between oral and subcutaneous routes of administration for meloxicam but indicates a higher bioavailability with the oral route of administration (Coetzee et al., 2009). Although the literature indicates a potential behavioral or performance benefit of these products, a need exists for a more externally valid model of evaluation in a commercial beef production setting. Therefore, the objective of this study was to determine the effects of flunixin meglumine or meloxicam in comparison to the negative control, when administered at the time of surgical castration, on the behavioral characteristics and performance in crossbred bull calves. Materials and Methods All activities related to this study were reviewed and approved by the Institutional Animal Care and Use Committee of the Veterinary and Biomedical Research Center, Inc. prior to study initiation (IACUC number VAC15024B). Study population and animal management A total of 252, intact male crossbred Bos taurus calves (mean ± SE BW = 176 ± 4.7 kg) were received into a commercial backgrounding yard in Kansas. Each animal was identified by an individual ear tag. Individual cattle were eligible for inclusion if they had a categorical attitude score (CAS; Table 1) of 0, presence of two descended testicles, and a minimum of a 30 d acclimation period. Following a brief transition period, cattle were fed once daily, a grower diet which included (as-fed basis): 52.7% wet distillers’ grain, 37.6% roughage, 8.4% cracked corn, and 1.3% micro/minerals mix (including 200 mg of monensin per animal per day [Rumensin, Elanco Animal Health, Greenfield, IN]). Standard operating procedures at the backgrounding yard were followed for cattle management and care. Table 1. Description of CAS to classify pain status of calves castrated using the Henderson tool technique and administered injectable flunixin meglumine, oral meloxicam or a sham control at the time of castration Categorical attitude score (CAS) . Description . Clinical appearance . 0 Clinically normal Stands and walks normally, no aversion to movement, appears bright and alert 1 Mild depression Appears slightly depressed, but responds quickly to handler when prompted 2 Moderate depression Head lowered, ears drooped, animal moves away slowly from handler when prompted 3 Severe depression Head lowered, ears drooped, animal is reluctant to move away from handler when prompted, appears to have low abdominal fill 4 Moribund Animal will not rise or move without kinetic pressure from handler. Animal is a candidate for humane euthanasia upon consultation with attending veterinarian Categorical attitude score (CAS) . Description . Clinical appearance . 0 Clinically normal Stands and walks normally, no aversion to movement, appears bright and alert 1 Mild depression Appears slightly depressed, but responds quickly to handler when prompted 2 Moderate depression Head lowered, ears drooped, animal moves away slowly from handler when prompted 3 Severe depression Head lowered, ears drooped, animal is reluctant to move away from handler when prompted, appears to have low abdominal fill 4 Moribund Animal will not rise or move without kinetic pressure from handler. Animal is a candidate for humane euthanasia upon consultation with attending veterinarian Open in new tab Table 1. Description of CAS to classify pain status of calves castrated using the Henderson tool technique and administered injectable flunixin meglumine, oral meloxicam or a sham control at the time of castration Categorical attitude score (CAS) . Description . Clinical appearance . 0 Clinically normal Stands and walks normally, no aversion to movement, appears bright and alert 1 Mild depression Appears slightly depressed, but responds quickly to handler when prompted 2 Moderate depression Head lowered, ears drooped, animal moves away slowly from handler when prompted 3 Severe depression Head lowered, ears drooped, animal is reluctant to move away from handler when prompted, appears to have low abdominal fill 4 Moribund Animal will not rise or move without kinetic pressure from handler. Animal is a candidate for humane euthanasia upon consultation with attending veterinarian Categorical attitude score (CAS) . Description . Clinical appearance . 0 Clinically normal Stands and walks normally, no aversion to movement, appears bright and alert 1 Mild depression Appears slightly depressed, but responds quickly to handler when prompted 2 Moderate depression Head lowered, ears drooped, animal moves away slowly from handler when prompted 3 Severe depression Head lowered, ears drooped, animal is reluctant to move away from handler when prompted, appears to have low abdominal fill 4 Moribund Animal will not rise or move without kinetic pressure from handler. Animal is a candidate for humane euthanasia upon consultation with attending veterinarian Open in new tab Treatment allocation and administration Individual animals were randomly allocated using computer software (Excel, Microsoft Corp., Redmond, WA) to one of three treatment groups within pen (n = 5 pens): control (CON), flunixin meglumine (FLU; 2.2 mg/kg BW via jugular intravenous injection), or meloxicam (MEL; 2.0 mg/kg BW per os). Treatments were administered immediately prior to castration on study day 0. CON was administered 0.9% sodium chloride (0.044 mL/kg BW via jugular intravenous injection) and whey protein powder (approximately 14 g, per os). The whey protein powder dose amount was selected as this was proximate to the average weight of the MEL test article (meloxicam) as supplied by 7.5 mg meloxicam tablets (Zydus Pharmaceuticals USA, Inc., Pennington, NJ) within a porcine gelatin capsule. Both meloxicam and whey powder were delivered to the respective treatment groups via a single, 24 mL porcine gelatin capsule (Torpac, Inc., Fairfield, NJ) using a commercially available stainless steel balling gun. Meloxicam doses were rounded to the next (higher) nearest 7.5 mg (whole tablet) dose. The FLU group also received whey protein (approximately 14 g, per os) and MEL, 0.9% sodium chloride (0.044 mL/kg BW) via jugular intravenous injection to ensure each animal received similar procedural manipulations, regardless of treatment group. All intravenous injections were administered through the jugular vein using an 18 gauge × 40 mm hypodermic needle. Castration procedure Castration was completed on study day 0, immediately after (approximately 30 s) the treatment administration while the animals were restrained in a hydraulic chute (Silencer Chutes, Moly Manufacturing, Inc., Lorraine, KS). Briefly, all foreign material was removed from the scrotum by a gloved hand and the scrotum was thoroughly scrubbed with a dilute chlorhexidine solution. Both testicles were isolated near the body wall, and the scrotum was incised using a Newberry knife (Jorgensen Lab, Loveland, CO). Approximately 50% of the length of the scrotum was incised perpendicular to the scrotal septum to expose both testicles with a single incision, and allow for adequate drainage during healing. The cremaster muscle of each testicle was broken down by blunt dissection, and the Henderson castration tool (Stone Manufacturing and Supply Company, Kansas City, MO) connected to a cordless drill was clamped to the spermatic cord proximal to the head of the epididymis. The electric drill was engaged in a clockwise direction to twist the spermatic cord until severed by spiral torsion as previously described (Coetzee et al., 2007; Webster et al., 2013). The same procedure was used to remove the second testicle. After removal of both testicles, the scrotum was disinfected with 1% iodine wound spray. The Henderson castration tool was disinfected with dilute chlorhexidine solution between each animal. Outcome measures and blinding A single observer, blinded to treatment group allocation, subjectively evaluated the bulls with each treatment equally represented on study days −1, 0 (6 h post-castration), 1, 2, 3, and 4 of the study. The observer was a trained DVM with vast experience in visual analog scale (VAS) scoring in cattle models. Observations were recorded at 08:00 a.m. daily with the observer standing in the pen. All animals were observed at each timepoint and scoring was completed while in the pen. The observer used the VAS and CAS as assessment methods for each animal evaluated at each time point. The VAS was assigned using a 100 mm horizontal line with “normal” labeled on the left end of the line and “moribund” at the right end of the horizontal line. A vertical mark was placed on the VAS line at the point, which in the blinded evaluator’s assessment, reflected the pain status of the animal between normal and moribund. The distance the observer marked on the horizontal line was measured with the use of a digital caliper (Fisher Scientific, Hampton, NH) to derive the raw data point for inclusion into the dataset. The CAS was assigned using a five-point scoring system defined in Table 1 just after the VAS was determined. Cattle performance measures were considered secondary outcomes and were collected by study personnel also blinded to the treatment group. Furthermore, personnel who administered treatments did not collect VAS, CAS, health, or performance data. Cattles were individually weighed on days 0 and 14 of the study. For purposes of this study, ADG was calculated as: ADG=(Day14Day 0)/14 days Statistical analysis Data were entered into a commercial software package (R Core Team 2015, Vienna, Australia). The VAS was evaluated as the proportion of scale an individual calf was scored at each observation period (Myles et al., 1999). A binary outcome variable was created for each animal for CAS. Calves receiving a CAS ≥ 1 were assigned a value of 1, and calves that received a CAS equal to 1 were assigned a value of 0. Generalized linear mixed models were used for the VAS and CAS outcomes and included treatment group, study day, and treatment by study day interaction. Random effects for repeated measures on individual calves within a pen were included for the VAS and CAS outcomes. Due to study design, the biological plausibility of castration pain, and outcome of interest, if treatment by study day interaction was not statistically significant (P > 0.10), day was included as a random effect in the final statistical model to evaluate the overall treatment effects. Effects with a P-value ≤0.10 were further explored. Potential differences within an individual study day between treatment groups were evaluated with pairwise comparisons. A P-value ≤0.05 was considered statistically significant for all pairwise comparisons. Box and whisker plots with median, first and third quartiles, minimum, and maximum VAS by CAS were evaluated. Continuous outcomes of initial BW and ADG were evaluated with individual mixed linear models. Linear models included a random effect for pen in analysis and treatment as a fixed effect. Results There was a (P < 0.01) treatment by study day interaction for VAS (Figure 1). Calves in the MEL group had decreased VAS scores on days 0 (P = 0.07), 1 (P < 0.01), 2 (P = 0.05), and 3 (P = 0.01) compared with CON calves. Calves in the FLU group had decreased VAS scores on days 0 (P < 0.01) and 1 (P < 0.01) compared with CON calves. No other comparisons among treatment groups within a day were statistically significant for VAS. Figure 1. Open in new tabDownload slide Model-adjusted least squares means (± SE) of VAS scores by treatment group and study day of calves castrated using the Henderson tool technique and concurrently administered injectable flunixin meglumine (FLU; 2.2 mg/kg BW via jugular intravenous injection), oral meloxicam (MEL; 2.0 mg/kg BW per os), or sham control products (CON) at the time of castration. The model included effects for repeated measures on individual calves within pen. Calves in the MEL group had decreased VAS scores on days 0 (P = 0.07), 1 (P < 0.01), 2 (P = 0.05), and 3 (P = 0.01) compared with CON calves. Calves in the FLU group had decreased VAS scores on days 0 (P < 0.01) and 1 (P < 0.01) compared with CON calves. No other comparisons among treatment groups, within a day were statistically significant for VAS. There was not a treatment by study day interaction for CAS. Study day remained in the final model as a random effect due to study design. Calves in the MEL group had decreased percentage of CAS ≥ 1 (P = 0.01) compared with CON calves (Figure 2). Calves in the FLU group had decreased percentage of CAS ≥ 1 (P = 0.05) compared with CON calves (Figure 2). No differences were identified in the percentage of CAS ≥ 1 in MEL group compared with FLU group (P = 0.58). Box and whisker plots of VAS by CAS were displayed in Figure 3. Figure 2. Open in new tabDownload slide Model-adjusted least squares mean (±SE) percentage of CAS ≥1 observations by treatment group of calves castrated using the Henderson tool technique and concurrently administered injectable flunixin meglumine (FLU; 2.2 mg/kg BW via jugular intravenous injection), oral meloxicam (MEL; 2.0 mg/kg BW per os), or sham control products (CON) at the time of castration. The model included effects for repeated measures on individual calves within pen. The model included random effects for repeated measures on individual calves within pen and study day. Treatment groups not connected by the same letter are significantly (P < 0.05) different. Figure 3. Open in new tabDownload slide Box and whisker plots with median, first and third quartiles, minimum, and maximum VAS by CAS that were surgically castrated using the Henderson tool technique. Average daily gain during the first 14 d after surgical castration tended (P = 0.09) to be associated with the treatment group. Average daily gain (± SE) of calves in the CON, FLU, and MEL treatment groups were 1.20 (± 0.31), 1.23 (± 0.31), and 1.40 (± 0.31) kg/day, respectively. Calves in the MEL group had greater ADG compared with CON calves from day 0 to day 14 (P = 0.04); MEL calves also tended (P = 0.09) to have greater ADG from day 0 to day 14 compared with FLU calves. No differences (P = 0.72) were identified in ADG for CON compared with FLU calves. Discussion The objective of this study was to evaluate the use of two of the most commonly prescribed NSAID products, utilizing an experimental model that closely resembled those practices which are commonly employed at commercial cattle feeding facilities. Based on these data, it seems clear that pain-associated behavior traits were diminished by both MEL and FLU; however, differences were also elucidated between their therapeutic duration, which is consistent with current data relating to the pharmacokinetic profile of oral meloxicam and intravenous flunixin meglumine (Coetzee, 2013; Fraccaro et al., 2013). The American Veterinary Medical Association (AVMA) policy on castration and dehorning states that because these procedures cause pain and discomfort, practices such as the use of the Animal Medical Drug Use and Clarification Act (AMDUCA)-permissible clinically effective medications are recommended when possible (AVMA, 2019). Injectable flunixin meglumine is, at current, only labeled for intravenous administration in cattle, though anecdotal reports suggest that it is widely known within the veterinary community, however, that injectable flunixin meglumine is often administered via non-approved routes (intramuscularly or subcutaneously) by personnel who are not proficient at intravenous jugular injection techniques. It is worth noting that altering the route of administration is not a justifiable basis for ELDU, and is considered illegal by the Animal Medical Drug Use Clarification Act (AMDUCA) and may cause local tissue reactions (U.S. Food and Drug Administration, 2014). This risk factor, along with the overall time required to intravenously administer the product, demonstrates a significant drawback to the use of injectable flunixin meglumine as an analgesic at the time of castration. Currently, meloxicam is not labeled for use in cattle, for any indication or by any route of administration in the United States, and, as previously stated at this time, there are no products labeled for analgesia in cattle at the time of castration. Taken as a whole, these factors justify the extra-label use of meloxicam under AMDUCA. One clear advantage to meloxicam over that of an injectable NSAID is the ability for the therapeutic to be administered per os (PO) as a single bolus, which requires much less training and skill to become proficient at as compared with intravenous administration techniques, while also decreasing the time required for administration, thus potentially also decreasing the amount of time the animal is restrained. Furthermore, administering therapy PO decreases, the number of injections the animal is subjected to, which is in compliance with initiatives such as the Beef Quality Assurance program (Beef Quality Assurance, 2014). A final key consideration must be the cost/benefit model of providing therapeutic analgesia. The cost of pain mitigation is a common factor identified by producers, which has affected the widespread adoption into the industry (Newton and O’Connor, 2013). A survey of practicing veterinarians in North America identified only 21% of respondents administered a systemic analgesic compound at the time of castration with flunixin meglumine being the most common product used. At the time of this study, the cost of meloxicam was $0.40/100 kg of BW and the cost of flunixin meglumine was $0.93/100 kg of BW. Some of the previous publications utilizing meloxicam have described a protocol where the treatment was administered approximately 24 h prior to the painful procedure (e.g., dehorning and castration; Coetzee et al., 2012), under that hypothesis this lead time was required to allow for the maximum therapeutic effect at the time of the painful procedure (the time to maximum plasma concentration is 12 to 24 h for meloxicam; Coetzee et al., 2009; Fraccaro et al., 2013). However, administering meloxicam to calves 24 h prior is wrought with many inefficiencies and is not likely an externally valid model. An additional day processing calves through a chute requires extra time, labor, is an additional stressor for the cattle, and has been identified as a potential reason for decreased animal performance (Voisinet et al., 1997; Cull et al., 2012, 2015; Francisco et al., 2012). In the current study, both the MEL and FLU treatment groups were administered immediately prior to surgical castration which represents a scenario more likely to be adopted by producers and recommended by veterinarians. The modest ADG improvement observed in the current study for MEL and FLU vs. CON was unexpected. Previous research has indicated calves dehorned and administered meloxicam spent more time near the feeder compared with placebo control calves which may be a potential reason for improved ADG in the current study (Theurer et al., 2012). Healthy animals also spent more time near the feeder compared with morbid animals (Sowell et al., 1999; Buhman et al., 2000; Theurer et al., 2013b; Jackson et al., 2016). A meta-analysis, which investigated the association between pain management and increased production outcomes, concluded that the body of work does not support the hypothesis of analgesics/pain interventions directly improving performance parameters (Newton and O’Connor, 2013). Newton and O’Connor (2013) also point out that the vast majority of the published data in this field are associated with short study periods and a relatively low number of experimental units. The study herein was short in duration, mostly attributable to the stage of production, but had a sample size much larger than many previous publications. The production benefits observed in this study are modest but do suggest that this topic warrants further investigation. Given the differences in the pharmacokinetics of the two drugs, an apparent longer duration of therapeutic effect in MEL compared with FLU was not surprising. The plasma half-life of meloxicam administered PO in calves has been shown to vary from 16 to 27 h (Wagner et al. 2021). In contrast, the reported half-life of flunixin meglumine administered intravenously ranges from 3 to 8 h in plasma (Anderson et al., 1990; Landoni et al., 1995; Coetzee et al., 2009; Fraccaro et al., 2013; Glynn et al., 2013). The longer plasma half-life of meloxicam compared with flunixin meglumine provides meloxicam a larger area under the curve which should reasonably translate to a longer period for potential analgesic therapeutic effect. Furthermore, the 2 mg/kg dose used in this study is double that previously published. It would be expected that the duration of effect for meloxicam would be an additional half-life and thus an additional 16 to 27 h. Further work is needed to better explain the effect a double dose may have on prostaglandin production. Flunixin meglumine administered intravenously is deposited directly into the systemic circulation where it can potentially provide analgesia almost instantaneously, whereas an orally administered therapy, such as the meloxicam used in this study, must be absorbed through the gastrointestinal tract before it can enter the systemic circulation to provide analgesia. Thus, resulting in a lag time between the drug administration and its clinical therapeutic effect, therefore, it was not surprising that FLU anecdotally appeared to elicit a more immediate effect on pain behaviors as compared with MEL. However, it is important to note that there were no statistical differences observed between MEL and FLU. Historically, scoring systems used for clinical evaluation have been based on discrete categorical outcomes, arranged in a loose ordinal system. Meaning that observers classify an animal into an individual category (or score) based upon the appearance and clinical signs of the animal. Results are tabulated but commonly are evaluated as binary outcomes to determine the probability of being classified as normal or abnormal (Theurer et al., 2013a). These types of scoring systems are much more qualitative in nature than they are quantitative, as the difference between each descriptive category is not necessarily the same. Additionally, the differences between each category can be interpreted to be of a different magnitude by different clinical scorers. In contrast, the VAS system provides a more sensitive scoring system to the observer. This continuous scale has the potential to more accurately identify and quantify subtle differences between treatments groups which may not be able to be captured with traditional categorical scoring systems (Welsh et al., 1993). The continuous VAS has been used to accurately and reliably identify dairy cows with sole ulcers in dairy cows (Flower and Weary, 2006). A VAS was utilized by two scorers to assess pain in cattle (steers and bulls) following processing at a feedyard (Martin et al., 2020). The authors reported a lower VAS for calves treated with transdermal flunixin. Furthermore, research in humans has identified that the VAS has the ability to accurately represent the magnitude of pain experienced in the human medical field (Myles et al., 1999). In the current study, the similarities between the VAS and CAS ≥1 outcomes as well as the distribution of the VAS score by CAS demonstrate that the VAS system was an effective scoring tool in this group of calves. Based on these results, it seems prudent that the VAS system warrants further research as a subjective measure of pain exhibited by beef calves. However, it is important to note that while the observer assigning CAS and VAS scores was blinded to the treatment group, the same observer assigned the CAS and VAS scores to each calf individually at the same time. Therefore, it must be noted that the potential for observation bias to occur was present since the observer did not assign CAS and VAS independently, but rather did so simultaneously. Conclusions Data indicated that MEL and FLU both improved behavior response in calves after surgical castration. Even while considering the different routes of administration, both products improved behavior response when administered concurrently to the painful stimuli. The VAS system appears to be an effective method of pain-associated behavioral assessment in this class of animals. Average daily gains were improved when meloxicam was provided. Method of administration, duration of effect, and the cost of treatment make meloxicam an attractive analgesic in calves when administered at the time of surgical castration. Abbreviations Abbreviations AMDUCA Animal Medical Drug Use and Clarification Act AVMA American Veterinary Medical Association CAS categorical attitude score CON control FLU flunixin meglumine IACUC Institutional Animal Care and Use Committee MEL meloxicam NSAID nonsteroidal anti-inflammatory drugs PO per os US FDA United States Food Drug Administration VAS visual analog scale Acknowledgment This study was internally funded by Midwest Veterinary Services, Inc., as there was no external funding or support for this project. Conflict of Interest Statement The authors declare no real or perceived conflicts of interest. Literature Cited Anderson , K. L. , C. A. Neff-Davis, L. E. Davis, and V. D. Bass. 1990 . Pharmacokinetics of flunixin meglumine in lactating cattle after single and multiple intramuscular and intravenous administrations. Am. J. Vet. Res . 51 ( 9 ): 1464 – 1467 . PMID: 2396794. Google Scholar PubMed OpenURL Placeholder Text WorldCat AVMA. 2019 . Castration and dehorning of cattle. American Veterinary Medical Association. [accessed February 16, 2019 ] https://www.avma.org/KB/Policies/Pages/Castration-and-Dehorning-of-Cattle.aspx. Beef Quality Assurance. 2014 . National beef quality assurance manual. https://www.bqa.org/Media/BQA/Docs/bqa_manual_final.pdf. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Buhman , M. J. , L. J. Perino, M. L. Galyean, T. E. Wittum, T. H. Montgomery, and R. S. Swingle. 2000 . Association between changes in eating and drinking behaviors and respiratory tract disease in newly arrived calves at a feedlot. Am. J. Vet. Res . 61 ( 10 ): 1163 – 1168 . doi:10.2460/ajvr.2000.61.1163 Google Scholar Crossref Search ADS PubMed WorldCat Coetzee , J. F. 2013 . Assessment and management of pain associated with castration in cattle. Vet. Clin. North Am. Food Anim. Pract . 29 ( 1 ): 75 – 101 . doi:10.1016/j.cvfa.2012.11.002 Google Scholar Crossref Search ADS PubMed WorldCat Coetzee , J. F. , L. N. Edwards, R. A. Mosher, N. M. Bello, A. M. O’Connor, B. Wang, B. Kukanich, and D. A. Blasi. 2012 . Effect of oral meloxicam on health and performance of beef steers relative to bulls castrated on arrival at the feedlot. J. Anim. Sci . 90 ( 3 ): 1026 – 1039 . doi:10.2527/jas.2011-4068 Google Scholar Crossref Search ADS PubMed WorldCat Coetzee , J. F. , R. Gehring, A. C. Bettenhausen, B. V. Lubbers, S. E. Toerber, D. U. Thomson, B. Kukanich, and M. D. Apley. 2007 . Attenuation of acute plasma cortisol response in calves following intravenous sodium salicylate administration prior to castration. J. Vet. Pharmacol. Ther . 30 ( 4 ): 305 – 313 . doi:10.1111/j.1365-2885.2007.00869.x Google Scholar Crossref Search ADS PubMed WorldCat Coetzee , J. F. , B. KuKanich, R. Mosher, and P. S. Allen. 2009 . Pharmacokinetics of intravenous and oral meloxicam in ruminant calves. Vet. Ther . 10 ( 4 ): E1 – E8 . PMID: 20425727. Google Scholar PubMed OpenURL Placeholder Text WorldCat Coetzee , J. F. , A. L. Nutsch, L. A. Barbur, and R. M. Bradburn. 2010 . A survey of castration methods and associated livestock management practices performed by bovine veterinarians in the United States. Bmc Vet. Res . 6 : 12 . doi:10.1186/1746-6148-6-12 Google Scholar Crossref Search ADS PubMed WorldCat Cull , C. A. , Z. D. Paddock, T. G. Nagaraja, N. M. Bello, A. H. Babcock, and D. G. Renter. 2012 . Efficacy of a vaccine and a direct-fed microbial against fecal shedding of Escherichia coli O157:H7 in a randomized pen-level field trial of commercial feedlot cattle. Vaccine 30 ( 43 ): 6210 – 6215 . doi:10.1016/j.vaccine.2012.05.080 Google Scholar Crossref Search ADS PubMed WorldCat Cull , C. A. , D. G. Renter, N. M. Bello, S. E. Ives, and A. H. Babcock. 2015 . Performance and carcass characteristics of commercial feedlot cattle from a study of vaccine and direct-fed microbial effects on Escherichia coli O157:H7 fecal shedding. J. Anim. Sci . 93 ( 6 ): 3144 – 3151 . doi:10.2527/jas.2015-8924 Google Scholar Crossref Search ADS PubMed WorldCat Engelhardt , G. , R. Bögel, C. Schnitzer, and R. Utzmann. 1996 . Meloxicam: influence on arachidonic acid metabolism. Part 1. In vitro findings. Biochem. Pharmacol . 51 ( 1 ): 21 – 28 . doi:10.1016/0006-2952(95)02111-6 Google Scholar Crossref Search ADS PubMed WorldCat Fajt , V. R. , S. A. Wagner, and B. Norby. 2011 . Analgesic drug administration and attitudes about analgesia in cattle among bovine practitioners in the United States. J. Am. Vet. Med. Assoc . 238 ( 6 ): 755 – 767 . doi:10.2460/javma.238.6.755 Google Scholar Crossref Search ADS PubMed WorldCat FDA . 2017 . Freedom of information summary original new animal drug application Banamine Transdermal. US FDA CVM . https://animaldrugsatfda.fda.gov/adafda/app/search/public/document/downloadFoi/1944. Flower , F. C. , and D. M. Weary. 2006 . Effect of hoof pathologies on subjective assessments of dairy cow gait. J. Dairy Sci . 89 ( 1 ): 139 – 146 . doi:10.3168/jds.S0022-0302(06)72077-X Google Scholar Crossref Search ADS PubMed WorldCat Fraccaro , E. , J. F. Coetzee, R. Odore, L. N. Edwards-Callaway, B. Kukanich, P. Badino, L. Bertolotti, H. Glynn, J. Dockweiler, K. Allen, and L. Bergamasco. 2013 . A study to compare circulating flunixin, meloxicam and gabapentin concentrations with prostaglandin E2 levels in calves undergoing dehorning. Res. Vet. Sci . 95 ( 1 ): 204 – 211 . doi:10.1016/j.rvsc.2013.01.018 Google Scholar Crossref Search ADS PubMed WorldCat Francisco , C. L. , R. F. Cooke, R. S. Marques, R. R. Mills, and D. W. Bohnert. 2012 . Effects of temperament and acclimation to handling on feedlot performance of Bos taurus feeder cattle originated from a rangeland-based cow-calf system. J. Anim. Sci . 90 ( 13 ): 5067 – 5077 . doi:10.2527/jas.2012-5447 Google Scholar Crossref Search ADS PubMed WorldCat Glynn , H. D. , J. F. Coetzee, L. N. Edwards-Callaway, J. C. Dockweiler, K. A. Allen, B. Lubbers, M. Jones, E. Fraccaro, L. L. Bergamasco, and B. KuKanich. 2013 . The pharmacokinetics and effects of meloxicam, gabapentin, and flunixin in postweaning dairy calves following dehorning with local anesthesia. J. Vet. Pharmacol. Ther . 36 : 550 – 561 . doi:10.1111/jvp.12042 Google Scholar Crossref Search ADS PubMed WorldCat Jackson , K. S. , G. E. Carstens, L. O. Tedeschi, and W. E. Pinchak. 2016 . Changes in feeding behavior patterns and dry matter intake before clinical symptoms associated with bovine respiratory disease in growing bulls. J. Anim. Sci . 94 ( 4 ): 1644 – 1652 . doi:10.2527/jas.2015-9993 Google Scholar Crossref Search ADS PubMed WorldCat Landoni , M. F. , F. M. Cunningham, and P. Lees. 1995 . Determination of pharmacokinetics and pharmacodynamics of flunixin in calves by use of pharmacokinetic/pharmacodynamic modeling. Am. J. Vet. Res . 56 ( 6 ): 786 – 794 . PMID: 7653889. Google Scholar PubMed OpenURL Placeholder Text WorldCat Martin , M. , M. D. Kleinhenz, K. E. Kleinhenz, E. J. Reppert, D. Blasi, T. Parks, A. Baysinger, J. Hutchenson, and J. F. Coetzee. 2020 . Comparison of the effect of tildipirosin administered alone or in combination with transdermal flunixin on the performance, health, activity and well-being of transported feedlot calves on arrival at the feedlot. Trans. Anim. Sci . doi:10.1093/tas/txaa005 Google Scholar OpenURL Placeholder Text WorldCat Myles , P. S. , S. Troedel, M. Boquest, and M. Reeves. 1999 . The pain visual analog scale: is it linear or nonlinear? Anesth. Analg . 89 ( 6 ): 1517 – 1520 . doi:10.1097/00000539-199912000-00038 Google Scholar Crossref Search ADS PubMed WorldCat Newton , H. P. , and A. M. O’Connor. 2013 . The economics of pain management. Vet. Clin. North Am. Food Anim. Pract . 29 ( 1 ): 229 – 250 . doi:10.1016/j.cvfa.2012.11.010 Google Scholar Crossref Search ADS PubMed WorldCat R Core Team . 2015 . R: A Language and Environment for Statistical Computing, Vienna, Austria . https://www.R-project.org/. Smith , G. 2013 . Extralabel use of anesthetic and analgesic compounds in cattle. Vet. Clin. North Am. Food Anim. Pract . 29 ( 1 ): 29 – 45 . doi:10.1016/j.cvfa.2012.11.003 Google Scholar Crossref Search ADS PubMed WorldCat Smith , G. W. , J. L. Davis, L. A. Tell, A. I. Webb, and J. E. Riviere. 2008 . Extralabel use of nonsteroidal anti-inflammatory drugs in cattle. J. Am. Vet. Med. Assoc . 232 ( 5 ): 697 – 701 . doi:10.2460/javma.232.5.697 Google Scholar Crossref Search ADS PubMed WorldCat Sowell , B. F. , M. E. Branine, J. G. Bowman, M. E. Hubbert, H. E. Sherwood, and W. Quimby. 1999 . Feeding and watering behavior of healthy and morbid steers in a commercial feedlot. J. Anim. Sci . 77 ( 5 ): 1105 – 1112 . doi:10.2527/1999.7751105x Google Scholar Crossref Search ADS PubMed WorldCat Stafford , K. J. , and D. J. Mellor. 2005 . The welfare significance of the castration of cattle: a review. N. Z. Vet. J . 53 : 271 – 278 . doi:10.1080/00480169.2005.36560 Google Scholar Crossref Search ADS PubMed WorldCat Stock , M. L. , and J. F. Coetzee. 2015 . Clinical pharmacology of analgesic drugs in cattle. Vet. Clin. North Am. Food Anim. Pract . 31 ( 1 ): 113 – 138 . doi:10.1016/j.cvfa.2014.11.002 Google Scholar Crossref Search ADS PubMed WorldCat Theurer , M. E. , D. E. Amrine, and B. J. White. 2013a . Remote noninvasive assessment of pain and health status in cattle. Vet. Clin. North Am. Food Anim. Pract . 29 ( 1 ): 59 – 74 . doi:10.1016/j.cvfa.2012.11.011 Google Scholar Crossref Search ADS WorldCat Theurer , M. E. , D. E. Anderson, B. J. White, M. D. Miesner, D. A. Mosier, J. F. Coetzee, J. Lakritz, and D. E. Amrine. 2013b . Effect of Mannheimia haemolytica pneumonia on behavior and physiologic responses of calves during high ambient environmental temperatures. J. Anim. Sci . 91 ( 8 ): 3917 – 3929 . doi:10.2527/jas.2012-5823 Google Scholar Crossref Search ADS WorldCat Theurer , M. E. , B. J. White, J. F. Coetzee, L. N. Edwards, R. A. Mosher, and C. A. Cull. 2012 . Assessment of behavioral changes associated with oral meloxicam administration at time of dehorning in calves using a remote triangulation device and accelerometers. Bmc Vet. Res . 8 : 48 . doi:10.1186/1746-6148-8-48 Google Scholar Crossref Search ADS PubMed WorldCat U.S. Food and Drug Administration. 2014 . Animal Medical Drug Use Clarification Act of 1994 (AMDUCA). [accessed May 4, 2016 ]. https://www.fda.gov/animal-veterinary/guidance-regulations/animal-medicinal-drug-use-clarification-act-1994-amduca. U.S. Food and Drug Administration, C. f. V. M. 2006 . Guidance for industry no. 123. Development of target animal safety and effectiveness data to support approval of non-steroidal anti-inflammatory drugs (NSAIDs) for use in animals. https://www.fda.gov/media/70191/download. United States Department of Agriculture, NASS. 2016 . Cattle. https://downloads.usda.library.cornell.edu/usda-esmis/files/h702q636h/1n79h639b/1z40kw116/Catt-01-29-2016.pdf. ISSN: 1948-9099. Voisinet , B. D. , T. Grandin, S. F. O’Connor, J. D. Tatum, and M. J. Deesing. 1997 . Bos indicus-cross feedlot cattle with excitable temperaments have tougher meat and a higher incidence of borderline dark cutters. Meat Sci . 46 ( 4 ): 367 – 377 . doi:10.1016/s0309-1740(97)00031-4 Google Scholar Crossref Search ADS PubMed WorldCat Wagner , B. K. , E. Nixon, I. Robles, R. E. Baynes, J. F. Coetzee, and M. D. Pairis-Garcia. 2021 . Non-steroidal anti-inflammatory drugs: pharmacokinetics and mitigation of procedural pain in cattle. Animals . 11 ( 2 ): 282 . doi:10.3390/ani11020282 Google Scholar Crossref Search ADS PubMed WorldCat Webster , H. B. , D. Morin, V. Jarrell, C. Shipley, L. Brown, A. Green, R. Wallace, and P. D. Constable. 2013 . Effects of local anesthesia and flunixin meglumine on the acute cortisol response, behavior, and performance of young dairy calves undergoing surgical castration. J. Dairy Sci . 96 ( 10 ): 6285 – 6300 . doi:10.3168/jds.2012-6238 Google Scholar Crossref Search ADS PubMed WorldCat Welsh , E. M. , G. Gettinby, and A. M. Nolan. 1993 . Comparison of a visual analogue scale and a numerical rating scale for assessment of lameness, using sheep as a model. Am. J. Vet. Res . 54 ( 6 ): 976 – 983 . PMID: 8323070. Google Scholar PubMed OpenURL Placeholder Text WorldCat © The Author(s) 2022. Published by Oxford University Press on behalf of the American Society of Animal Science. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected] © The Author(s) 2022. Published by Oxford University Press on behalf of the American Society of Animal Science.

Lenoir, Guillaume; Flatres-Grall, Loïc; Friggens, Nicolas C; David, Ingrid

doi: 10.1093/jas/skac157pmid: 35511420

Abstract The objective was to determine operational proxies for robustness based on data collected routinely on farm that allow phenotyping of these traits in fattening pigs, and to estimate their genetic parameters. A total of 7,256 pigs, from two Piétrain paternal lines (Pie and Pie NN), were tested at the AXIOM boar testing station (Azay-sur-Indre, France) from 2019 to 2021. During the fattening period (from 75 to 150 d of age), individual performance indicators were recorded (growth, backfat, loin depth, feed intake, and feed conversion ratio [FCR]) together with indicators such as insufficient growth, observable defect, symptoms of diseases, and antibiotic and anti-inflammatory injections. These indicators were combined into three categorical robustness scores: R1, R2, and R3. Genetic parameters were estimated using an animal linear model. The robustness score R2 (selectable or not selectable animal) that combined information from status at testing and mortality had the highest heritability estimates of 0.08 ± 0.03 for Pie NN line and a value of 0.09 ± 0.02 for Pie line, compared with traits R1 and R3. The score R3 that combines information from the score R2 with antibiotic and anti-inflammatory injections presented slightly lower heritability estimates (0.05 ± 0.02 to 0.07 ± 0.03). Genetic correlations between R2 and R3 were high and favorable (0.93 ± 0.04 to 0.95 ± 0.03) and R2 and R3 can be considered identical with regard to the confidence interval. These two robustness scores were also highly and favorably genetically correlated with initial body weight and average daily gain, and unfavorably correlated with daily feed intake (ranging from 0.73 ± 0.06 to 0.90 ± 0.08). Estimates of genetic correlations of R2 and R3 with backfat depth and raw FCR (not standardized between starting and finishing weights) were moderate and unfavorable (0.20 ± 0.13 to 0.46 ± 0.20). A part of these genetic correlations, that are of low precision due to the number of data available, have to be confirmed on larger datasets. The results showed the interest of using routine phenotypes collected on farm to build simple robustness indicators that can be applied in breeding. Lay Summary The objective was to determine operational proxies for robustness based on data collected routinely on farm that allow phenotyping of these traits in fattening pigs (from approximately 75 to 150 d of age), and to estimate their genetic parameters. A total of 7,256 pigs, from two Piétrain paternal lines (Pie and Pie NN), were tested. Individual performance indicators were recorded together with indicators such as insufficient growth, observable defects, symptoms of diseases, and antibiotic and anti-inflammatory injections. These indicators were combined into three categorical robustness scores: R1, R2, and R3. The robustness score R2 (selectable or not selectable animal) that combined information from status at testing and mortality had the highest heritability of 0.08 ± 0.03 for Pie NN line and a value of 0.09 ± 0.02 for Pie line. This robustness score was also highly and favorably genetically correlated with initial body weight and average daily gain, and unfavorably correlated with daily feed intake in both lines (ranging from 0.73 ± 0.06 to 0.90 ± 0.08). Estimates of genetic correlations of R2 with backfat depth and feed conversion ratio were moderate and unfavorable (0.20 ± 0.13 to 0.46 ± 0.20). The results showed the interest of using routine phenotypes collected on farm to build simple robustness indicators that can be applied in breeding. Introduction In Europe, livestock farming faces new challenges related to a rapidly changing economic, societal, and environmental context. Societal pressure to “eat healthier” is changing the way pigs are raised and, in particular, leads to a decrease in the use of antibiotics. In France, for example, the level of exposure of pigs to antimicrobials (ALEA) decreased by 41% from 2012 to 2016 (Hémonic et al., 2019). In this context, there will be a greater reliance on the innate robustness of farmed animals. The more general context of global warming implies an increase in the frequency of extreme events, such as heatwaves or droughts (Hansen et al., 2012) having direct (temperature) and indirect impacts (availability of raw materials for feed production) on animals’ rearing environments. All these challenges require having animals able to adapt to these new conditions, which implies an improvement of robustness while maintaining a high level of production. In parallel, improving animal robustness meets the economic expectations of the operators, especially by increasing the viability and reducing treatment costs (Phocas et al., 2016). There is no real consensus on the definition of robustness as well as on the ways to phenotype it. It is not the aim of the present study to add to the list of definitions but rather to evaluate potential proxies of robustness. Nevertheless, our approach is informed by the definition of robustness adapted to the context of artificial selection of Knap (2005). He defined robustness as “the ability to combine a high production potential with resilience to stressors, allowing for unproblematic expression of a high production potential in a wide variety of environmental conditions.” Generally, the production potential is associated with a phenotype of interest, such as growth, feed conversion ratio (FCR), etc. Today, traits included in breeding goals that may be associated with robustness are mainly related to the health status of animals, including resistance to diseases and mortality during a specific period, or to the longevity of reproductive animals (Knap, 2005; Berghof et al., 2019; Knap and Doeschl-Wilson, 2020). Incorporating one or more operational proxies to evaluate the robustness of growing pigs in genetic selection would therefore be of value for the development of more sustainable breeding goals (Berghof et al., 2019). At first sight, animals that have the best performance in a given environment, compared with their contemporaries reared in the same environment, could be considered to be the most robust because they perform well and thus seem to be most adapted to this environment. However, this approach is too narrow as it does not include the costs of achieving this “robustness,” which may be hidden in good and stable environments. It seems important to include more direct measures of other robustness components. Studies have already approached this subject but mainly focused on health-related traits that reflect disease resistance. For example, in rabbits, nonspecific disease resistance traits, based on routinely collected phenotypes, show nonzero heritabilities from 0.04 to 0.11 (Gunia et al., 2018), but without simultaneously incorporating other robustness components. The objective of this study was to determine a set of operational proxies of robustness for fattening pigs (from 75 to 150 d of age) combining aspects of growth, survival, health, and medication; based on phenotypes commonly available on farm; and to evaluate their genetic determinism. In this context, these proxies should reflect the ability of an animal to express or adapt its production potential in the face of changes in the environment relative to other animals that have been raised under the same conditions. Materials and Methods Specific Experimental Animal Care and Use Committee approval was not needed because all the data used in this study were obtained from preexisting databases provided by AXIOM. The data used were from animals raised under commercial conditions that were cared for according to EU-Council directive 2008/120/EC of 18 December 2008 laying down minimum standards for the protection of pigs (http://data.europa.eu/eli/dir/2008/120/oj). Populations Animals from two paternal lines of the Axiom company were used in this study: Piétrain Français (Pie) and Piétrain NN Français free from halothane-sensitivity (Pie NN). These lines are selected on paternal traits for more than 10 generations. In both cases, the objective is to improve the average daily growth (ADG) while reducing FCR during the fattening period. The selection objective is also to meet European market requirements for carcass qualities at 100 kg by reducing the backfat thickness (BF) and improving loin thickness. The animals considered in this study were all males (5,116 Pie and 2,140 Pie NN) raised from January 2019 to April 2021 at the boar testing station of the breeding company AXIOM Genetics (Azay-sur-Indre, France). The station consisted of 2 quarantine rooms, 2 postweaning rooms, and 10 fattening rooms with 12 identical pens each, housing a maximum of 14 pigs per pen, leading to a total capacity of 2,638 places. Each group, from the same week of introduction in the station, was divided into two fattening rooms (24 pens with 14 pigs). Sick pigs were treated with individual medication according to veterinary requirements. The station was not equipped with an air-cooling system. The studied males were born in six different farms (four farms for Pie and two farms for Pie NN) integrated into the AXIOM breeding scheme and that comply with AXIOM’s biosafety and health requirements (monitoring, vaccination plan, etc.), that are negative for monitored diseases (Porcine Reproductive and Respiratory Syndrome, Brucellosis, Classical Swine Fever, Aujesky’s disease, major serotypes of Actinobacillus pleuropneumonia, Porcine Epidemic Diarrhea, Transmissible gastroenteritis, and Swine dysentery) and vaccinated for Mycoplasma pneumoniae and PCV2. These animals came from 1,462 litters in Pie line (3.5 ± 1.8 piglets per litter) and from 951 litters in Pie NN line (2.3 ± 1.3 piglets per litter). They were born from 182 sires in Pie line (28.1 ± 31.8 piglets per sire) and 88 sires in Pie NN line (25.2 ± 16.5 piglets per sire). The pedigrees contained 11,325 animals across 22 generations for Pie and 3,944 animals across 24 generations for Pie NN. To limit the risk of confounding between environmental (i.e., fattening group) and genetic effects, the sires from the two lines were used at least in two mating groups in each farm and in two different farms. Each fattening group consists of animals sourced from one and three farrowing farms in the Pie line and from one or two farrowing farms in the Pie NN line. Pigs from both lines entered the boar testing station at an average age of 27.3 ± 2.2 d with an average body weight (BW) of 8.5 ± 1.7 kg for the two lines at the rate of one group of 336 piglets every 3 wk. They were raised in air-filtered quarantine rooms for 5 wk in pens of 14 animals from the same line and birth farm. These groups of 14 pigs were never modified at the different stages of breeding. During this quarantine period, corresponding to the time required for seroconversion control, animals were controlled for monitored diseases: serological control and observation of symptoms. In the case of positive animals for monitored diseases, the whole group was excluded from the farm. Then, animals were raised in postweaning rooms for 2 wk and transferred to fattening rooms when they were 75.3 ± 3.4 d of age (34.5 ± 6.2 kg BW). Pigs were raised in fattening rooms for 74.8 ± 4.0 d until the individual testing at around 149.7 ± 4.1 d of age (108.9 ± 11.5 kg BW). Fattening rooms were equipped with an automatic feeding system (AFS): Nedap pig performance testing feeding station (Nedap N.V., Groenlo, the Netherlands). Each pen had one water nipple available for the animals. Animals were fed ad libitum with commercial diets adapted to their physiological needs. The provided diets were nonlimiting in amino acids. Information recorded during the fattening period Each animal was individually weighted on arrival in the fattening room (initial body weight, IBW). During the fattening period, BW and feed intake (FI) were recorded each time the animal went into the AFS. In addition, each treatment received by the animal and associated symptoms were recorded, as well as the date of death, if necessary. When the average weight of the group was approximately 100 kg, individual tests were performed. Measurements made during the test were: body weight (TBW), average ultrasonic BF (= mean of three measurements in mm), and ultrasonic longissimus dorsi thickness (LD = one measurement in mm). The BF and LD measures were transformed to correspond to their values at 100 kg live weight (BF100 and LD100, respectively) to compare animals at the equivalent weight. This transformation was done by applying linear coefficients that multiply by the difference between 100 kg and TBW. Coefficients used are 0.04 mm/kg for BF100 and 0.27 mm/kg for LD100 (Sourdioux et al., 2009). Visual observation of the animals was then carried out by the technician in charge of the measurements in order to note the morphological defects, anomalies, and clinical signs of disease according to a frame of reference (Supplementary Appendix 1; Institut Technique du Porc, 2004), noted as “observable defects.” These observations were made by the same person within any given fattening group, and by a total of four technicians over the studied period. To avoid deviations in notations, they used the same reference and were trained together each year. Any systematic differences between technicians would be absorbed in the fattening group effect in the statistical model. Part of these observations was used to construct the robustness traits. Animals weighing less than 70 kg were considered to have too poor growth and these were excluded from the test. This threshold was defined by the French Pork and Pig Institute in their specifications for on-farm testing (Institut Technique du Porc, 2004). These animals were only noted with the observation: “Out of test” and the various performance traits were not recorded for them. The ADG was estimated only for animals with TBW greater than or equal to 70 kg, and calculated as the difference between TBW and IBW divided by the number of days elapsed between the two weighings. The FCR was calculated as the ratio between the total FI during the fattening period and the weight gain (TBW − IBW), expressed in kg/kg. The average daily feed intake (DFI) was calculated as the total FI during the period divided by the number of days elapsed. The residual feed intake (RFI) was also estimated for each animal as the deviation between the recorded DFI and the potential average daily feed intake (PDFI) predicted from requirements for maintenance and production. Based on the method proposed by Labroue et al. (1999), the PDFI was estimated by linear regression, with the lm function in R (R Core Team, 2018), of DFI on average metabolic weight (AMW), ADG, and BF100. The AMW was estimated for each animal using the formula proposed by Noblet et al. (1991): AMW= (TBW1.6−IBW1.6)1.6(TBW−IBW). The estimation of PDFI was computed separately for each line and without including fixed effects. Robustness traits Three synthetic phenotypes to characterize the robustness of the candidates were defined from the measurements performed during the individual test, and from the medical treatments recorded during the testing period (Table 1). The objectives of these synthetic traits were to describe the ability of the animal to be measured at the end of the individual testing present, that is, to be alive and weighing at least 70 kg, and to be in good health without observable defects. The trait R1 corresponded to the distinction used at present in the AXIOM testing protocol to differentiate candidates that can be tested (Note = 1) from those that are dead or weighing less than 70 kg on the day of the individual test (Note = 0). Individual mortality was not available in the database. Consequently, it was not possible to analyze directly this trait. The trait R2 differentiated animals that were selectable, tested, and without any observable defect on the day of testing (Note = 1), from those that were not tested or tested and had an observable defect (Note = 0). We considered it as an observable defect on the day of testing, factors such as weak development and similar, were estimated to relate to the robustness of the animal (see Supplementary Appendix 1 for full description). The trait R3 was a decomposition of the trait R2 in which the category of “selectable” animals was differentiated into those pigs that received at least one antibiotic or anti-inflammatory injection during the testing period (Note = 1) and those that did not receive any injection (Note = 2). For R3, we considered the levels as equidistant as has been commonly done (Varona et al., 1999; Pérez-Cabal and Charfeddine, 2015). We did not include symptoms in the trait definition due to the subjectivity of the observations. Table 1. Description of robustness traits studied Variable . Modality . Entitled . Comment . R1 0 Absent Animal alive but weighing less than 70 kg (not controlled) or dead. 1 Present Animal alive and weighing 70 kg or more (controlled). R2 0 Not selectable Animal « Absent (R1) » or « Present (R1) » with a negative observation (body condition, health status (abscess, respiratory problem, diarrhea, etc.), cannibalism, poor body development) 1 Selectable Animal « Present (R1) » without negative observation R3 0 Not selectable Animal « Not selectable (R2) » 1 Selectable with medicine Animal « Selectable (R2) » with at least one antibiotic or anti-inflammatory injection during the fattening period 2 Selectable without medicine Animal « Selectable (R2) » without any medicine injection during the fattening period Variable . Modality . Entitled . Comment . R1 0 Absent Animal alive but weighing less than 70 kg (not controlled) or dead. 1 Present Animal alive and weighing 70 kg or more (controlled). R2 0 Not selectable Animal « Absent (R1) » or « Present (R1) » with a negative observation (body condition, health status (abscess, respiratory problem, diarrhea, etc.), cannibalism, poor body development) 1 Selectable Animal « Present (R1) » without negative observation R3 0 Not selectable Animal « Not selectable (R2) » 1 Selectable with medicine Animal « Selectable (R2) » with at least one antibiotic or anti-inflammatory injection during the fattening period 2 Selectable without medicine Animal « Selectable (R2) » without any medicine injection during the fattening period Open in new tab Table 1. Description of robustness traits studied Variable . Modality . Entitled . Comment . R1 0 Absent Animal alive but weighing less than 70 kg (not controlled) or dead. 1 Present Animal alive and weighing 70 kg or more (controlled). R2 0 Not selectable Animal « Absent (R1) » or « Present (R1) » with a negative observation (body condition, health status (abscess, respiratory problem, diarrhea, etc.), cannibalism, poor body development) 1 Selectable Animal « Present (R1) » without negative observation R3 0 Not selectable Animal « Not selectable (R2) » 1 Selectable with medicine Animal « Selectable (R2) » with at least one antibiotic or anti-inflammatory injection during the fattening period 2 Selectable without medicine Animal « Selectable (R2) » without any medicine injection during the fattening period Variable . Modality . Entitled . Comment . R1 0 Absent Animal alive but weighing less than 70 kg (not controlled) or dead. 1 Present Animal alive and weighing 70 kg or more (controlled). R2 0 Not selectable Animal « Absent (R1) » or « Present (R1) » with a negative observation (body condition, health status (abscess, respiratory problem, diarrhea, etc.), cannibalism, poor body development) 1 Selectable Animal « Present (R1) » without negative observation R3 0 Not selectable Animal « Not selectable (R2) » 1 Selectable with medicine Animal « Selectable (R2) » with at least one antibiotic or anti-inflammatory injection during the fattening period 2 Selectable without medicine Animal « Selectable (R2) » without any medicine injection during the fattening period Open in new tab In addition, the area between curves (ABC) index estimated during fattening period, developed by Revilla et al. (2022) which the authors called a resilience phenotype, was also calculated using weight measured by AFS for each animal alive at the end of the fattening period. The datasets analyzed by Revilla et al. (2022) were collected on the same farm from September 2015 to July 2019. The trait ABC was the accumulated difference of area between the unperturbed growth curve and the perturbed curve. The ABC index had no unit. The unperturbed growth curve of each animal was modeled using the Gompertz equation. The perturbed curve was constructed using linear interpolation of body weight measurements recorded by AFS. In comparison to the method proposed by Revilla et al. (2022), some modifications have been made to the data pretreatment when analyzing all data from each AFS within fattening group. This was done to detect inconsistencies linked to the AFS machine. A linear regression of weight on the number of days of test was applied. The standard deviation of the residual value was calculated for each day for each AFS within fattening group. If more than 20% of the weights measured on AFS in a fattening group were greater or less than 4 standard deviations, then records of the AFS within the fattening group were removed from the data set. Animals with less than 20 d of measurements in total were removed from the analysis. The ratio of the number of animals after cleaning procedure and the raw dataset was 0.93 in Pie line and 0.92 in Pie NN line. Statistical analysis Differences in the phenotypic means between the lines were tested using analysis of variance on R software (R Core Team, 2018). To compare the differences and frequencies in the three robustness scores among the two lines, a Chi-square on R was performed. Statistical significance was set a priori at P ≤ 0.05. Genetic parameters estimation Each trait was analyzed with ASREML 3.0 software (Gilmour et al., 2009), using the restricted maximum likelihood (REML) method. Each line was analyzed separately. Firstly, to select fixed and random effects, all traits were analyzed using a single trait model. The global linear mixed model was defined as: y=μ+Xβ+Vb+Wl+ Zu+e where y was the vector of phenotypes for the considered trait (R1, R2, and R3 considered as continuous phenotypes, IBW, ADG, LD100, BF100, FCR, DFI, and ABC); μ is the overall mean, β was the vector of fixed effects; b was the vector of random fattening group effect, with ∼N(0,Ibσb2) ⁠, where Iwas the identity matrix of appropriate size; l was a vector of the common litter random effects with ∼N(0,Ilσl2) ⁠; u was the vector of additive genetic random effects with ∼N(0,Aσu2) ⁠, where A was the pedigree-based relationship matrix; X ⁠, V ⁠, W ⁠, and Z were the known incidence matrices for fixed, fattening group effects, litter effect, and animal genetic, respectively; and e is the vector of residual random effects with ∼N(0,Ieσe2) ⁠. For all estimated traits, the fixed effects tested at an α-risk of 5% using the Wald F statistic of ASReml (Gilmour et al., 2009) were the birth farm for Pie and Pie NN and halothane-sensitivity gene status for Pie line. The significance of common litter random effect was tested by using likelihood ratio test with R statistical software (R Core Team, 2018) from log-likelihood values estimated on ASReml software (Gilmour et al., 2009). In the two lines, the fattening group effect and common litter effect were significant (P < 0.05) for all tested traits expected for traits ABC, FCR, and RFI with the common litter effect in Pie NN line. In the second step of the analysis, to follow the assumption of the BLUP method, which should be applied to a nonselected base population, and to estimate the covariance between traits, a series of multi-traits models including the four traits under selection (ADG, BF100, LD100, and FCR) and the nonselected traits to be estimated (R1, R2, R3, IBW, DFI, RFI, and ABC) were applied to the data. First, variance and covariance components were estimated with a four-trait linear animal model including ADG, FCR, BF100, and LD100 traits, to estimate heritabilities and genetic correlations of traits under selection. Second, to estimate heritability for each nonselected trait (R1, R2, R3, IBW, DFI, RFI, and ABC) and their genetic correlations with the traits under selection, five-trait linear animal models including the four traits under selection and one trait to be estimated were used. Third, to estimate genetic correlations between the nonselected traits (R1, R2, R3, ABC, IBW, DFI, and RFI), six-trait linear animal models including the four traits under selection as well as the two traits for which the genetic correlation is estimated were performed. Heritability (h²) was calculated as the ratio of animal genetic variance to the total phenotypic variance, that is, the sum of the genetic additive variance, environmental variances (fattening group, litter if necessary), and the residual variance, estimated with the four-trait model for the traits under selection and with the five-trait models for the nonselected traits. Results Phenotypic means and distributions Means of TBW were similar between the two lines (Table 2). The Pie NN animals had significantly lower average values (P < 0.05) for IBW (−0.4 kg), ADG (−17 g/d), DFI (−23 g/d), and LD100 (−4.8 mm) and significantly higher average values for BF100 (+0.6 mm) and FCR (+0.01 kg/kg) than Pie. The mean and SD for ABC values were significantly higher for Pie NN animals compared with Pie (+5,223, i.e., +20.5% of ABC), indicating more important deviations between unperturbed and perturbed growth in line Pie NN, suggesting that these are average less robust animals compared with the Pie line. The distributions of the traits R1 and R2 were similar (P > 0.1) between Pie and Pie NN (Figure 1). Approximately 95% of the animals introduced in fattening rooms were “Present” (Trait R1—Note = 1), on the day of individual testing and around 80% were “Selectable” (Trait R2—Note = 1). The mortality rate over the fattening period for Pie pigs (3.75%) was significantly higher than for Pie NN pigs (2.42%). Proportions of animals with observable defects at individual testing were 13.6% and 15.5% in Pie and Pie NN lines, respectively. For the trait R3, the Pie line had a significantly higher proportion (P < 0.001) of animals “Selectable with medicine” (Trait R3—Note = 1) than the Pie NN line (32.2% vs. 19.7%, respectively). Table 2. Descriptive statistics (mean and SD) for area between curves and production traits for each line1 and significance level of difference (P) Trait, unit2 . Pie (n = 5,116) . Pie NN (n = 2,140) . P3 . Mean . SD . Mean . SD . IBW, kg 34.5 6.1 34.1 5.7 * TBW, kg 108.9 11.4 108.8 11.5 ADG, g/d 1,009 104 992 108 * FCR, kg/kg 2.25 0.18 2.26 0.19 * DFI, g/d 2,263 268 2240 287 * RFI, g/d 0 150 0 159 BF100, mm 6.0 0.8 6.6 0.8 * LD100, mm 72.8 5.1 68.0 5.3 * ABC 25,503 21,603 30,726 24,764 * Trait, unit2 . Pie (n = 5,116) . Pie NN (n = 2,140) . P3 . Mean . SD . Mean . SD . IBW, kg 34.5 6.1 34.1 5.7 * TBW, kg 108.9 11.4 108.8 11.5 ADG, g/d 1,009 104 992 108 * FCR, kg/kg 2.25 0.18 2.26 0.19 * DFI, g/d 2,263 268 2240 287 * RFI, g/d 0 150 0 159 BF100, mm 6.0 0.8 6.6 0.8 * LD100, mm 72.8 5.1 68.0 5.3 * ABC 25,503 21,603 30,726 24,764 * Pie, Piétrain Français; Pie NN, Piétrain NN Français free from halothane-sensitivity. IBW, initial body weight; TBW, testing body weight; ADG, average daily gain; FCR, feed conversion ratio; DFI, average daily feed intake; RFI, residual feed intake; BF100, backfat thickness estimated at 100 kg live weight; LD100, longissimus dorsi thickness estimated at 100 kg live weight; ABC, resilience index. P value for the difference between least squares means of Pie and Pie NN lines. P < 0.05. Open in new tab Table 2. Descriptive statistics (mean and SD) for area between curves and production traits for each line1 and significance level of difference (P) Trait, unit2 . Pie (n = 5,116) . Pie NN (n = 2,140) . P3 . Mean . SD . Mean . SD . IBW, kg 34.5 6.1 34.1 5.7 * TBW, kg 108.9 11.4 108.8 11.5 ADG, g/d 1,009 104 992 108 * FCR, kg/kg 2.25 0.18 2.26 0.19 * DFI, g/d 2,263 268 2240 287 * RFI, g/d 0 150 0 159 BF100, mm 6.0 0.8 6.6 0.8 * LD100, mm 72.8 5.1 68.0 5.3 * ABC 25,503 21,603 30,726 24,764 * Trait, unit2 . Pie (n = 5,116) . Pie NN (n = 2,140) . P3 . Mean . SD . Mean . SD . IBW, kg 34.5 6.1 34.1 5.7 * TBW, kg 108.9 11.4 108.8 11.5 ADG, g/d 1,009 104 992 108 * FCR, kg/kg 2.25 0.18 2.26 0.19 * DFI, g/d 2,263 268 2240 287 * RFI, g/d 0 150 0 159 BF100, mm 6.0 0.8 6.6 0.8 * LD100, mm 72.8 5.1 68.0 5.3 * ABC 25,503 21,603 30,726 24,764 * Pie, Piétrain Français; Pie NN, Piétrain NN Français free from halothane-sensitivity. IBW, initial body weight; TBW, testing body weight; ADG, average daily gain; FCR, feed conversion ratio; DFI, average daily feed intake; RFI, residual feed intake; BF100, backfat thickness estimated at 100 kg live weight; LD100, longissimus dorsi thickness estimated at 100 kg live weight; ABC, resilience index. P value for the difference between least squares means of Pie and Pie NN lines. P < 0.05. Open in new tab Figure 1. Open in new tabDownload slide Distribution of modalities for the three robustness traits (R1, R2, and R3) for the Pie and Pie NN lines. Pie, Piétrain Français; Pie NN, Piétrain NN Français free from halothane-sensitivity. Variance–Covariance components The heritability estimates for robustness traits R1, R2, and R3 were low and in the same range for the two lines, ranging from 0.03 ± 0.01 to 0.09 ± 0.02 (Table 3). Heritability estimates for R2 and R3 tended to be slightly higher than for R1 in each line. Heritability estimates for the ABC index were low for both Pie (0.09 ± 0.03) and Pie NN (0.06 ± 0.03). Heritability estimates were low to moderate in the Pie and Pie NN lines for the traits under selection (ADG, FCR, BF100, and LD100), and also for IBW, DFI, and RFI, ranging from 0.13 ± 0.03 to 0.34 ± 0.05. The higher standard errors in Pie NN were due to the smaller dataset for this line. The fattening group effect ranged from 0.02 ± 0.01 to 0.38 ± 0.07 for the studied traits, with the highest estimates being for LD100 in both lines. The proportion of variance due to common litter effects was similar for all traits, ranging from 0.03 ± 0.01 to 0.08 ± 0.03, except for IBW in the two lines and ABC in Pie line that had the highest proportion of phenotypic variance explained by litter effect. Table 3. Estimates of heritability (h²), fattening group effect ratio (b²), common litter effect (c²), and phenotypic variance (Vp) for the traits recorded (± standard error) for each line1 Trait2 . Pie . Pie NN . h² . b² . c² . Vp . h² . b² . c² . Vp . R13 0.03 ± 0.01 0.02 ± 0.01 0.03 ± 0.01 0.054 ± 0.001 0.06 ± 0.03 0.04 ± 0.01 0.06 ± 0.02 0.045 ± 0.002 R23 0.09 ± 0.02 0.03 ± 0.01 0.03 ± 0.01 0.157 ± 0.004 0.08 ± 0.03 0.02 ± 0.01 0.06 ± 0.02 0.162 ± 0.005 R33 0.05 ± 0.02 0.07 ± 0.02 0.07 ± 0.01 0.590 ± 0.016 0.07 ± 0.03 0.03 ± 0.01 0.07 ± 0.02 0.648 ± 0.021 ABC3,4 0.09 ± 0.03 0.06 ± 0.02 0.16 ± 0.02 4.95 × 108 ± 1.59 × 107 0.06 ± 0.02 0.05 ± 0.02 6.20 × 108 ± 2.36 × 107 IBW3 0.34 ± 0.05 0.18 ± 0.04 0.13 ± 0.02 36.38 ± 1.88 0.33 ± 0.06 0.20 ± 0.04 0.16 ± 0.03 34.88 ± 2.17 ADG4 0.21 ± 0.04 0.13 ± 0.03 0.07 ± 0.02 10,694 ± 462 0.32 ± 0.06 0.11 ± 0.03 0.08 ± 0.03 12,274 ± 623 FCR4 0.23 ± 0.04 0.15 ± 0.03 0.04 ± 0.01 0.034 ± 0.002 0.15 ± 0.04 0.13 ± 0.03 0.037 ± 0.002 DFI3 0.29 ± 0.05 0.16 ± 0.04 0.08 ± 0.02 73,364 ± 3,661 0.31 ± 0.07 0.16 ± 0.04 0.08 ± 0.03 87,972 ± 5,299 RFI3 0.19 ± 0.04 0.16 ± 0.04 0.06 ± 0.02 23,575 ± 1,159 0.13 ± 0.04 0.12 ± 0.03 26,363 ± 1301 BF1004 0.31 ± 0.04 0.12 ± 0.03 0.05 ± 0.02 0.624 ± 0.027 0.29 ± 0.06 0.12 ± 0.03 0.07 ± 0.03 0.717 ± 0.037 LD1004 0.17 ± 0.03 0.43 ± 0.06 0.03 ± 0.01 31.47 ± 3.54 0.25 ± 0.05 0.38 ± 0.07 0.03 ± 0.02 28.45 ± 3.08 Trait2 . Pie . Pie NN . h² . b² . c² . Vp . h² . b² . c² . Vp . R13 0.03 ± 0.01 0.02 ± 0.01 0.03 ± 0.01 0.054 ± 0.001 0.06 ± 0.03 0.04 ± 0.01 0.06 ± 0.02 0.045 ± 0.002 R23 0.09 ± 0.02 0.03 ± 0.01 0.03 ± 0.01 0.157 ± 0.004 0.08 ± 0.03 0.02 ± 0.01 0.06 ± 0.02 0.162 ± 0.005 R33 0.05 ± 0.02 0.07 ± 0.02 0.07 ± 0.01 0.590 ± 0.016 0.07 ± 0.03 0.03 ± 0.01 0.07 ± 0.02 0.648 ± 0.021 ABC3,4 0.09 ± 0.03 0.06 ± 0.02 0.16 ± 0.02 4.95 × 108 ± 1.59 × 107 0.06 ± 0.02 0.05 ± 0.02 6.20 × 108 ± 2.36 × 107 IBW3 0.34 ± 0.05 0.18 ± 0.04 0.13 ± 0.02 36.38 ± 1.88 0.33 ± 0.06 0.20 ± 0.04 0.16 ± 0.03 34.88 ± 2.17 ADG4 0.21 ± 0.04 0.13 ± 0.03 0.07 ± 0.02 10,694 ± 462 0.32 ± 0.06 0.11 ± 0.03 0.08 ± 0.03 12,274 ± 623 FCR4 0.23 ± 0.04 0.15 ± 0.03 0.04 ± 0.01 0.034 ± 0.002 0.15 ± 0.04 0.13 ± 0.03 0.037 ± 0.002 DFI3 0.29 ± 0.05 0.16 ± 0.04 0.08 ± 0.02 73,364 ± 3,661 0.31 ± 0.07 0.16 ± 0.04 0.08 ± 0.03 87,972 ± 5,299 RFI3 0.19 ± 0.04 0.16 ± 0.04 0.06 ± 0.02 23,575 ± 1,159 0.13 ± 0.04 0.12 ± 0.03 26,363 ± 1301 BF1004 0.31 ± 0.04 0.12 ± 0.03 0.05 ± 0.02 0.624 ± 0.027 0.29 ± 0.06 0.12 ± 0.03 0.07 ± 0.03 0.717 ± 0.037 LD1004 0.17 ± 0.03 0.43 ± 0.06 0.03 ± 0.01 31.47 ± 3.54 0.25 ± 0.05 0.38 ± 0.07 0.03 ± 0.02 28.45 ± 3.08 Pie, Piétrain Français; Pie NN, Piétrain NN Français free from halothane-sensitivity. IBW, initial body weight; TBW, testing body weight; ADG, average daily gain; FCR, feed conversion ratio; DFI, average daily feed intake; RFI, residual feed intake; BF100, backfat thickness estimated at 100 kg live weight; LD100, longissimus dorsi thickness estimated at 100 kg live weight; ABC, resilience index. Estimates from a five-trait multiple trait model (ADG, FCR, BF100, LD100, and the trait under consideration). Estimates from a four-trait multiple trait model (ADG, FCR, BF100, and LD100). Open in new tab Table 3. Estimates of heritability (h²), fattening group effect ratio (b²), common litter effect (c²), and phenotypic variance (Vp) for the traits recorded (± standard error) for each line1 Trait2 . Pie . Pie NN . h² . b² . c² . Vp . h² . b² . c² . Vp . R13 0.03 ± 0.01 0.02 ± 0.01 0.03 ± 0.01 0.054 ± 0.001 0.06 ± 0.03 0.04 ± 0.01 0.06 ± 0.02 0.045 ± 0.002 R23 0.09 ± 0.02 0.03 ± 0.01 0.03 ± 0.01 0.157 ± 0.004 0.08 ± 0.03 0.02 ± 0.01 0.06 ± 0.02 0.162 ± 0.005 R33 0.05 ± 0.02 0.07 ± 0.02 0.07 ± 0.01 0.590 ± 0.016 0.07 ± 0.03 0.03 ± 0.01 0.07 ± 0.02 0.648 ± 0.021 ABC3,4 0.09 ± 0.03 0.06 ± 0.02 0.16 ± 0.02 4.95 × 108 ± 1.59 × 107 0.06 ± 0.02 0.05 ± 0.02 6.20 × 108 ± 2.36 × 107 IBW3 0.34 ± 0.05 0.18 ± 0.04 0.13 ± 0.02 36.38 ± 1.88 0.33 ± 0.06 0.20 ± 0.04 0.16 ± 0.03 34.88 ± 2.17 ADG4 0.21 ± 0.04 0.13 ± 0.03 0.07 ± 0.02 10,694 ± 462 0.32 ± 0.06 0.11 ± 0.03 0.08 ± 0.03 12,274 ± 623 FCR4 0.23 ± 0.04 0.15 ± 0.03 0.04 ± 0.01 0.034 ± 0.002 0.15 ± 0.04 0.13 ± 0.03 0.037 ± 0.002 DFI3 0.29 ± 0.05 0.16 ± 0.04 0.08 ± 0.02 73,364 ± 3,661 0.31 ± 0.07 0.16 ± 0.04 0.08 ± 0.03 87,972 ± 5,299 RFI3 0.19 ± 0.04 0.16 ± 0.04 0.06 ± 0.02 23,575 ± 1,159 0.13 ± 0.04 0.12 ± 0.03 26,363 ± 1301 BF1004 0.31 ± 0.04 0.12 ± 0.03 0.05 ± 0.02 0.624 ± 0.027 0.29 ± 0.06 0.12 ± 0.03 0.07 ± 0.03 0.717 ± 0.037 LD1004 0.17 ± 0.03 0.43 ± 0.06 0.03 ± 0.01 31.47 ± 3.54 0.25 ± 0.05 0.38 ± 0.07 0.03 ± 0.02 28.45 ± 3.08 Trait2 . Pie . Pie NN . h² . b² . c² . Vp . h² . b² . c² . Vp . R13 0.03 ± 0.01 0.02 ± 0.01 0.03 ± 0.01 0.054 ± 0.001 0.06 ± 0.03 0.04 ± 0.01 0.06 ± 0.02 0.045 ± 0.002 R23 0.09 ± 0.02 0.03 ± 0.01 0.03 ± 0.01 0.157 ± 0.004 0.08 ± 0.03 0.02 ± 0.01 0.06 ± 0.02 0.162 ± 0.005 R33 0.05 ± 0.02 0.07 ± 0.02 0.07 ± 0.01 0.590 ± 0.016 0.07 ± 0.03 0.03 ± 0.01 0.07 ± 0.02 0.648 ± 0.021 ABC3,4 0.09 ± 0.03 0.06 ± 0.02 0.16 ± 0.02 4.95 × 108 ± 1.59 × 107 0.06 ± 0.02 0.05 ± 0.02 6.20 × 108 ± 2.36 × 107 IBW3 0.34 ± 0.05 0.18 ± 0.04 0.13 ± 0.02 36.38 ± 1.88 0.33 ± 0.06 0.20 ± 0.04 0.16 ± 0.03 34.88 ± 2.17 ADG4 0.21 ± 0.04 0.13 ± 0.03 0.07 ± 0.02 10,694 ± 462 0.32 ± 0.06 0.11 ± 0.03 0.08 ± 0.03 12,274 ± 623 FCR4 0.23 ± 0.04 0.15 ± 0.03 0.04 ± 0.01 0.034 ± 0.002 0.15 ± 0.04 0.13 ± 0.03 0.037 ± 0.002 DFI3 0.29 ± 0.05 0.16 ± 0.04 0.08 ± 0.02 73,364 ± 3,661 0.31 ± 0.07 0.16 ± 0.04 0.08 ± 0.03 87,972 ± 5,299 RFI3 0.19 ± 0.04 0.16 ± 0.04 0.06 ± 0.02 23,575 ± 1,159 0.13 ± 0.04 0.12 ± 0.03 26,363 ± 1301 BF1004 0.31 ± 0.04 0.12 ± 0.03 0.05 ± 0.02 0.624 ± 0.027 0.29 ± 0.06 0.12 ± 0.03 0.07 ± 0.03 0.717 ± 0.037 LD1004 0.17 ± 0.03 0.43 ± 0.06 0.03 ± 0.01 31.47 ± 3.54 0.25 ± 0.05 0.38 ± 0.07 0.03 ± 0.02 28.45 ± 3.08 Pie, Piétrain Français; Pie NN, Piétrain NN Français free from halothane-sensitivity. IBW, initial body weight; TBW, testing body weight; ADG, average daily gain; FCR, feed conversion ratio; DFI, average daily feed intake; RFI, residual feed intake; BF100, backfat thickness estimated at 100 kg live weight; LD100, longissimus dorsi thickness estimated at 100 kg live weight; ABC, resilience index. Estimates from a five-trait multiple trait model (ADG, FCR, BF100, LD100, and the trait under consideration). Estimates from a four-trait multiple trait model (ADG, FCR, BF100, and LD100). Open in new tab Genetic correlations between R1 and the two other robustness traits were low to moderate in Pie NN line, ranging from 0.25 ± 0.32 to 0.41 ± 0.30 (Table 5), and higher in Pie line, ranging from 0.42 ± 0.28 to 0.57 ± 0.36 (Table 4). Several estimates of genetic correlations had large standard errors and should be interpreted with caution. In both paternal lines, the genetic correlation between R2 and R3 was high (0.95 ± 0.04 and 0.92 ± 0.06, for Pie NN and Pie lines, respectively). The genetic correlation between ABC and the robustness traits tended to be negative in the Pie NN line, ranging from −0.03 ± 0.33 to −0.21 ± 0.38, and in the Pie line, ranging from −0.08 ± 0.26 to −0.22 ± 0.22, none of these correlations were significantly different from 0. In both lines, the traits R2 and R3 were highly correlated with ADG (correlations higher than 0.76), and moderately correlated with FCR, ranging from 0.32 ± 0.18 to 0.51 ± 0.25. The trait R1 had low correlations with ADG, which were 0.22 ± 0.25 in Pie line and 0.31 ± 0.25 in Pie NN line. The carcass traits (BF100 and LD100) tended to be positively correlated with the three robustness traits (estimates ranged from 0.11 ± 0.21 to 0.44 ± 0.25). For the nonselected traits, R2 and R3 were moderate to highly correlated with IBW and DFI (correlations higher than 0.45 ± 0.18). Correlations of R1 with IBW and DFI were null or moderate, ranging from −0.02 ± 0.23 to 0.33 ± 0.27, in both lines. Estimates of genetic correlations of RFI with robustness traits were not significantly different than 0 in both lines. Estimates of genetic correlations of ABC with other traits had large standard errors and showed estimates close to 0, except for IBW in both lines and for LD100 in Pie line with negative correlations. In addition, the genetic correlations between all studied traits are presented in Supplementary Appendix 2 for Pie and in Supplementary Appendix 3for Pie NN. Table 4. Estimates of genetic correlations (r²a ± standard error) between robustness traits (R1, R2, and R3), area between curves, and production traits for Piétrain line (Pie) Trait1 . R1 . R2 . R3 . ABC . R1 0.57 ± 0.282 0.42 ± 0.362 −0.17 ± 0.182 R2 0.57 ± 0.282 0.92 ± 0.062 −0.22 ± 0.262 ABC −0.17 ± 0.182 −0.22 ± 0.262 −0.08 ± 0.292 IBW 0.18 ± 0.222 0.50 ± 0.152 0.45 ± 0.182 −0.19 ± 0.182 ADG 0.22 ± 0.253 0.79 ± 0.083 0.78 ± 0.123 0.00 ± 0.193 FCR 0.21 ± 0.313 0.39 ± 0.153 0.32 ± 0.183 −0.10 ± 0.183 DFI 0.33 ± 0.272 0.73 ± 0.112 0.72 ± 0.122 −0.02 ± 0.162 RFI 0.23 ± 0.172 0.10 ± 0.202 0.07 ± 0.222 −0.05 ± 0.102 BF100 0.21 ± 0.233 0.29 ± 0.143 0.29 ± 0.173 0.01 ± 0.173 LD100 0.42 ± 0.233 0.15 ± 0.153 0.14 ± 0.183 −0.30 ± 0.183 Trait1 . R1 . R2 . R3 . ABC . R1 0.57 ± 0.282 0.42 ± 0.362 −0.17 ± 0.182 R2 0.57 ± 0.282 0.92 ± 0.062 −0.22 ± 0.262 ABC −0.17 ± 0.182 −0.22 ± 0.262 −0.08 ± 0.292 IBW 0.18 ± 0.222 0.50 ± 0.152 0.45 ± 0.182 −0.19 ± 0.182 ADG 0.22 ± 0.253 0.79 ± 0.083 0.78 ± 0.123 0.00 ± 0.193 FCR 0.21 ± 0.313 0.39 ± 0.153 0.32 ± 0.183 −0.10 ± 0.183 DFI 0.33 ± 0.272 0.73 ± 0.112 0.72 ± 0.122 −0.02 ± 0.162 RFI 0.23 ± 0.172 0.10 ± 0.202 0.07 ± 0.222 −0.05 ± 0.102 BF100 0.21 ± 0.233 0.29 ± 0.143 0.29 ± 0.173 0.01 ± 0.173 LD100 0.42 ± 0.233 0.15 ± 0.153 0.14 ± 0.183 −0.30 ± 0.183 IBW, initial body weight; TBW, testing body weight; ADG, average daily gain; FCR, feed conversion ratio; DFI, average daily feed intake; RFI, residual feed intake; BF100, backfat thickness estimated at 100 kg live weight; LD100, longissimus dorsi thickness estimated at 100 kg live weight; ABC, resilience index. Estimates from a six-trait multiple trait model (ADG, FCR, BF100, LD100, and the two traits under consideration). Estimates from a five-traits multiple trait model (ADG, FCR, BF100, LD100, and the trait under consideration). Open in new tab Table 4. Estimates of genetic correlations (r²a ± standard error) between robustness traits (R1, R2, and R3), area between curves, and production traits for Piétrain line (Pie) Trait1 . R1 . R2 . R3 . ABC . R1 0.57 ± 0.282 0.42 ± 0.362 −0.17 ± 0.182 R2 0.57 ± 0.282 0.92 ± 0.062 −0.22 ± 0.262 ABC −0.17 ± 0.182 −0.22 ± 0.262 −0.08 ± 0.292 IBW 0.18 ± 0.222 0.50 ± 0.152 0.45 ± 0.182 −0.19 ± 0.182 ADG 0.22 ± 0.253 0.79 ± 0.083 0.78 ± 0.123 0.00 ± 0.193 FCR 0.21 ± 0.313 0.39 ± 0.153 0.32 ± 0.183 −0.10 ± 0.183 DFI 0.33 ± 0.272 0.73 ± 0.112 0.72 ± 0.122 −0.02 ± 0.162 RFI 0.23 ± 0.172 0.10 ± 0.202 0.07 ± 0.222 −0.05 ± 0.102 BF100 0.21 ± 0.233 0.29 ± 0.143 0.29 ± 0.173 0.01 ± 0.173 LD100 0.42 ± 0.233 0.15 ± 0.153 0.14 ± 0.183 −0.30 ± 0.183 Trait1 . R1 . R2 . R3 . ABC . R1 0.57 ± 0.282 0.42 ± 0.362 −0.17 ± 0.182 R2 0.57 ± 0.282 0.92 ± 0.062 −0.22 ± 0.262 ABC −0.17 ± 0.182 −0.22 ± 0.262 −0.08 ± 0.292 IBW 0.18 ± 0.222 0.50 ± 0.152 0.45 ± 0.182 −0.19 ± 0.182 ADG 0.22 ± 0.253 0.79 ± 0.083 0.78 ± 0.123 0.00 ± 0.193 FCR 0.21 ± 0.313 0.39 ± 0.153 0.32 ± 0.183 −0.10 ± 0.183 DFI 0.33 ± 0.272 0.73 ± 0.112 0.72 ± 0.122 −0.02 ± 0.162 RFI 0.23 ± 0.172 0.10 ± 0.202 0.07 ± 0.222 −0.05 ± 0.102 BF100 0.21 ± 0.233 0.29 ± 0.143 0.29 ± 0.173 0.01 ± 0.173 LD100 0.42 ± 0.233 0.15 ± 0.153 0.14 ± 0.183 −0.30 ± 0.183 IBW, initial body weight; TBW, testing body weight; ADG, average daily gain; FCR, feed conversion ratio; DFI, average daily feed intake; RFI, residual feed intake; BF100, backfat thickness estimated at 100 kg live weight; LD100, longissimus dorsi thickness estimated at 100 kg live weight; ABC, resilience index. Estimates from a six-trait multiple trait model (ADG, FCR, BF100, LD100, and the two traits under consideration). Estimates from a five-traits multiple trait model (ADG, FCR, BF100, LD100, and the trait under consideration). Open in new tab Table 5. Estimates of genetic correlations (r²a ± standard error) between robustness traits (R1, R2, and R3), area between curves, and production traits for Piétrain NN line (Pie NN) Trait1 . R1 . R2 . R3 . ABC . R1 0.41 ± 0.302 0.25 ± 0.322 −0.21 ± 0.382 R2 0.41 ± 0.302 0.95 ± 0.042 −0.03 ± 0.332 ABC −0.21 ± 0.382 −0.03 ± 0.332 −0.10 ± 0.312 IBW −0.02 ± 0.232 0.89 ± 0.142 0.78 ± 0.152 −0.23 ± 0.192 ADG 0.31 ± 0.253 0.86 ± 0.113 0.76 ± 0.143 0.01 ± 0.263 FCR 0.21 ± 0.313 0.51 ± 0.253 0.42 ± 0.263 0.08 ± 0.313 DFI 0.32 ± 0.252 0.91 ± 0.072 0.81 ± 0.152 −0.02 ± 0.162 RFI 0.15 ± 0.352 0.09 ± 0.252 −0.05 ± 0.312 0.19 ± 0.332 BF100 0.35 ± 0.203 0.44 ± 0.253 0.23 ± 0.213 −0.05 ± 0.253 LD100 0.14 ± 0.243 0.11 ± 0.213 0.42 ± 0.263 0.08 ± 0.313 Trait1 . R1 . R2 . R3 . ABC . R1 0.41 ± 0.302 0.25 ± 0.322 −0.21 ± 0.382 R2 0.41 ± 0.302 0.95 ± 0.042 −0.03 ± 0.332 ABC −0.21 ± 0.382 −0.03 ± 0.332 −0.10 ± 0.312 IBW −0.02 ± 0.232 0.89 ± 0.142 0.78 ± 0.152 −0.23 ± 0.192 ADG 0.31 ± 0.253 0.86 ± 0.113 0.76 ± 0.143 0.01 ± 0.263 FCR 0.21 ± 0.313 0.51 ± 0.253 0.42 ± 0.263 0.08 ± 0.313 DFI 0.32 ± 0.252 0.91 ± 0.072 0.81 ± 0.152 −0.02 ± 0.162 RFI 0.15 ± 0.352 0.09 ± 0.252 −0.05 ± 0.312 0.19 ± 0.332 BF100 0.35 ± 0.203 0.44 ± 0.253 0.23 ± 0.213 −0.05 ± 0.253 LD100 0.14 ± 0.243 0.11 ± 0.213 0.42 ± 0.263 0.08 ± 0.313 IBW, initial body weight; TBW, testing body weight; ADG, average daily gain; FCR, feed conversion ratio; DFI, average daily feed intake; RFI, residual feed intake; BF100, backfat thickness estimated at 100 kg live weight; LD100, longissimus dorsi thickness estimated at 100 kg live weight; ABC, resilience index. Estimates from a six-trait multiple trait model (ADG, FCR, BF100, LD100, and the two traits under consideration). Estimates from a five-trait multiple trait model (ADG, FCR, BF100, LD100, and the trait under consideration). Open in new tab Table 5. Estimates of genetic correlations (r²a ± standard error) between robustness traits (R1, R2, and R3), area between curves, and production traits for Piétrain NN line (Pie NN) Trait1 . R1 . R2 . R3 . ABC . R1 0.41 ± 0.302 0.25 ± 0.322 −0.21 ± 0.382 R2 0.41 ± 0.302 0.95 ± 0.042 −0.03 ± 0.332 ABC −0.21 ± 0.382 −0.03 ± 0.332 −0.10 ± 0.312 IBW −0.02 ± 0.232 0.89 ± 0.142 0.78 ± 0.152 −0.23 ± 0.192 ADG 0.31 ± 0.253 0.86 ± 0.113 0.76 ± 0.143 0.01 ± 0.263 FCR 0.21 ± 0.313 0.51 ± 0.253 0.42 ± 0.263 0.08 ± 0.313 DFI 0.32 ± 0.252 0.91 ± 0.072 0.81 ± 0.152 −0.02 ± 0.162 RFI 0.15 ± 0.352 0.09 ± 0.252 −0.05 ± 0.312 0.19 ± 0.332 BF100 0.35 ± 0.203 0.44 ± 0.253 0.23 ± 0.213 −0.05 ± 0.253 LD100 0.14 ± 0.243 0.11 ± 0.213 0.42 ± 0.263 0.08 ± 0.313 Trait1 . R1 . R2 . R3 . ABC . R1 0.41 ± 0.302 0.25 ± 0.322 −0.21 ± 0.382 R2 0.41 ± 0.302 0.95 ± 0.042 −0.03 ± 0.332 ABC −0.21 ± 0.382 −0.03 ± 0.332 −0.10 ± 0.312 IBW −0.02 ± 0.232 0.89 ± 0.142 0.78 ± 0.152 −0.23 ± 0.192 ADG 0.31 ± 0.253 0.86 ± 0.113 0.76 ± 0.143 0.01 ± 0.263 FCR 0.21 ± 0.313 0.51 ± 0.253 0.42 ± 0.263 0.08 ± 0.313 DFI 0.32 ± 0.252 0.91 ± 0.072 0.81 ± 0.152 −0.02 ± 0.162 RFI 0.15 ± 0.352 0.09 ± 0.252 −0.05 ± 0.312 0.19 ± 0.332 BF100 0.35 ± 0.203 0.44 ± 0.253 0.23 ± 0.213 −0.05 ± 0.253 LD100 0.14 ± 0.243 0.11 ± 0.213 0.42 ± 0.263 0.08 ± 0.313 IBW, initial body weight; TBW, testing body weight; ADG, average daily gain; FCR, feed conversion ratio; DFI, average daily feed intake; RFI, residual feed intake; BF100, backfat thickness estimated at 100 kg live weight; LD100, longissimus dorsi thickness estimated at 100 kg live weight; ABC, resilience index. Estimates from a six-trait multiple trait model (ADG, FCR, BF100, LD100, and the two traits under consideration). Estimates from a five-trait multiple trait model (ADG, FCR, BF100, LD100, and the trait under consideration). Open in new tab Discussion Genetic parameters for robustness traits The heritabilities for the traits R1, R2, and R3 in the present study were low but not null, with the exception of R1 in Pie line (related to the standard error of the estimate). The heritability estimates from our study were in the same range as those presented in different publications estimated at an individual level (Gunia et al., 2015, 2018; Putz et al., 2019; Shrestha et al., 2020) or at the full-sibs level (Gorssen et al., 2021). However, it should be noted that most literature references to similar traits have focused on traits related to the resistance to nonspecific or specific diseases, or related to the use of antibiotics. The heritability estimates for R1 in the two breeds were of the same order of magnitude as the values reported by Perez et al. (2021) on two survival traits (juvenile and late) in turkeys raised under classical production conditions, 0.06 ± 0.01 and 0.04 ± 0.03, respectively. In growing rabbits, the heritability for infectious mortality estimated by Gunia et al. (2015) was 0.043 (±0.004). Heritabilities of R2 and R3 traits tended to be higher than those of the R1 trait, maybe related to low occurrence of phenotype “Absent” for trait R1. Gunia et al. (2018) estimated a similar heritability in rabbits for the trait resistance to nonspecific disease in the selection environment (0.04 ± 0.01). The present study was carried out in a standard breeding environment, that is, designed to minimize exposure to environmental challenges. In some studies, the animals were reared under challenging conditions, which seems to allow better phenotyping of the robustness of the animals. This may result in the estimation of higher heritabilities. Indeed, Gunia et al. (2018) estimated higher heritabilities for resistance to nonspecific disease in a challenging environment (0.08 ± 0.02) than in the standard selection environment. Under challenging conditions in rabbits, Shrestha et al. (2020) showed a heritability of the resistance to pasteurellosis of 0.16 ± 0.06. Putz et al. (2019) estimated the heritability for mortality traits for fattening pigs raised under disease challenging conditions to be 0.13 ± 0.03. The definition of this trait was close to that for R1, which had a slightly higher heritability. It is expected that challenging conditions better reveal variation in robustness (Theilgaard et al., 2007; Gunia et al., 2018). However, when choosing the selection environment, there is a need to balance between conditions that allow growth potential to be expressed and conditions that favor the expression of robustness. This is a relevant question for future selection strategies that aim to produce efficient and robust animals. Advantages and limits of robustness traits Our objective was to build proxies of robustness based on information readily available in the context of commercial pig breeding. These proxies have to meet the expectations of pig farmers, that is to say, they identify animals that were present for testing in good health, with reasonable growth rates, and with the least amount of medical injections. In this context, we decided to combine the underlying traits into scores to build the three robustness traits, rather than focusing on specific traits such as mortality or disease resilience. This choice was pragmatic because working on specific traits will multiply the number of traits to be included in the breeding goal. The advantage of using such pragmatic measures is that they can be deployed on large scale, if shown to be useful. Among the robustness traits, R2 was the trait with the highest heritability estimate in the two lines. It was highly genetically correlated (≥0.92 ± 0.06) with R3 but required less information in order to be calculated. Thus, R2 meets the objective of finding an operational trait to select in order to have live and healthy animals at the end of the period. A limit of this robustness trait is the difficulty of estimating the impact of the genetic evolution of the synthetic trait on each of its underlying traits. As such further investigation on the impact of the improvement of this robustness trait on mortality or on disease occurrence could be useful. The use of these types of additional information, not currently available in the databases, would require improved data management systems. Furthermore, estimation of the economic value in the breeding goal of such synthetic traits is an important issue. It would be interesting to estimate the economic value of genetic evolution of the tested robustness traits in order to define a weighting in the breeding goal. Berghof et al. (2019) have published an interesting approach for estimating the economic value of resilience traits based on cost reductions of labor and treatments. Improving robustness traits also meets societal expectations, in particular animal welfare, the economic value of which is difficult to quantify. Fattening group fitted as a random effect The fattening group included as a random effect in the models describes the common environmental conditions encountered by all the animals of a group entering into the station at the same date and having been raised under the same environmental conditions, including disturbances. What we call the fattening group in this article can also be more classically called the contemporary group, as described by Van Vleck (1987). The risk associated with treating the contemporary group as a random effect is to obtain biased breeding values if there is a nonrandom association between contemporary groups and sires (Visscher and Goddard, 1993). Babot et al. (2003) showed that the estimate of genetic progress could be biased when there was an environmental trend. However, considering the contemporary group as a random effect avoids a too important loss of information encountered when it is treated as a fixed effect (Visscher and Goddard, 1993). Inclusion of fattening group as a random effect with additive genetic effects was chosen as it was expected to avoid overestimating heritabilities. Binary traits: threshold vs. linear models The analysis of R1, R2, and R3 traits was carried out using a linear model, whereas they are categorical traits. Theoretically, the use of linear models to analyze categorical data is not optimal, the appropriate method being the threshold model (Gianola, 1982). However, to integrate these traits in multi-traits analysis to estimate genetic correlations and to perform a genetic evaluation, it is necessary to analyze them with a linear model to overcome convergence issues and long computing times (Kadarmideen et al., 2000). It has been shown that the linear model can be a good approximation of the threshold model under certain conditions. Meijering and Gianola (1985) showed similar heritability estimates between the two methods for binary traits, when the prevalence of the analyzed traits was between 25% and 75%. The trait R1 did not meet this condition with a prevalence of 4.8% and 5.7%, while R2 was close to the condition with a prevalence of 19.3% and 20.2%. To evaluate the consequences of applying a linear model for R1, R2, and R3 instead of a threshold model, we compared the linear and threshold models for each of these three traits analyzed separately. With the threshold model, heritabilities were estimated on the observed scale and after applying the transformation proposed by Gianola (1982). For R1, the estimates with threshold model and single-trait linear model were, respectively, 0.02 ± 0.01 and 0.02 ± 0.01 for the Pie line and 0.03 ± 0.02 and 0.06 ± 0.03 for the Pie NN line. For R2, the heritabilities from the threshold model and single-trait linear model were 0.04 ± 0.01 and 0.05 ± 0.02 in the Pie line and 0.05 ± 0.02 and 0.08 ± 0.02 in the Pie NN line. For R3, heritabilities with the threshold model and single-trait linear model were 0.04 ± 0.02 and 0.03 ± 0.02, respectively, in the Pie line, and 0.08 ± 0.04 and 0.07 ± 0.03 in the Pie NN line. For the trait R3, the correlations between estimated breeding values (EBVs) estimated from the linear model and EBVs estimated from the threshold model were 0.99 for Pie and 0.98 for Pie NN. This validates the use of equidistant levels for the three categories in R3. Thus, we found no evidence that the use of the linear model was inappropriate for analyzing R1, R2, and R3. Heritability estimates for production traits Heritability estimates for ADG and DFI were consistent with those reported in the literature for Piétrain or Large-White pigs raised in similar environmental conditions, which varied from 0.29 ± 0.02 to 0.48 ± 0.06 and from 0.31 ± 0.05 to 0.53 ± 0.06 (Saintilan et al., 2013; Gilbert et al., 2017; Déru et al., 2020; Gorssen et al., 2021). For carcass traits (BF100 and LD100), heritabilities were also consistent with the values estimated by Sourdioux et al. (2009) and Saintilan et al. (2013) in the Pietrain breed (BF100: 0.38 to 0.48; LD100: 0.25 to 0.34). Our estimates of heritability for FCR and RFI in Pie and Pie NN lines were lower, especially for Pie NN, than the heritabilities presented by Saintilan et al. (2013) and Déru et al. (2020), which varied from 0.33 ± 0.06 to 0.34 ± 0.05, and from 0.40 ± 0.06 to 0.47 ± 0.08, respectively. However, the heritability estimate for FCR in the Pie line was close to the values estimated by Gilbert et al. (2017), Putz et al. (2019), and Gorssen et al. (2021); from 0.24 ± 0.04 to 0.35 ± 0.07. For FCR and RFI traits, the lower heritabilities for the Pie NN line were related to a lower genetic variance than for Pie, respectively, 0.0054 and 0.0104 for FCR, and 3,686 and 6,667 for RFI. Heritability estimates for ABC index For the trait ABC, the heritability for the Pie NN line was consistent with that published by Revilla et al. (2022; 0.03 ± 0.016), but we found a slightly higher heritability in the Pie line (0.09 ± 0.03 vs. 0.04 ± 0.01). This difference is the result of a lower phenotypic variance in both lines compared with those reported by Revilla et al. (2022). In the present study, the data were recorded during a different time period and an improved outlier detection procedure was used on the observations collected by AFS, which reduced the contribution of erroneous measures to the phenotypic variance, and consequently reduced the residual variance when estimating variance parameters. Genetic correlations between robustness and production traits The two growth traits (IBW and ADG) were moderate to strongly correlated with R2 and R3. The correlations with IBW showed that growth during postweaning, that is, pretest period, had an impact on the robustness scores evaluated during the fattening period. In this context, Putz et al. (2019) showed that the genetic correlation of ADG with mortality was close to 0 while the genetic correlation with the number of antibiotic treatments was favorable and strong (from −0.68 ± 0.42 to −0.70 ± 0.13). It seems that the growth of less robust animals is more impacted by environmental perturbations. It is also important to take into account that growth has been a major selection trait in both breeds for over 20 yr, and lack of growth or weak body development were major causes of culling at testing. In this situation, an animal’s ability to be robust is strongly linked to its ability to express optimal growth regardless of the environment. Nonetheless, even if the correlation is strong, it is different from 1, which implies that the traits R2 and R3 add an additional information regarding the robustness of the animal compared with growth traits. Thus, if the selection is made using these traits, they would allow us to improve animals’ robustness more than if the selection is made only on growth traits. There was a moderate and unfavorable relationship between the robustness traits and the FCR, although the precision of the estimates remains low. This could be related to the positive correlation between ADG and FCR, which was affected by the way these two traits were estimated. They were measured over an identical time period for all individuals but were not standardized between starting and finishing weights (ADG 30 to 110 kg). Accordingly, some of the animals tested reached their mature weight before the end of the testing period, which led to a drop in feed conversion even if they had a previously strong growth. Within these two pig populations, there were two different types of animals with low FCR: those which had a strong growth but did not approach their mature weight during the testing period, and those with a low DFI associated with low growth. In parallel, the genetic correlations of R2 and R3 with DFI were strong. This could indicate that the most robust animals during the fattening period are not the most efficient because they allocate a part of nutrients to nonproductive functions. This antagonism between short-term efficiency and robustness had been put forward by Friggens et al. (2017). Genetic correlations between robustness and BF100 were slightly unfavorable, with low precision, particularly in the Pie line. We can suppose that the capacity to be robust could be associated with more important body reserves allowing the animal to face perturbations. The genetic correlations between the robustness traits and the RFI were close to 0 or slightly unfavorable in the Pie and Pie NN lines. For the relation between RFI and robustness, it is hypothesized that selection for low RFI may limit the animals’ ability to allocate nutrients to resilience functions for dealing with perturbations (Gilbert et al., 2017). In contrast, several studies have shown, through divergent selection experiments on RFI, that there can be favorable effects of lines with low RFI on sensitivity to the PRRS virus (Dunkelberger et al., 2015) or on the risk of being culled between 70 d of age and slaughter (Gilbert et al., 2017). Genetic correlations of robustness traits with the ABC were close to zero and difficult to interpret, due to low precision. The trait based on a dynamic analysis of the evolution of the weight (ABC) approach was relatively independent of the criteria created from the static data (R1, R2, and R3). In view of the strong or moderate link between the robustness criteria, the DFI and the FCR, it would be interesting to investigate the link between the dynamics of ingestion or allocation of animals and their ability to cope with disturbances, that is, their robustness. The robustness traits that we proposed are built on single measurements represented the effects of the accumulations of good or bad events during the measured period (Friggens et al., 2017). A dynamic analysis of the data collected by the automatic feeders would make it possible to have an analysis of this accumulation that is dynamic and probably better able to identify the finer criteria of robustness. Conclusion This study showed that it is possible to set up a selection based on robustness in growing pigs from robustness scores (R2 and R3) calculated from data available routinely on farms. However, the low heritabilities offer limited hope for rapid genetic improvement. The trait R2 would seem the most interesting because it is more heritable and requires less information to be calculated. The introduction of the R2 trait in the breeding goal of paternal lines is relevant but would require further investigation with respect to the potential genetic gain achievable in a multi-trait breeding goal. At this stage, the trait R3 is less relevant, but its determination could be upgraded by adding additional information on the various other assistance provided by the breeder, to identify animals that have the ability to express or adapt their production potential without help. In this study, we focused on the evaluation of robustness over a short period of the animal’s life but it is necessary to investigate such traits over the whole lifespan. Abbreviations Abbreviations ABC area between curves ADG average daily growth AFS automatic feeding system AMW average metabolic weight BF backfat thickness BF100 backfat thickness estimated at 100 kg live weight BW body weight DFI daily feed intake FCR feed conversion ratio FI feed intake IBW initial body weight LD longissimus dorsi thickness LD100 longissimus dorsi thickness estimated at 100 kg live weight PDFI potential average daily feed intake Pie Piétrain Français Pie NN Piétrain NN Français free from halothane-sensitivity RFI residual feed intake TBW body weight at individual testing Acknowledgments We acknowledge Rafael Muñoz-Tamayo for his constructive discussion on this project. We also acknowledge the technicians of AXIOM’s testing station at La Garenne for their implication for data collection. We also thank the two anonymous reviewers for their critical review and helpful comments. This project was supported by ANRT (Association Nationale Recherche Technologie) with a CIFRE Doctoral fellowship (2019/0705). Conflict of interest statement G.L. and L.F.-G. were employed by the company AXIOM. There is no conflict of interest to be declared. LITERATURE CITED Babot , D. , J. L. Noguera, L. Alfonso, and J. Estany. 2003 . Fixed or random contemporary groups in genetic evaluation for litter size in pigs using a single trait repeatability animal model . J. Anim. Breed. Genet . 120 : 12 – 22 . doi:10.1046/j.1439-0388.2003.00372.x Google Scholar Crossref Search ADS WorldCat Berghof , T. V. L. , M. Poppe, and H. A. Mulder. 2019 . Opportunities to improve resilience in animal breeding programs . Front. Genet. 9 : 692 . doi:10.3389/fgene.2018.00692. Available from https://www.frontiersin.org/articles/10.3389/fgene.2018.00692/full Google Scholar Crossref Search ADS PubMed WorldCat Déru , V. , A. Bouquet, C. Hassenfratz, B. Blanchet, C. Carillier-Jacquin, and H. Gilbert. 2020 . Impact of a high-fibre diet on genetic parameters of production traits in growing pigs . Animal 14 : 2236 – 2245 . doi:10.1017/S1751731120001275 Google Scholar Crossref Search ADS PubMed WorldCat Dunkelberger , J. R. , N. J. Boddicker, N. V. L. Serão, J. M. Young, R. R. R. Rowland, and J. C. M. Dekkers. 2015 . Response of pigs divergently selected for residual feed intake to experimental infection with the PRRS virus . Livest. Sci . 177 : 132 – 141 . doi:10.1016/j.livsci.2015.04.014 Google Scholar Crossref Search ADS WorldCat Friggens , N. C. , F. Blanc, D. P. Berry, and L. Puillet. 2017 . Review: deciphering animal robustness. A synthesis to facilitate its use in livestock breeding and management . Animal 11 : 2237 – 2251 . doi:10.1017/S175173111700088X Google Scholar Crossref Search ADS PubMed WorldCat Gianola , D. 1982 . Theory and analysis of threshold characters . J. Anim. Sci . 54 : 1079 – 1096 . doi:10.2527/jas1982.5451079x Google Scholar Crossref Search ADS WorldCat Gilbert , H. , Y. Billon, L. Brossard, J. Faure, P. Gatellier, F. Gondret, E. Labussière, B. Lebret, L. Lefaucheur, N. Le Floch, et al. . 2017 . Review: divergent selection for residual feed intake in the growing pig . Animal 11 : 1427 – 1439 . doi:10.1017/S175173111600286X Google Scholar Crossref Search ADS PubMed WorldCat Gilmour , A. R. , B. J. Gogel, B. R. Cullis, and R. Thompson. 2009 . ASREML user guide release 3.0; 310 pp . Available from https://www.vsni.co.uk/downloads/asreml/release3/UserGuide.pdf Gorssen , W. , D. Maes, R. Meyermans, J. Depuydt, S. Janssens, and N. Buys. 2021 . High heritabilities for antibiotic usage show potential to breed for disease resistance in finishing pigs . Antibiotics 10 : 829 . doi:10.3390/antibiotics10070829 Google Scholar Crossref Search ADS PubMed WorldCat Gunia , M. , I. David, J. Hurtaud, M. Maupin, H. Gilbert, and H. Garreau. 2015 . Resistance to infectious diseases is a heritable trait in rabbits1 . J. Anim. Sci . 93 : 5631 – 5638 . doi:10.2527/jas.2015-9377 Google Scholar Crossref Search ADS PubMed WorldCat Gunia , M. , I. David, J. Hurtaud, M. Maupin, H. Gilbert, and H. Garreau. 2018 . Genetic parameters for resistance to non-specific diseases and production traits measured in challenging and selection environments; application to a rabbit case . Front. Genet . 9 : 467 . doi:10.3389/fgene.2018.00467 Google Scholar Crossref Search ADS PubMed WorldCat Hansen , J. , M. Sato, and R. Ruedy. 2012 . Perception of climate change . Proc. Natl. Acad. Sci . 109 : E2415 . doi:10.1073/pnas.1205276109 Google Scholar OpenURL Placeholder Text WorldCat Hémonic , A. , A. Poissonnet, C. Chauvin, and I. Corrégé. 2019 . Evolution des usages d’antibiotiques dans les élevages de porcs en France entre 2010 et 2016 au travers des panels INAPORC. In: 51st Journées Rech. Porcine Fr. Paris, France. p. 277–282. Institut Technique du Porc . 2004 . La sélection porcine collective en France . Paris (France) : ITP . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kadarmideen , H. N. , R. Thompson, and G. Simm. 2000 . Linear and threshold model genetic parameters for disease, fertility and milk production in dairy cattle . Anim. Sci . 71 : 411 – 419 . doi:10.1017/S1357729800055338 Google Scholar Crossref Search ADS WorldCat Knap , P. W. 2005 . Breeding robust pigs . Aust. J. Exp. Agric . 45 : 763 – 773 . doi:10.1071/EA05041 Google Scholar Crossref Search ADS WorldCat Knap , P. W. , and A. Doeschl-Wilson. 2020 . Why breed disease-resilient livestock, and how? Genet. Sel. Evol . 52 : 60 . doi:10.1186/s12711-020-00580-4 Google Scholar Crossref Search ADS PubMed WorldCat Labroue , F. , L. Maignel, P. Sellier, and J. Noblet. 1999 . Consommation résiduelle chez le porc en croissance alimenté à volonté Méthode de calcul et variabilité génétique. In: 31st Journées Rech. Porcine Fr. Paris, France. p. 167–174. Meijering , A. , and D. Gianola. 1985 . Linear versus nonlinear methods of sire evaluation for categorical traits: a simulation study . Génét. Sél. Évol . 17 : 115 – 132 . doi:10.1186/1297-9686-17-1-115 Google Scholar Crossref Search ADS WorldCat Noblet , J. , C. Karège, and S. Dubois. 1991 . Influence of growth potential on energy requirements for maintenance in growing pigs. In: Wenk, C., and M. Boessinger, editors. Energy metabolism in farm animals . Zurich, Switzerland: ETH-Zentrum; p. 107–110 (EAAP Pub. No. 58). Available from https://hal.inrae.fr/hal-02849658 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Perez , B. C. , J. Shaddick, S. A. S. Van Der Klein, K. Alves, M. C. A. M. Bink, A. Gueret, and O. W. Willems. 2021 . Genetic parameters for liveability traits in turkeys. In: 72nd Annual Meeting of the European Federation of Animal Science. Davos, Switzerland ; p. 265 . Pérez-Cabal , M. A. , and N. Charfeddine. 2015 . Models for genetic evaluations of claw health traits in Spanish dairy cattle . J. Dairy Sci . 98 : 8186 – 8194 . doi:10.3168/jds.2015-9562 Google Scholar Crossref Search ADS PubMed WorldCat Phocas , F. , C. Belloc, J. Bidanel, L. Delaby, J. Y. Dourmad, B. Dumont, P. Ezanno, L. Fortun-Lamothe, G. Foucras, B. Frappat, et al. . 2016 . Review: towards the agroecological management of ruminants, pigs and poultry through the development of sustainable breeding programmes. II. Breeding strategies . Animal 10 : 1760 – 1769 . doi:10.1017/S1751731116001051 Google Scholar Crossref Search ADS PubMed WorldCat Putz , A. M. , J. C. S. Harding, M. K. Dyck, F. Fortin, G. S. Plastow, J. C. M. Dekkers, and PigGen Canada. 2019 . Novel resilience phenotypes using feed intake data from a natural disease challenge model in wean-to-finish pigs . Front. Genet . 9 : 660 . doi:10.3389/fgene.2018.00660 Google Scholar Crossref Search ADS PubMed WorldCat R Core Team . 2018 . R: a language and environment for statistical computing . Vienna (Austria) : R Foundation for Statistical Computing . Available from https://www.R-project.org/ Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Revilla , M. , G. Lenoir, L. Flatres-Grall, R. Muñoz-Tamayo, and N. C. Friggens. 2022 . Quantifying growth perturbations over the fattening period in swine via mathematical modelling . Peer Community J . 2 :1–16. doi:10.24072/pcjournal.82 Available from https://peercommunityjournal.org/articles/10.24072/pcjournal.82/ Google Scholar OpenURL Placeholder Text WorldCat Saintilan , R. , I. Mérour, L. Brossard, T. Tribout, J. Y. Dourmad, P. Sellier, J. Bidanel, J. van Milgen, and H. Gilbert. 2013 . Genetics of residual feed intake in growing pigs: relationships with production traits, and nitrogen and phosphorus excretion traits1 . J. Anim. Sci . 91 : 2542 – 2554 . doi:10.2527/jas.2012-5687 Google Scholar Crossref Search ADS PubMed WorldCat Shrestha , M. , H. Garreau, E. Balmisse, B. Bed’hom, I. David, E. Guitton, E. Helloin, G. Lenoir, M. Maupin, R. Robert, et al. . 2020 . Genetic parameters of resistance to pasteurellosis using novel response traits in rabbits . Genet. Sel. Evol . 52 : 34 . doi:10.1186/s12711-020-00552-8 Google Scholar Crossref Search ADS PubMed WorldCat Sourdioux , M. , G. Lenoir, L. Guery, D. Bahon, T. Tribout, and J. P. Bidanel. 2009 . Estimation des paramètres génétiques pour des critères de croissance et carcasse en race Piétrain et en lignée composite Piétrain négative halothane. In: 41st Journées Rech. Porcine Fr. Paris, France. p. 25–26. Theilgaard , P. , J. P. Sánchez, J. J. Pascual, P. Berg, N. C. Friggens, and M. Baselga. 2007 . Late reproductive senescence in a rabbit line hyper selected for reproductive longevity, and its association with body reserves . Genet. Sel. Evol . 39 : 207 . doi:10.1186/1297-9686-39-2-207 Google Scholar Crossref Search ADS PubMed WorldCat Van Vleck , L. D. 1987 . Contemporary groups for genetic evaluations . J. Dairy Sci . 70 : 2456 – 2464 . doi:10.3168/jds.S0022-0302(87)80309-0 Google Scholar Crossref Search ADS PubMed WorldCat Varona , L. , I. Misztal, and J. K. Bertrand. 1999 . Threshold-linear versus linear-linear analysis of birth weight and calving ease using an animal model: II. Comparison of models . J. Anim. Sci . 77 : 2003 – 2007 . doi:10.2527/1999.7782003x Google Scholar Crossref Search ADS PubMed WorldCat Visscher , P. M. , and M. E. Goddard. 1993 . Fixed and random contemporary groups . J. Dairy Sci . 76 : 1444 – 1454 . doi:10.3168/jds.S0022-0302(93)77475-5 Google Scholar Crossref Search ADS WorldCat © The Author(s) 2022. Published by Oxford University Press on behalf of the American Society of Animal Science. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact [email protected] © The Author(s) 2022. Published by Oxford University Press on behalf of the American Society of Animal Science.

Xiao, Ping; Goodarzi, Parniyan; Pezeshki, Adel; Hagen, Darren E

doi: 10.1093/jas/skac156pmid: 35552417

Low birth weight (LBW) is associated with metabolic disorders in early life. While dietary l-tryptophan (Trp) can ameliorate postprandial plasma triglycerides (TG) disposal in LBW piglets, the genetic and biological basis underlying Trp-caused alterations in lipid metabolism is poorly understood. In this study, we collected 24 liver samples from 1-mo-old LBW and normal birth weight (NBW) piglets supplemented with different concentrations of dietary Trp (NBW with 0% Trp, N0; LBW with 0% Trp, L0; LBW with 0.4% Trp, L4; LBW with 0.8% Trp, L8; N = 6 in each group.) and conducted systematic, transcriptome-wide analysis using RNA sequencing (RNA-seq). We identified 39 differentially expressed genes (DEG) between N0 and L0, and genes within “increased dose effect” clusters based on dose-series expression profile analysis, enriched in fatty acid response of gene ontology (GO) biological process (BP). We then identified RNA-binding proteins including SRSF1, DAZAP1, PUM2, PCBP3, IGF2BP2, and IGF2BP3 significantly (P < 0.05) enriched in alternative splicing events (ASE) in comparison with L0 as control. There were significant positive and negative relationships between candidate genes from co-expression networks (including PID1, ANKRD44, RUSC1, and CYP2J34) and postprandial plasma TG concentration. Further, we determined whether these candidate hub genes were also significantly associated with metabolic and cardiovascular traits in humans via human phenome-wide association study (Phe-WAS), and analysis of mammalian orthologs suggests a functional conservation between human and pig. Our work demonstrates that transcriptomic changes during dietary Trp supplementation in LBW piglets. We detected candidate genes and related BP that may play roles on lipid metabolism restoration. These findings will help to better understand the amino acid support in LBW metabolic complications.

Kim, Eunjoo; Kyoung, Hyunjin; Hyung Koh, Nae; Lee, Hanbae; Lee, Seonmin; Kim, Yonghee; Il Park, Kyeong; Min Heo, Jung; Song, Minho

doi: 10.1093/jas/skac122pmid: 35404458

The present study investigated the effects of live yeast cultures (LYC) on growth performance, gut health indicators, and immune responses in broiler chickens. A total of 720 mixed-sex broilers (40 birds/pen; 9 replicates/treatment) were randomly allocated to two dietary treatments: (1) a basal diet based on corn–soybean meal (CON) and (2) CON with 1 g/kg LYC. At 35 d of age, one bird per replicate pen was chosen for biopsy. LYC group tended (P < 0.10) to increase average daily gain during the grower phase compared with CON group. Broilers fed LYC diet had increased (P = 0.046) duodenal villus height and area but reduced (P = 0.003) duodenal crypt depth compared with those fed CON diet. Birds fed LYC diet presented alleviated (P < 0.05) serum TNF-α, IL-1β, and IL-6 levels compared with those fed CON diet. Further, birds fed LYC diet exhibited upregulated (P < 0.05) ileal tight junction-related proteins and pro-inflammatory cytokines in the ileal tissue compared with those fed CON diet. Inverse Simpson’s diversity (P = 0.038) revealed that birds fed CON diet had a more diverse microbiota community in the ileal digesta, compared with those fed LYC diet, while no significant difference between the treatments on Chao1 and Shannon’s indices was observed. Based on the weighted UniFrac distance, the PCoA showed that microbiota in the ileal digesta of the LYC group was different from that of the CON group. LYC group increased the abundance of the phyla Firmicutes and genera Lactobacillus, Prevotella, and Enterococcus compared with CON group. The present study demonstrated that supplemental LYC as a feed additive provide supportive effects on enhancing gut functionality by improving the upper intestinal morphology and gut integrity, and modulating the immune system and microbiota communities of birds.

Goulart, Rodrigo S; Tedeschi, Luis O; Silva, Saulo L; Leme, Paulo R; de Alencar, Maurício M; Lanna, Dante P D

doi: 10.1093/jas/skac125pmid: 35417561

This study determined the energy requirement for maintenance of purebred Nellore cattle and its crossbreds using data from a comparative slaughter trial in which animals were raised under the same plane of nutrition from birth through slaughter and born from a single commercial Nellore cowherd. A total of 79 castrated steers (361 ± 54 kg initial body weight [BW]) were used in a completely randomized design by age (22 mo ± 23 d of age) with four genetic groups (GG): Nellore (NL), ½ Angus × ½ Nellore (AN), ½ Canchim × ½ Nellore (CN), and ½ Simmental × ½ Nellore (SN). The experimental design provided ranges in metabolizable energy (ME) intake (MEI), BW, and average daily gain needed to develop regression equations to predict net energy for maintenance (NEm) requirements. Four steers of each GG were slaughtered to determine the initial body composition. The remaining 63 steers were assigned to different nutritional treatments (NT) by GG; ad libitum or limit-fed treatments (receiving 70% of the daily feed of the ad libitum treatment of the same GG). Full BW was recorded at birth, weaning, 12, 18, and 22 mo. In the feedlot, steers were fed for 101 d a diet containing (DM basis) 60% corn silage and 40% concentrate. No difference in age at weaning (P = 0.534) and slaughter (P = 0.179 and P = 0.896, for GG and NT, respectively) were observed. AN steers were heavier at weaning weight, yearling weight and had higher empty BW (EBW; P = 0.007, P = 0.014, and P < 0.001, respectively) in comparison to NL, CN, and SN. There were no interactions (P > 0.05) between GG and NT for any variable evaluated. When fed ad libitum, AN steers had higher daily MEI (Mcal/d; P < 0.001) in comparison to NL, CN, and SN. On a constant age basis, differences were observed on body composition (P < 0.05) between GG. The slope (P = 0.600) and intercept (P = 0.702) of the regression of log heat production on MEI were similar among GG. Evaluating at the same age and the same frame size, there were no differences in NEm requirement between Nellore and AN (P = 0.528), CN (P = 0.671), and SN (P = 0.706). The combined data indicated a NEm requirement of 86.8 kcal/d/kg0.75 EBW and a ME required for maintenance requirement had a common value of 137.53 kcal/d/kg0.75 EBW. The efficiency of energy utilization for maintenance and the efficiency of energy utilization for growth values were similar among GG (P > 0.05 and P > 0.05, respectively) and were on average 63.2% and 26.0%, respectively. However, although not statistically different, the NEm values from NL showed a decrease in NEm of 5.76% compared with AN steers.