Advertisement

Phenotypic and genetic analysis of milk and serum element concentrations in dairy cows

Open AccessPublished:October 03, 2019DOI:https://doi.org/10.3168/jds.2019-16960

      ABSTRACT

      Enhancing micronutrient (i.e., mineral and vitamin) concentrations within milk and serum from dairy cows is important for both the health of the cow and the nutritive value of the milk for human consumption. However, a good understanding of the genetics underlying the micronutrient content in dairy cattle is needed to facilitate such enhancements through feeding or breeding practices. In this study, milk (n = 950) and serum (n = 766) samples were collected from Holstein-Friesian dairy cows (n = 479) on 19 occasions over a 59-mo period and analyzed for concentrations of important elements. Additionally, a subset of 256 milk samples was analyzed for concentrations of vitamin B12. Cows belonged to 2 genetic lines (average and highest genetic merit for milk fat plus protein yield) and were assigned to 1 of 2 diets based on either a by-product or homegrown ration. Univariate models accounting for repeated records were used to analyze element and vitamin B12 data and investigate the effect of genotype and feeding system as well as derive estimates of variance components and genetic parameters. Bivariate models were used to study correlations both within and between milk and serum. Only concentrations of Hg in milk were seen to be affected by genotype, with higher concentrations in cows with high genetic merit. In contrast, element concentrations were influenced by feeding system such that cows fed the homegrown diet had increased milk concentrations of Ca, Cu, I, Mn, Mo, P, and K and increased serum concentrations of Cd, Cu, Fe, Mo, and V. Cows on the by-product diet had increased milk concentrations of Mg, Se, and Na and increased serum concentrations of P and Se. Heritability (h2) estimates were obtained for 6 milk and 4 serum elements, including Mg (h2milk = 0.30), K (h2serum = 0.18), Ca (h2milk = 0.20; h2serum = 0.12), Mn (h2milk = 0.14), Cu (h2serum = 0.22), Zn (h2milk = 0.24), Se (h2milk = 0.15; h2serum = 0.10), and Mo (h2milk = 0.19). Significant estimates of repeatability were observed in all milk and serum quantity elements (Na, Mg, P, K, and Ca) as well as 5 milk and 7 serum trace elements. Only K in milk and serum was found to have a significant positive genetic and phenotypic correlation (0.52 and 0.22, respectively). Significant phenotypic associations were noted between milk and serum Ca (0.17), Mo (0.19), and Na (−0.79). Additional multivariate analyses between measures within sample type (i.e., milk or serum) revealed significant positive associations, both phenotypic and genetic, between some of the elements. In milk, Se was genetically correlated with Ca (0.63), Mg (0.59), Mn (0.40), P (0.53), and Zn (0.52), whereas in serum, V showed strong genetic associations with Cd (0.71), Ca (0.53), Mn (0.63), Mo (0.57), P (0.42), K (0.45), and Hg (−0.44). These results provide evidence that element concentrations in milk and blood of dairy cows are significantly influenced by both diet and genetics and demonstrate the potential for genetic selection and dietary manipulation to alter nutrient concentration to improve both cow health and the healthfulness of milk for human consumption.

      Key words

      INTRODUCTION

      Micronutrients are required throughout life and consist of vitamins and minerals that are essential for maintaining normal body function and health in humans and other animals (
      • FAO/WHO (Food and Agriculture Organization of the United Nations/World Health Organization)
      Vitamin and Mineral Requirements in Human Nutrition.
      ;
      • Gernand A.D.
      • Schulze K.J.
      • Stewart C.P.
      • West K.P.
      • Christian P.
      Micronutrient deficiencies in pregnancy worldwide: Health effects and prevention.
      ). Neither humans nor other animals can synthesize micronutrients within the body; therefore, micronutrients must be obtained from the diet. Suboptimal intakes of these vitamins and minerals can affect normal growth and development, reducing performance as well as increasing susceptibility to, for example, Keshan disease in humans and white muscle disease in cattle due to Se deficiency (
      • Muth O.H.
      • Oldfield J.E.
      • Remmert L.F.
      • Schubert J.R.
      Effects of selenium and vitamin E on white muscle disease.
      ;
      • Yu W.H.
      A study of nutritional and bio-geochemical factors in the occurrence and development of Keshan disease.
      ).
      Whereas vitamins are organic molecules, minerals such as P, Ca, Fe, Zn, Se, and I are inorganic but are required only in very small amounts. Minerals can be further classified into trace elements (e.g., Fe, Zn, Se, and I) that are required in low amounts and quantity elements (e.g., Mg, P, K, and Ca) that are required in larger amounts. When intakes of quantity or trace elements or vitamins from the diet are insufficient, deficiencies can arise that can compromise animal and human health and increase the risk of disease. Indeed, concern is currently growing that a sizable portion of the human population does not meet micronutrient reference nutrient intake (RNI) values (i.e., the amount of nutrient required to prevent deficiency;
      • Rooke J.
      • Flockhart J.
      • Sparks N.
      The potential for increasing the concentrations of micro-nutrients relevant to human nutrition in meat, milk and eggs.
      ;
      • Givens D.I.
      • Livingstone K.M.
      • Pickering J.E.
      • Fekete A.
      • Dougkas A.
      • Elwood P.C.
      Milk: White elixir or white poison? An examination of the associations between dairy consumption and disease in human subjects.
      ).
      Dairy products, such as milk and cheese, are important sources of minerals and vitamins and contribute substantially to dietary intakes of Ca, P, I, Zn, and Mg (60, 60, 55, 18, and 10% of RNI, respectively) as well as vitamins A, B2, and B12 (26, 52, and 150% of RNI, respectively;
      • Kliem K.E.
      • Givens D.I.
      Dairy products in the food chain: Their impact on health.
      ). Importantly, Mg and Ca are increasingly significant factors in bone development, especially in children (
      • Givens D.I.
      • Livingstone K.M.
      • Pickering J.E.
      • Fekete A.
      • Dougkas A.
      • Elwood P.C.
      Milk: White elixir or white poison? An examination of the associations between dairy consumption and disease in human subjects.
      ). Furthermore, the circulating concentrations of these minerals and vitamins in the blood and milk of dairy cows likely relate to the fitness of the animal given their important roles in numerous physiological and immunological processes (
      • Percival S.S.
      Copper and immunity.
      ;
      • Doherty C.P.
      Host-pathogen interactions: The role of iron.
      ;
      • Maggini S.
      • Wintergerst E.S.
      • Beveridge S.
      • Hornig D.H.
      Selected vitamins and trace elements support immune function by strengthening epithelial barriers and cellular and humoral immune responses.
      ;
      • Hoffmann P.R.
      • Berry M.J.
      The influence of selenium on immune responses.
      ;
      • Prasad A.S.
      Zinc in human health: Effect of zinc on immune cells.
      ;
      • Alpert P.T.
      The role of vitamins and minerals on the immune system.
      ). Therefore, identifying breeding strategies within the cow to increase milk micronutrient concentrations as well as optimizing micronutrient concentrations within the cow herself should be of ultimate benefit to both the cow and the human dairy product consumer. Moreover, because heavy metals such as Pb, Hg, and Cd, which have potential adverse effects on health, can also be found in milk (
      • Rey-Crespo F.
      • Miranda M.
      • López-Alonso M.
      Essential trace and toxic element concentrations in organic and conventional milk in NW Spain.
      ), breeding strategies should also be commensurate with reducing, or at least not increasing, concentrations of these metals where possible. Dietary manipulation of mineral concentrations in livestock has been demonstrated, yet there is a relative lack of knowledge concerning the contribution of cow genetics to variation in concentrations of elements (including heavy metals) and vitamins within the blood and milk of dairy cows (
      • Rooke J.
      • Flockhart J.
      • Sparks N.
      The potential for increasing the concentrations of micro-nutrients relevant to human nutrition in meat, milk and eggs.
      ).
      The aim of this study was to carry out a phenotypic and genetic analysis of mineral, vitamin, and heavy metal concentrations in dairy cow milk and serum to determine (1) the effect of genotype and diet on individual element and vitamin B12 concentrations, (2) whether relationships exist between element concentrations (including vitamin B12) in milk and serum, and (3) whether variation between animals exists that would permit selection for optimized element and vitamin B12 concentrations that would benefit the health of both the cow and the human consumer.

      MATERIALS AND METHODS

      Animals

      Animals involved in this study were from the Langhill pedigree herd of Holstein-Friesian dairy cows (n = 479) housed at the Scotland's Rural College Dairy Research Centre in Dumfries, Scotland, between 2012 and 2016. All cows were part of a long-term (ongoing) selection experiment for genotype × environment following a 2 × 2 approach (
      • Veerkamp R.F.
      • Simm G.
      • Oldham J.D.
      Effects of interaction between genotype and feeding system on milk production, feed intake, efficiency and body tissue mobilization in dairy cows.
      ). Briefly, the herd has been divided equally between 2 distinct genetic lines (control and select) selected since 1970 and assigned to 1 of 2 diets based on differing rations. The control line has been bred to bulls of UK national average genetic merit for kilograms of fat plus protein yield. In contrast, the high genetic merit select line (top 5% genetic merit) has been bred from bulls with the highest genetic merit for kilograms of fat plus protein yield. The 2 diet groups consist of a low-forage, high-energy ration based on by-products and minimal land use, simulating high-input commercial systems, and a high-forage, lower-energy ration based on homegrown components and using the maximum amount of land available, thus simulating low-input grazing systems (
      • Pryce J.E.
      • Nielsen B.L.
      • Veerkamp R.F.
      • Simm G.
      Genotype and feeding system effects and interactions for health and fertility traits in dairy cattle.
      ;
      • Roberts D.J.
      • March M.D.
      Feeding systems for dairy cows: Homegrown versus by-products feeds.
      ).
      The homegrown (HG) ration consisted of components grown exclusively on farm and included grazed grass, grass silage, red clover silage, forage maize, lucerne silage, crimped wheat, and beans. Additionally, the HG ration was balanced with purchased minerals. In contrast, the by-product (BP) ration consisted of biscuit meal, sugar beet pulp, chopped straw, breakfast cereal, wheat distillers dark grains, soybean meal (Hipro 50%, Tarff Valley Ltd., Ringford, UK), Vitagold (KW Feeds, Ayrshire, UK), protected fat (Megalac, Volac International Ltd., Hertfordshire, UK), molasses, and minerals. Mineral compositions of the HG and BP diets are presented in Supplemental Table S1 (https://doi.org/10.3168/jds.2019-16960). Target milk yields of select cows on the low- and high-energy diets are 7,500 and 13,000 L/lactation, respectively; the UK average per cow per lactation is approximately 7,557 ().

      Ethics Statement

      Blood samples were collected in accordance with UK Home Office regulations (PPL no. 60/4278 Dairy Systems, Environment and Nutrition), and procedures were approved by the Scotland's Rural College Animal Experimentation Committee. The study was restricted to routine on-farm observations and measurements that did not inconvenience or stress the animals.

      Sampling Protocol

      Samples used in the present study were collected across several years and seasons from the same ongoing experimental system; 385 (of 479) cows had 2 or more samples. Furthermore, samples were selected such that they accounted for genotype and management of cows to give a balanced representation of the herd. In total, 950 milk samples and 766 serum samples were collected for analysis of element and vitamin B12 (milk only) concentrations. Further information regarding sample collection is presented in Supplemental Table S2 (https://doi.org/10.3168/jds.2019-16960).

      Milk Samples

      Cows in the Langhill herd are milked 3 times daily (morning, midday, and evening), and for the present study milk samples were taken from the morning milking (at the same time as any blood sampling). Milk samples were collected on 16 separate occasions between June 2012 and January 2015 and included summer and winter periods. All milk samples were whole milk except for 256 samples that were from skim milk collected as part of a previous project (
      • Denholm S.J.
      • McNeilly T.N.
      • Banos G.
      • Coffey M.P.
      • Russell G.C.
      • Bagnall A.
      • Mitchell M.C.
      • Wall E.
      Estimating genetic and phenotypic parameters of cellular immune-associated traits in dairy cows.
      ,
      • Denholm S.J.
      • McNeilly T.N.
      • Banos G.
      • Coffey M.P.
      • Russell G.C.
      • Bagnall A.
      • Mitchell M.C.
      • Wall E.
      Immune-associated traits measured in milk of Holstein-Friesian cows as proxies for blood serum measurements.
      ). For these latter samples, milk was first centrifuged at 3,000 × g for 30 min at 4°C and the skim milk fraction was retained from below the fat layer using a fine-tipped pastette. All samples were stored at −20°C before analysis.

      Blood Samples

      Whole-blood samples were collected on 13 separate occasions between April 2013 and May 2016 and included summer and winter periods. Samples were collected into plain Vacutainers (BD, Reading, UK); blood was allowed to coagulate before centrifugation at 2,000 × g for 10 min at 18°C (using a refrigerated centrifuge), and the serum was retained. All samples were stored at −20°C before analysis.

      Analysis of Quantity and Trace Element and Heavy Metal Concentrations

      All milk and serum samples were analyzed, and concentrations of circulating quantity elements (Na, Mg, P, K, and Ca), trace elements (V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Se, Mo, and I), and heavy metals (Cd, Hg, and Pb) were recorded. Milk samples (1.0 mL) were digested in nitric acid (8.0 mL; 65% vol/vol) using the MARS 6 digestion system (CEM Corp., Matthews, NC) and then stored overnight at room temperature. The temperature of the samples was increased from room temperature to 210°C and then held for 10 min before cooling.
      Serum samples (50 μL) were added to hydrogen peroxide (10 μL; 30% wt/wt) and nitric acid (40 μL; 65% vol/vol) and then digested at 85°C for 40 min. Digested samples were diluted in decomposition matrix before inductively coupled plasma MS (ICP-MS) analysis. The decomposition matrix was nitric acid (2% vol/vol) and hydrochloric acid (0.5% vol/vol) in distilled deionized water (Millipore UK, Watford, UK), which was used for preparation of all solutions.
      The measured isotopes analyzed by ICP-MS were 23Na, 24Mg, 31P, 39K, 44Ca, 51V, 52Cr, 55Mn, 56Fe, 59Co, 60Ni, 63Cu, 66Zn, 78Se, 95Mo, 127I, 111Cd, 202Hg, and 208Pb. All element standards were used in stock solutions of 1,000 mg/L, which served for preparation of calibration solutions and internal standard solution. The ICP-MS measurements were carried out using an Agilent 7700X spectrometer (Agilent Technologies UK, Wokingham, UK) equipped with a MicroMist nebulizer and nickel sampler and skimmer cones. The flow of mineral standards (National Institute of Standards and Technology, Gaithersburg, MD) and samples was joined together with a flow of erbium internal standard solution (1 mg/L). The mixed flow (~500 μL/min) was delivered by the peristaltic pump to the nebulizer of the ICP-MS setup. The duration of the ICP-MS analysis was 3.0 min. Data acquisition was 1 point, 5 replicates, 100 sweeps per replicate. Milk and mussel reference materials were obtained from LGC Group (Middlesex, UK).
      For the quantification of I in milk samples, these were first digested at 90°C in 5% tetramethylammonium hydroxide (TMAH) for 3 h and then cooled. The TMAH (≥97%; Sigma-Aldrich, Dorset, UK) was diluted to 5% using ultrapure water (18.2 MΩ × cm; Elga, Pure-Flex, Cheshire, UK). The I content in the milk samples was then determined by ICP-MS (7700X, Agilent Technologies) in standard analysis mode using external calibration. The stock standard solution was gravimetrically prepared in-house from high-purity potassium iodide (+99.99%; Thermo Fisher Scientific, Altrincham, UK) in 5% TMAH. The calibration standards were prepared by serial dilution of this stock using 5% TMAH, and a tellurium internal standard was added at the same level as in the samples to a final concentration of 150 ng/mL. The method accuracy was monitored using ERM-BD150 skim milk powder certified reference material (LGC Standards, Middlesex, UK) with a certified I content of 1,730 ± 140 μg/kg (dry weight basis).

      Analysis of Vitamin B12

      Milk samples collected between April 2013 and January 2015 (11 sample points yielding 256 samples; n = 64) were analyzed for vitamin B12 concentrations. Vitamin B12 in undiluted milk was measured using a commercial competitive assay (RIDAscreen, R-Biopharm, Pfungstadt, Germany). Absorbance was measured at 450 nm, where values are inversely related to vitamin B12 concentration using standards in the range of 0 to 30 µg/L. The detection limit of the assay was 0.5 µg/L.

      Data Preparation and Preprocessing

      Element and vitamin B12 data were combined with individual animal information before being subjected to quality control measures. For the purposes of the present study, the interquartile range (IQR) was calculated for each trait (x) with any concentrations falling out, with Q1 − 1.5 × IQR < x < Q3 − 1.5 × IQR (where Q1 and Q3 are the first and third quartiles, respectively) considered an outlier and removed from the data set. To ensure normality, all data were log-transformed before analysis. The final data set consisted of 938 milk and 754 serum records, with 385 (of 479) cows having more than 1 record.

      Statistical Analyses

      Data were analyzed using a mixed-effects linear animal model (Equation 1). Genetic relationships between individuals within the data set were accounted for by fitting a pedigree relationship matrix. Full pedigree information spanning 7 generations was available for all cows in the study:
      y = Xa + Z1b + Z2c + e.
      [1]


      Here, y is a vector of trait observations (i.e., mineral, vitamin, heavy metal); a is a vector of fixed effects; b is a vector of random additive genetic effects; c is a vector of permanent environmental effects; e is a vector of random residual effects; and X, Z1, and Z2 are incidence matrices linking phenotypic records to fixed and additive genetic and permanent environmental effects, respectively.
      Fixed effects included diet group, genetic group, lactation number, week in milk, year × season of calving interaction, and year × month of record interaction. Cow was fitted as a random effect to account for the additive genetic effect of the nth individual cow (pedigree data for 2,793 animals were included). To account for repeated observations per cow, the permanent environmental effect of the nth individual cow was also fitted as a random effect. All analyses were carried out using ASReml version 3 (
      • Gilmour A.R.
      • Gogel B.
      • Cullis B.
      • Thompson R.
      ASReml User Guide. Release 3.0.
      ).

      RESULTS

      Summary Statistics

      Table 1, Table 2 summarize the element and vitamin B12 data generated from milk and serum samples, respectively. Trait variability was determined by calculating the coefficient of determination (%). Within-milk variability was in the range of 22% (Zn) to 156% (Co) for trace elements and 11% (K) to 23% (Na) for quantity elements. In serum, variability ranged from 24% (Se) to 133% (Mn) for trace elements and 7% (Na) to 32% (P) for quantity elements. Variability was greater in serum compared with milk and in trace elements compared with quantity elements.
      Table 1Descriptive statistics of the milk element and vitamin B12 data
      ItemCountCows
      Number of cows/total cows with at least 2 observations of trait.
      MeanSDMinimumMaximumCV (%)
      Quantity elements (mg/L)
       Na878202/318364.1683.97148.92602.9723.06
       Mg908206/320113.1915.5869.74155.9813.76
       P904205/318946.18135.04586.191,300.7614.27
       K892202/3171,774.39191.811,239.382,301.4310.81
       Ca900203/3181,192.18181.14679.461,684.8615.19
      Trace elements (μg/L)
       V186103/1842.701.960.067.2272.43
       Cr726188/29825.5420.140.29104.6878.87
       Mn889204/32038.0814.491.0683.4638.05
       Fe848194/3121,077.02886.2641.224,017.0382.29
       Co41180/1644.306.710.0630.40156.12
       Ni321115/232427.28463.020.862,113.01108.37
       Cu916210/324120.7778.200.96386.8264.75
       Zn919208/3214,348.23938.971,916.916,833.1721.59
       Se916208/32319.104.676.8331.8124.46
       Mo888200/31437.2012.146.2671.8632.63
       I448131/2671,448.35610.52161.203,290.9042.15
      Heavy metals (μg/L)
       Cd733205/3110.160.100.010.4561.80
       Hg417188/3010.310.260.011.0483.35
       Pb829203/3164.823.170.0215.7265.75
      Vitamin B12 (μg/L)24763/640.820.380.501.7547.08
      1 Number of cows/total cows with at least 2 observations of trait.
      Table 2Descriptive statistics of the serum element data
      ItemCountCows
      Number of cows/total cows with at least 2 observations of trait.
      MeanSDMinimumMaximumCV (%)
      Quantity elements (mg/L)
       Na1691/1683,133.18210.642,533.073,757.496.72
       Mg734215/32223.015.119.7936.6222.22
       P743214/323139.0744.9720.75260.9832.33
       K732215/323206.0235.54105.38302.4517.25
       Ca707214/318105.3717.7658.03153.3616.86
      Trace elements (μg/L)
       V584213/2140.680.210.151.2531.15
       Cr22163/2140.890.760.004.2385.71
       Mn626215/2596.468.610.0736.70133.22
       Fe654213/2632,179.441,144.67292.156,157.9152.52
       Co610214/2301.340.680.114.3950.73
       Ni392201/2123.852.710.0112.1470.31
       Cu726215/315601.02195.51108.521,113.4632.53
       Zn626214/258910.90330.06201.572,109.5136.23
       Se717214/32275.9618.5826.11124.5924.45
       Mo662165/32114.9213.621.2662.8891.30
      Heavy metals (μg/L)
       Cd666215/2830.110.110.000.46100.76
       Hg265139/1513.433.220.0014.8993.97
       Pb738215/32583.2381.740.01367.5698.20
      1 Number of cows/total cows with at least 2 observations of trait.

      Effect of Genotype and Diet on Element Concentrations

      The effect of both genetic line and diet group on concentrations of micronutrients and heavy metals in milk and serum is summarized in Table 3, with P-values representing whether a significant difference was observed between predicted mean concentrations. Analyses revealed a predisposition for increased concentrations of Hg in the milk of select cows (mean = 0.19 μg/L, P = 0.01). No significant effect of genotype on the concentration of any other element in either milk or serum was observed. In contrast, diet was found to have a significant and varying effect on the concentration of 10 milk and 7 serum minerals. Cows on the HG diet had higher milk concentrations of Ca (mean = 1,257.02 mg/L, P < 0.001), Cu (mean = 61.39 μg/L, P = 0.01), I (mean = 1,346.82 μg/L, P < 0.001), K (mean = 1,739.23 mg/L, P < 0.001), Mn (mean = 36.82 μg/L, P < 0.001), Mo (mean = 41.89 μg/L, P < 0.001), and P (mean = 954.42 mg/L, P < 0.001) compared with those on the BP diet (Table 3). Conversely, cows on the BP diet had higher milk concentrations of Mg (mean = 116.50 mg/L, P < 0.001), Na (mean = 375.40 mg/L, P < 0.001), and Se (mean = 20.09 μg/L, P < 0.001). Regarding serum elements, cows on the HG diet showed higher concentrations of Cd (mean = 0.09 μg/L, P < 0.001), Cu (mean = 484.69 μg/L, P < 0.001), Fe (mean = 1,787.19 μg/L, P < 0.02), and Mo (mean = 17.49 μg/L, P < 0.001) compared with BP-fed cows, which showed higher serum concentrations of P (mean = 154.04 mg/L, P < 0.001), Se (mean = 73.03 μg/L, P < 0.001), and V (mean = 0.53 μg/L, P < 0.001).
      Table 3Effect of diet and genotype on element concentrations in milk and serum
      Predicted mean values were obtained via univariate models (accounting for all other sources of systematic variation). Only predicted mean concentrations that were significantly different (P < 0.05) are presented.
      ItemPredicted mean
      The homegrown ration consisted of components grown exclusively on farm and included grazed grass, grass silage, red clover silage, forage maize, lucerne silage, crimped wheat, and beans. Additionally, it was balanced with purchased minerals. In contrast, the by-product ration consisted of biscuit meal, sugar beet pulp, chopped straw, breakfast cereal, wheat distillers dark grains, soybean meal (Hipro 50%, Tarff Valley Ltd., Ringford, UK), Vitagold (KW Feeds, Ayrshire, UK), protected fat (Megalac, Volac International Ltd., Hertfordshire, UK), molasses, and minerals. The control line had been bred to bulls of UK national average genetic merit for kilograms of fat plus protein yield. In contrast, the high genetic merit select line (top 5% genetic merit) had been bred from bulls with the highest genetic merit for kilograms of fat plus protein yield.
      SED
      Standard error of the difference.
      P-value
      By-productHomegrown
      Milk
       Quantity elements (mg/L)
      Na375.40338.321.02<0.001
      Mg116.50110.811.01<0.001
      P878.40954.421.01<0.001
      K1,661.211,739.231.01<0.001
      Ca1,107.321,257.021.01<0.001
       Trace elements (μg/L)
      Mn27.4936.811.03<0.001
      Cu54.5561.391.050.009
      Se20.0915.141.01<0.001
      Mo32.9041.891.02<0.001
      I1,082.791,346.821.04<0.001
      Serum
       Quantity element (mg/L)
      P154.04108.801.02<0.001
       Trace elements (μg/L)
      V0.530.421.03<0.001
      Fe1,663.201,787.191.030.019
      Cu418.01484.691.02<0.001
      Se73.0264.651.02<0.001
      Mo5.4617.491.05<0.001
       Heavy metal (μg/L)
      Cd0.040.091.09<0.001
      MilkControl lineSelect line
       Heavy metal (μg/L)
      Hg0.140.191.110.010
      1 Predicted mean values were obtained via univariate models (accounting for all other sources of systematic variation). Only predicted mean concentrations that were significantly different (P < 0.05) are presented.
      2 The homegrown ration consisted of components grown exclusively on farm and included grazed grass, grass silage, red clover silage, forage maize, lucerne silage, crimped wheat, and beans. Additionally, it was balanced with purchased minerals. In contrast, the by-product ration consisted of biscuit meal, sugar beet pulp, chopped straw, breakfast cereal, wheat distillers dark grains, soybean meal (Hipro 50%, Tarff Valley Ltd., Ringford, UK), Vitagold (KW Feeds, Ayrshire, UK), protected fat (Megalac, Volac International Ltd., Hertfordshire, UK), molasses, and minerals. The control line had been bred to bulls of UK national average genetic merit for kilograms of fat plus protein yield. In contrast, the high genetic merit select line (top 5% genetic merit) had been bred from bulls with the highest genetic merit for kilograms of fat plus protein yield.
      3 Standard error of the difference.

      Variance Components

      Variance components of the milk and serum elements are presented in Table 4, Table 5, respectively. Additive genetic variance was small for both milk and serum traits, and in most cases genetic variance was higher in serum traits compared with those in milk. Heritability (h2) estimates were obtained for 17 of the 20 milk traits, 6 of which were significant (Mg, Ca, Mn, Zn, Se, and I). In serum, heritability estimates were obtained for 16 of 18 traits, of which 4 were significant (K, Ca, Cu, and Se). Significant element heritabilities appeared to be greater in milk traits compared with their corresponding serum trait. The highest heritability in milk and serum was observed in Mg (h2 = 0.30, P = 0.002) and Cu (h2 = 0.22, P < 0.001), respectively. Milk and serum quantity elements were all moderately to highly repeatable, and we observed significant repeatability in 5 milk and 10 serum trace elements.
      Table 4Variance components and heritability (h
      Significantly different from zero at P < 0.05.
      ) estimates of the milk elements and vitamin B12 data
      ItemSDh
      Significantly different from zero at P < 0.05.
      (SE)
      Repeatability (SE)
      Additive geneticPermanent environmentTotal phenotypic
      Quantity elements (mg/L)
       Na0.0000.0060.039NE
      Not estimable due to additive genetic variance >0.
      0.16
      Significantly different from zero at P < 0.05.
      (0.040)
       Mg0.0050.0000.0160.30
      Significantly different from zero at P < 0.05.
      (0.090)
      0.30
      Significantly different from zero at P < 0.05.
      (0.044)
       P0.0020.0020.0160.12 (0.070)0.22
      Significantly different from zero at P < 0.05.
      (0.042)
       K0.0010.0020.0100.11 (0.078)0.27
      Significantly different from zero at P < 0.05.
      (0.045)
       Ca0.0030.0000.0150.20
      Significantly different from zero at P < 0.05.
      (0.078)
      0.22
      Significantly different from zero at P < 0.05.
      (0.043)
      Trace elements (μg/L)
       V0.0630.0000.9780.06 (0.161)0.06 (0.161)
       Cr0.0130.0000.4500.03 (0.035)0.03 (0.035)
       Mn0.0200.0000.1390.14
      Significantly different from zero at P < 0.05.
      (0.039)
      0.14
      Significantly different from zero at P < 0.05.
      (0.039)
       Fe0.0060.0000.5310.01 (0.028)0.01 (0.028)
       Co0.0390.0000.9630.04 (0.056)0.04 (0.056)
       Ni0.0490.0001.2470.04 (0.093)0.04 (0.093)
       Cu0.0180.0000.4130.04 (0.028)0.04 (0.028)
       Zn0.0110.0090.0470.24
      Significantly different from zero at P < 0.05.
      (0.116)
      0.43
      Significantly different from zero at P < 0.05.
      (0.042)
       Se0.0050.0010.0330.15
      Significantly different from zero at P < 0.05.
      (0.072)
      0.18
      Significantly different from zero at P < 0.05.
      (0.041)
       Mo0.0110.0000.0600.19
      Significantly different from zero at P < 0.05.
      (0.041)
      0.19
      Significantly different from zero at P < 0.05.
      (0.041)
       I0.0310.0030.1400.22 (0.123)0.24
      Significantly different from zero at P < 0.05.
      (0.081)
      Heavy metals (μg/L)
       Cd0.0000.0000.511NENE
       Pb0.0000.0000.577NENE
       Hg0.0380.0001.0090.04 (0.064)0.04 (0.064)
      Vitamin B12 (μg/L)0.0340.0080.3460.10 (0.226)0.12 (0.095)
      1 Not estimable due to additive genetic variance >0.
      * Significantly different from zero at P < 0.05.
      Table 5Variance components and heritability (h
      Significantly different from zero at P < 0.05.
      ) estimates of the serum elements data
      ItemSDh
      Significantly different from zero at P < 0.05.
      (SE)
      Repeatability (SE)
      Additive geneticPermanent environmentTotal phenotypic
      Quantity elements (mg/L)
       Na0.0010.0030.0040.34 (0.234)1.00
      Significantly different from zero at P < 0.05.
      (0.002)
       Mg0.0070.0000.0470.14 (0.083)0.14
      Significantly different from zero at P < 0.05.
      (0.050)
       P0.0060.0070.0620.09 (0.079)0.20
      Significantly different from zero at P < 0.05.
      (0.046)
       K0.0040.0000.0240.18
      Significantly different from zero at P < 0.05.
      (0.051)
      0.18
      Significantly different from zero at P < 0.05.
      (0.051)
       Ca0.0030.0000.0250.12
      Significantly different from zero at P < 0.05.
      (0.049)
      0.12
      Significantly different from zero at P < 0.05.
      (0.049)
      Trace elements (μg/L)
       V0.0070.0170.0730.09 (0.121)0.33
      Significantly different from zero at P < 0.05.
      (0.059)
       Cr0.0000.0001.215NE
      Not estimable due to additive genetic variance >0.
      NE
       Mn0.0120.0000.3240.04 (0.044)0.04 (0.044)
       Fe0.0120.0000.1300.09 (0.046)0.09 (0.046)
       Co0.0090.0070.1240.07 (0.100)0.14
      Significantly different from zero at P < 0.05.
      (0.055)
       Ni0.2450.1130.8300.30 (0.183)0.43
      Significantly different from zero at P < 0.05.
      (0.076)
       Cu0.0180.0000.0840.22
      Significantly different from zero at P < 0.05.
      (0.051)
      0.22
      Significantly different from zero at P < 0.05.
      (0.051)
       Zn0.0070.0050.0670.11 (0.095)0.18
      Significantly different from zero at P < 0.05.
      (0.054)
       Se0.0050.0000.0530.10
      Significantly different from zero at P < 0.05.
      (0.047)
      0.10
      Significantly different from zero at P < 0.05.
      (0.047)
       Mo0.0000.0930.335NE0.28
      Significantly different from zero at P < 0.05.
      (0.057)
      Heavy metals (μg/L)
       Cd0.0440.1820.9090.05 (0.086)0.25
      Significantly different from zero at P < 0.05.
      (0.059)
       Pb0.0620.1530.5460.11 (0.103)0.39
      Significantly different from zero at P < 0.05.
      (0.051)
       Hg0.7090.6131.4330.49 (0.347)0.92
      Significantly different from zero at P < 0.05.
      (0.017)
      1 Not estimable due to additive genetic variance >0.
      * Significantly different from zero at P < 0.05.

      Associations Within Milk or Serum Elements

      Correlations between element concentrations within milk are presented in Table 6, and those within serum are presented in Table 7. Strong positive genetic correlations (significantly different from zero at P < 0.05) were observed within milk between the quantity elements, in particular Ca, Mg, and P (Table 6). Magnesium was positively associated with both Ca (r = 0.45, P < 0.001) and P (r = 0.49, P < 0.001); a positive association between P and Ca was also observed (r = 0.61, P < 0.001). Moreover, strong positive genetic correlations were observed for Se with Ca (r = 0.63, P < 0.001), Mg (r = 0.59, P < 0.001), Mn (r = 0.40, P = 0.034), P (r = 0.53, P < 0.001), and Zn (r = 0.52, P < 0.001). Consistent phenotypic relationships were also observed between the milk micronutrients and are presented in full in Table 6.
      Table 6Correlations between element concentrations and vitamin B12 within milk
      Additive genetic correlations are presented above the diagonal, and phenotypic correlations are presented below. Corresponding SE are given in parentheses.
      ItemNaMgPKCaVCrMnFeCoNiCuZnSeMoIHgB12
      Na0.16−0.13−0.260.030.56−0.60−0.04−0.020.32
      Not estimable.
      −0.27−0.150.09−0.06−0.210.780.17
      (0.16)(0.21)(0.20)(0.19)(0.88)(0.47)(0.22)(0.58)(0.55)(0.36)(0.16)(0.20)(0.19)(0.24)(0.63)(0.40)
      Mg0.36
      Significantly different from zero at P < 0.05.
      0.49
      Significantly different from zero at P < 0.05.
      0.200.45
      Significantly different from zero at P < 0.05.
      −0.47−0.840.200.05−0.120.56−0.240.170.59
      Significantly different from zero at P < 0.05.
      0.130.10−0.01−0.05
      (0.04)(0.11)(0.13)(0.12)(0.91)(0.59)(0.17)(0.45)(0.38)(0.85)(0.26)(0.11)(0.11)(0.15)(0.19)(0.44)(0.31)
      P0.39
      Significantly different from zero at P < 0.05.
      0.56
      Significantly different from zero at P < 0.05.
      −0.040.61
      Significantly different from zero at P < 0.05.
      −0.61−0.110.55
      Significantly different from zero at P < 0.05.
      0.250.310.310.230.53
      Significantly different from zero at P < 0.05.
      0.22−0.27−0.070.21
      (0.04)(0.03)(0.17)(0.09)(1.28)(0.37)(0.16)(0.53)(0.44)(0.30)(0.12)(0.13)(0.16)(0.20)(0.50)(0.34)
      K0.19
      Significantly different from zero at P < 0.05.
      0.35
      Significantly different from zero at P < 0.05.
      0.54
      Significantly different from zero at P < 0.05.
      0.02−0.570.270.290.310.460.15−0.19−0.05−0.090.020.120.130.18
      (0.04)(0.04)(0.03)(0.16)(0.75)(0.38)(0.17)(0.58)(0.47)(0.53)(0.27)(0.13)(0.17)(0.16)(0.20)(0.49)(0.37)
      Ca0.30
      Significantly different from zero at P < 0.05.
      0.57
      Significantly different from zero at P < 0.05.
      0.75
      Significantly different from zero at P < 0.05.
      0.43
      Significantly different from zero at P < 0.05.
      −0.290.010.57
      Significantly different from zero at P < 0.05.
      −0.07−0.150.930.370.25
      Significantly different from zero at P < 0.05.
      0.63
      Significantly different from zero at P < 0.05.
      0.14−0.050.220.15
      (0.04)(0.03)(0.02)(0.04)(0.79)(0.39)(0.17)(0.54)(0.43)(0.89)(0.29)(0.12)(0.12)(0.16)(0.21)(0.70)(0.36)
      V−0.190.01−0.23
      Significantly different from zero at P < 0.05.
      −0.070.020.700.720.840.42−0.080.32−0.290.460.36
      (0.10)(0.10)(0.10)(0.11)(0.11)(1.44)(1.10)(2.11)(1.42)(0.52)(0.80)(0.97)(1.04)(1.37)
      Cr0.07−0.020.010.02−0.050.02−0.04−0.62−0.76−0.16−0.620.03−0.850.07
      (0.04)(0.04)(0.04)(0.04)(0.04)(0.11)(0.40)(1.65)(1.55)(0.30)(0.44)(0.35)(0.70)(1.03)
      Mn0.19
      Significantly different from zero at P < 0.05.
      0.17
      Significantly different from zero at P < 0.05.
      0.22
      Significantly different from zero at P < 0.05.
      0.080.16
      Significantly different from zero at P < 0.05.
      −0.080.21
      Significantly different from zero at P < 0.05.
      0.010.130.550.240.41
      Significantly different from zero at P < 0.05.
      0.40
      Significantly different from zero at P < 0.05.
      0.26−0.02−0.420.33
      (0.04)(0.04)(0.04)(0.04)(0.04)(0.09)(0.04)(0.69)(0.37)(0.52)(0.31)(0.14)(0.18)(0.18)(0.24)(0.53)(0.40)
      Fe0.10
      Significantly different from zero at P < 0.05.
      0.040.070.05−0.02−0.060.40
      Significantly different from zero at P < 0.05.
      0.42
      Significantly different from zero at P < 0.05.
      0.830.660.050.240.30−0.34
      (0.04)(0.04)(0.04)(0.04)(0.04)(0.10)(0.04)(0.03)(1.04)(1.05)(0.43)(0.50)(0.59)(0.74)
      Co0.070.110.14
      Significantly different from zero at P < 0.05.
      0.04−0.04−0.000.32
      Significantly different from zero at P < 0.05.
      0.37
      Significantly different from zero at P < 0.05.
      0.59
      Significantly different from zero at P < 0.05.
      −0.420.300.26−0.76−0.44−0.82
      (0.06)(0.06)(0.06)(0.06)(0.06)(0.12)(0.06)(0.05)(0.03)(0.69)(0.51)(0.40)(1.14)(0.57)(1.19)
      Ni−0.23
      Significantly different from zero at P < 0.05.
      −0.110.04−0.000.100.090.17
      Significantly different from zero at P < 0.05.
      0.50
      Significantly different from zero at P < 0.05.
      0.01−0.050.380.99
      (0.07)(0.07)(0.07)(0.07)(0.07)(0.07)(0.07)(0.05)(0.39)(0.43)(0.89)(2.42)
      Cu0.040.020.01−0.09
      Significantly different from zero at P < 0.05.
      0.040.090.03−0.09
      Significantly different from zero at P < 0.05.
      −0.24
      Significantly different from zero at P < 0.05.
      0.200.27−0.21−0.640.19
      (0.04)(0.04)(0.04)(0.04)(0.04)(0.10)(0.04)(0.04)(0.05)(0.24)(0.29)(0.29)(0.35)(0.78)
      Zn0.22
      Significantly different from zero at P < 0.05.
      0.37
      Significantly different from zero at P < 0.05.
      0.38
      Significantly different from zero at P < 0.05.
      0.15
      Significantly different from zero at P < 0.05.
      0.35
      Significantly different from zero at P < 0.05.
      0.000.13
      Significantly different from zero at P < 0.05.
      0.25
      Significantly different from zero at P < 0.05.
      0.15
      Significantly different from zero at P < 0.05.
      0.22
      Significantly different from zero at P < 0.05.
      −0.130.11
      Significantly different from zero at P < 0.05.
      0.52
      Significantly different from zero at P < 0.05.
      0.16−0.04−0.280.28
      (0.04)(0.04)(0.04)(0.04)(0.04)(0.10)(0.04)(0.04)(0.04)(0.06)(0.07)(0.04)(0.10)(0.13)(0.17)(0.48)(0.30)
      Se0.32
      Significantly different from zero at P < 0.05.
      0.49
      Significantly different from zero at P < 0.05.
      0.43
      Significantly different from zero at P < 0.05.
      0.15
      Significantly different from zero at P < 0.05.
      0.42
      Significantly different from zero at P < 0.05.
      −0.030.050.22
      Significantly different from zero at P < 0.05.
      0.10
      Significantly different from zero at P < 0.05.
      0.23
      Significantly different from zero at P < 0.05.
      −0.19
      Significantly different from zero at P < 0.05.
      0.030.48
      Significantly different from zero at P < 0.05.
      0.250.08−0.11−0.01
      (0.04)(0.03)(0.03)(0.04)(0.03)(0.10)(0.04)(0.04)(0.04)(0.05)(0.07)(0.04)(0.03)(0.17)(0.22)(0.47)(0.36)
      Mo0.11
      Significantly different from zero at P < 0.05.
      0.20
      Significantly different from zero at P < 0.05.
      0.26
      Significantly different from zero at P < 0.05.
      0.070.23
      Significantly different from zero at P < 0.05.
      −0.000.14
      Significantly different from zero at P < 0.05.
      0.29
      Significantly different from zero at P < 0.05.
      0.25
      Significantly different from zero at P < 0.05.
      0.37
      Significantly different from zero at P < 0.05.
      0.14
      Significantly different from zero at P < 0.05.
      0.040.23
      Significantly different from zero at P < 0.05.
      0.21
      Significantly different from zero at P < 0.05.
      −0.12−0.47−0.14
      (0.04)(0.04)(0.04)(0.04)(0.04)(0.10)(0.04)(0.04)(0.04)(0.05)(0.07)(0.04)(0.04)(0.04)(0.22)(0.44)(0.34)
      I−0.060.01−0.050.08−0.020.010.03−0.050.03−0.05−0.01−0.17
      Significantly different from zero at P < 0.05.
      −0.04−0.000.14
      Significantly different from zero at P < 0.05.
      −0.40
      (0.05)(0.06)(0.06)(0.06)(0.06)(0.17)(0.07)(0.06)(0.06)(0.12)(0.09)(0.06)(0.06)(0.06)(0.06)(0.50)
      Hg0.07−0.05−0.16
      Significantly different from zero at P < 0.05.
      −0.11−0.14
      Significantly different from zero at P < 0.05.
      0.040.13
      Significantly different from zero at P < 0.05.
      −0.040.070.060.100.010.000.03−0.22
      Significantly different from zero at P < 0.05.
      −0.93
      (0.06)(0.06)(0.06)(0.06)(0.06)(0.15)(0.06)(0.05)(0.08)(0.11)(0.05)(0.06)(0.06)(0.06)(0.09)(0.95)
      B120.13−0.000.040.070.06−0.010.050.030.190.020.16
      Significantly different from zero at P < 0.05.
      0.08−0.10
      (0.09)(0.08)(0.09)(0.09)(0.09)(0.17)(0.08)(0.10)(0.10)(0.06)(0.08)(0.07)(0.09)
      1 Additive genetic correlations are presented above the diagonal, and phenotypic correlations are presented below. Corresponding SE are given in parentheses.
      2 Not estimable.
      * Significantly different from zero at P < 0.05.
      Table 7Correlations between element concentrations within serum
      Additive genetic correlations are presented above the diagonal, and phenotypic correlations are presented below. Corresponding SE are given in parentheses.
      ItemNaMgPKCaVMnFeCoNiCuZnSeMoCdHgPb
      Na0.60
      Significantly different from zero at P < 0.05.
      Not estimable.
      0.58
      Significantly different from zero at P < 0.05.
      0.570.240.500.450.270.40−0.150.180.590.140.670.98
      Significantly different from zero at P < 0.05.
      (0.37)(0.24)(0.56)(0.29)(0.66)(0.40)(0.39)(0.35)(0.34)(0.36)(0.337)(0.33)(0.35)(0.07)
      Mg0.68
      Significantly different from zero at P < 0.05.
      0.43
      Significantly different from zero at P < 0.05.
      0.53
      Significantly different from zero at P < 0.05.
      0.49
      Significantly different from zero at P < 0.05.
      0.280.010.77
      Significantly different from zero at P < 0.05.
      −0.080.140.190.51
      Significantly different from zero at P < 0.05.
      −0.260.36−0.180.05−0.01
      (0.05)(0.16)(0.17)(0.20)(0.19)(0.49)0.17)(0.29)(0.22)(0.20)(0.17)(0.31)(0.20)(0.25)(0.21)(0.19)
      P0.68
      Significantly different from zero at P < 0.05.
      0.66
      Significantly different from zero at P < 0.05.
      0.67
      Significantly different from zero at P < 0.05.
      0.42
      Significantly different from zero at P < 0.05.
      −0.380.32−0.05−0.010.010.030.32−0.020.03−0.120.00
      (0.03)(0.11)(0.11)(0.15)(1.04)(0.24)(0.24)(0.19)(0.18)(0.21)(0.18)(0.20)(0.22)(0.18)(0.16)
      K0.71
      Significantly different from zero at P < 0.05.
      0.64
      Significantly different from zero at P < 0.05.
      0.71
      Significantly different from zero at P < 0.05.
      0.63
      Significantly different from zero at P < 0.05.
      0.45
      Significantly different from zero at P < 0.05.
      −0.220.390.050.100.170.230.200.55
      Significantly different from zero at P < 0.05.
      0.42−0.200.05
      (0.05)(0.03)(0.02)(0.14)(0.16)(0.49)(0.25)(0.25)(0.20)(0.19)(0.20)(0.24)(0.18)(0.21)(0.18)(0.17)
      Ca0.63
      Significantly different from zero at P < 0.05.
      0.70
      Significantly different from zero at P < 0.05.
      0.82
      Significantly different from zero at P < 0.05.
      0.77
      Significantly different from zero at P < 0.05.
      0.53
      Significantly different from zero at P < 0.05.
      −0.210.49−0.060.070.260.340.080.48
      Significantly different from zero at P < 0.05.
      0.59
      Significantly different from zero at P < 0.05.
      −0.220.09
      (0.06)(0.02)(0.02)(0.02)(0.17)(0.70)(0.25)(0.30)(0.24)(0.19)(0.17)(0.31)(0.22)(0.25)(0.22)(0.21)
      V0.090.37
      Significantly different from zero at P < 0.05.
      0.46
      Significantly different from zero at P < 0.05.
      0.42
      Significantly different from zero at P < 0.05.
      0.43
      Significantly different from zero at P < 0.05.
      0.63
      Significantly different from zero at P < 0.05.
      0.18−0.050.310.030.210.210.57
      Significantly different from zero at P < 0.05.
      0.71
      Significantly different from zero at P < 0.05.
      −0.44
      Significantly different from zero at P < 0.05.
      −0.09
      (0.25)(0.04)(0.04)(0.04)(0.04)(0.29)(0.24)(0.23)(0.18)(0.18)(0.18)(0.21)(0.14)(0.14)(0.14)(0.15)
      Mn0.34
      Significantly different from zero at P < 0.05.
      0.26
      Significantly different from zero at P < 0.05.
      0.30
      Significantly different from zero at P < 0.05.
      0.23
      Significantly different from zero at P < 0.05.
      0.26
      Significantly different from zero at P < 0.05.
      0.35
      Significantly different from zero at P < 0.05.
      0.020.23−0.15−0.030.50−0.17−0.200.280.08−0.40
      (0.17)(0.04)(0.04)(0.04)(0.04)(0.04)(0.60)(0.49)(0.45)(0.43)(0.43)(0.58)(0.48)(0.44)(0.39)(0.44)
      Fe0.230.61
      Significantly different from zero at P < 0.05.
      0.55
      Significantly different from zero at P < 0.05.
      0.42
      Significantly different from zero at P < 0.05.
      0.55
      Significantly different from zero at P < 0.05.
      0.34
      Significantly different from zero at P < 0.05.
      0.29
      Significantly different from zero at P < 0.05.
      0.11−0.210.170.43−0.040.37−0.52−0.15−0.12
      (0.15)(0.03)(0.03)(0.04)(0.03)(0.04)(0.04)(0.34)(0.27)(0.25)(0.24)(0.34)(0.24)(0.30)(0.25)(0.24)
      Co0.330.57
      Significantly different from zero at P < 0.05.
      0.59
      Significantly different from zero at P < 0.05.
      0.45
      Significantly different from zero at P < 0.05.
      0.47
      Significantly different from zero at P < 0.05.
      0.32
      Significantly different from zero at P < 0.05.
      0.31
      Significantly different from zero at P < 0.05.
      0.54
      Significantly different from zero at P < 0.05.
      0.430.08−0.160.10−0.47−0.26−0.67
      Significantly different from zero at P < 0.05.
      −0.20
      (0.19)(0.03)(0.03)(0.04)(0.04)(0.05)(0.04)(0.03)(0.24)(0.22)(0.30)(0.29)(0.28)(0.28)(0.22)(0.20)
      Ni−0.33−0.01−0.02−0.030.020.070.100.010.050.110.100.410.210.22−0.35
      Significantly different from zero at P < 0.05.
      0.29
      (0.50)(0.06)(0.06)(0.06)(0.06)(0.07)(0.06)(0.06)(0.06)(0.18)(0.21)(0.25)(0.21)(0.21)(0.14)(0.16)
      Cu0.37
      Significantly different from zero at P < 0.05.
      0.61
      Significantly different from zero at P < 0.05.
      0.53
      Significantly different from zero at P < 0.05.
      0.56
      Significantly different from zero at P < 0.05.
      0.67
      Significantly different from zero at P < 0.05.
      0.35
      Significantly different from zero at P < 0.05.
      0.26
      Significantly different from zero at P < 0.05.
      0.47
      Significantly different from zero at P < 0.05.
      0.49
      Significantly different from zero at P < 0.05.
      0.060.220.110.19−0.07−0.30−0.01
      (0.10)(0.03)(0.03)(0.03)(0.03)(0.05)(0.04)(0.04)(0.04)(0.06)(0.18)(0.23)(0.19)(0.21)(0.18)(0.16)
      Zn−0.280.70
      Significantly different from zero at P < 0.05.
      0.62
      Significantly different from zero at P < 0.05.
      0.54
      Significantly different from zero at P < 0.05.
      0.70
      Significantly different from zero at P < 0.05.
      0.39
      Significantly different from zero at P < 0.05.
      0.34
      Significantly different from zero at P < 0.05.
      0.61
      Significantly different from zero at P < 0.05.
      0.55
      Significantly different from zero at P < 0.05.
      0.060.59
      Significantly different from zero at P < 0.05.
      −0.070.21−0.04−0.150.51
      Significantly different from zero at P < 0.05.
      (0.19)(0.02)(0.03)(0.03)(0.03)(0.04)(0.04)(0.03)(0.04)(0.06)(0.03)(0.24)(0.22)(0.24)(0.19)(0.17)
      Se0.71
      Significantly different from zero at P < 0.05.
      0.58
      Significantly different from zero at P < 0.05.
      0.73
      Significantly different from zero at P < 0.05.
      0.60
      Significantly different from zero at P < 0.05.
      0.75
      Significantly different from zero at P < 0.05.
      0.37
      Significantly different from zero at P < 0.05.
      0.24
      Significantly different from zero at P < 0.05.
      0.48
      Significantly different from zero at P < 0.05.
      0.58
      Significantly different from zero at P < 0.05.
      −0.010.66
      Significantly different from zero at P < 0.05.
      0.59
      Significantly different from zero at P < 0.05.
      0.31−0.20−0.190.32
      (0.05)(0.03)(0.02)(0.03)(0.02)(0.04)(0.04)(0.04)(0.03)(0.06)(0.03)(0.03)(0.26)(0.28)(0.24)(0.21)
      Mo0.32
      Significantly different from zero at P < 0.05.
      0.33
      Significantly different from zero at P < 0.05.
      0.18
      Significantly different from zero at P < 0.05.
      0.23
      Significantly different from zero at P < 0.05.
      0.16
      Significantly different from zero at P < 0.05.
      0.35
      Significantly different from zero at P < 0.05.
      0.100.32
      Significantly different from zero at P < 0.05.
      0.24
      Significantly different from zero at P < 0.05.
      0.050.31
      Significantly different from zero at P < 0.05.
      0.30
      Significantly different from zero at P < 0.05.
      0.14
      Significantly different from zero at P < 0.05.
      0.10−0.210.49
      Significantly different from zero at P < 0.05.
      (0.10)(0.04)(0.05)(0.04)(0.05)(0.05)(0.05)(0.04)(0.05)(0.07)(0.04)(0.05)(0.05)(0.22)(0.19)(0.15)
      Cd0.020.14
      Significantly different from zero at P < 0.05.
      0.09
      Significantly different from zero at P < 0.05.
      0.16
      Significantly different from zero at P < 0.05.
      0.14
      Significantly different from zero at P < 0.05.
      0.34
      Significantly different from zero at P < 0.05.
      0.16
      Significantly different from zero at P < 0.05.
      0.080.13
      Significantly different from zero at P < 0.05.
      0.110.10
      Significantly different from zero at P < 0.05.
      0.15
      Significantly different from zero at P < 0.05.
      0.050.23
      Significantly different from zero at P < 0.05.
      −0.23−0.24
      (0.16)(0.00)(0.05)(0.04)(0.05)(0.04)(0.04)(0.05)(0.05)(0.07)(0.05)(0.05)(0.05)(0.05)(0.17)(0.18)
      Hg0.050.110.04−0.01−0.20
      Significantly different from zero at P < 0.05.
      0.030.01−0.17
      Significantly different from zero at P < 0.05.
      −0.19
      Significantly different from zero at P < 0.05.
      −0.05−0.04−0.01−0.10−0.10−0.11
      (0.08)(0.08)(0.07)(0.07)(0.08)(0.07)(0.07)(0.07)(0.09)(0.08)(0.08)(0.07)(0.09)(0.07)(0.13)
      Pb0.86
      Significantly different from zero at P < 0.05.
      0.11
      Significantly different from zero at P < 0.05.
      0.21
      Significantly different from zero at P < 0.05.
      0.13
      Significantly different from zero at P < 0.05.
      0.16
      Significantly different from zero at P < 0.05.
      0.13
      Significantly different from zero at P < 0.05.
      0.10
      Significantly different from zero at P < 0.05.
      0.14
      Significantly different from zero at P < 0.05.
      0.11
      Significantly different from zero at P < 0.05.
      0.19
      Significantly different from zero at P < 0.05.
      0.10
      Significantly different from zero at P < 0.05.
      0.24
      Significantly different from zero at P < 0.05.
      0.19
      Significantly different from zero at P < 0.05.
      0.26
      Significantly different from zero at P < 0.05.
      −0.01−0.04
      (0.03)(0.05)(0.05)(0.05)(0.04)(0.05)(0.05)(0.05)(0.05)(0.06)(0.05)(0.05)(0.04)(0.05)(0.05)(0.08)
      1 Additive genetic correlations are presented above the diagonal, and phenotypic correlations are presented below. Corresponding SE are given in parentheses.
      2 Not estimable.
      * Significantly different from zero at P < 0.05.
      Within serum, a similar set of genetic relationships was observed between the quantity elements, although no significant genetic relationships were observed with Se (Table 7). The most genetically correlated nutrient in serum was V, which showed strong associations with Cd (r = 0.71, P = 0.003), Ca (r = 0.53, P < 0.001), Mn (r = 0.63, P = 0.039), Hg (r = −0.44, P = 0.003), Mo (r = 0.57, P < 0.001), P (r = 0.42, P = 0.006), and K (r = 0.45, P = 0.006). Phenotypically, V showed moderate to strong correlations with all other serum nutrients except for Cr, Hg, Ni, and Na.

      Associations Between Milk and Serum Elements

      Genetic correlations of elements between milk and serum are presented in Table 8, and phenotypic correlations are shown in Table 9 (results from the full analysis are available in Supplemental Table S3, https://doi.org/10.3168/jds.2019-16960). Significant additive genetic correlations (significantly different from zero at P < 0.05) were found to exist between several of the milk and serum elements, with most being positive (Table 8). The strongest negative associations were observed between serum Ni with milk V (r = −0.98, P = 0.008), Co (r = −0.86, P = 0.011), and Na (r = −0.62, P = 0.017). Potassium was the only element that was found to have a significant correlation between concentrations recorded in milk and serum (r = 0.45, P = 0.025). Further, milk K was found to be significantly positively correlated with serum Mg (r = 0.53, P = 0.008), Co (r = 0.48, P = 0.039), Mo (r = 0.45, P = 0.020), and Cd (r = 0.43, P = 0.035). Moreover, serum Mg was highly correlated with milk Ca (r = 0.54, P = 0.014), Mn (r = 0.57, P = 0.028), and P (r = 0.49, P = 0.029). Associations involving heavy metals were observed only between serum Cd and milk Se, K, and Mn.
      Table 8Additive genetic correlations (r) between milk and serum element concentrations with corresponding SE and P-values
      Only correlations significantly different from zero at P < 0.05 are presented.
      MilkSerumr (SE)P-value
      Na
      Quantity element.
      (mg/L)
      Ca
      Quantity element.
      (mg/L)
      0.56 (0.275)0.049
      Na
      Quantity element.
      (mg/L)
      Ni
      Trace element.
      (μg/L)
      −0.62 (0.244)0.017
      Mg
      Quantity element.
      (mg/L)
      Fe
      Trace element.
      (μg/L)
      0.65 (0.266)0.021
      P
      Quantity element.
      (mg/L)
      Mg
      Quantity element.
      (mg/L)
      0.49 (0.215)0.029
      K
      Quantity element.
      (mg/L)
      Cd
      Heavy metal.
      (μg/L)
      0.43 (0.194)0.035
      K
      Quantity element.
      (mg/L)
      Co
      Trace element.
      (μg/L)
      0.48 (0.222)0.039
      K
      Quantity element.
      (mg/L)
      Mg
      Quantity element.
      (mg/L)
      0.53 (0.189)0.008
      K
      Quantity element.
      (mg/L)
      Mo
      Trace element.
      (μg/L)
      0.45 (0.182)0.020
      K
      Quantity element.
      (mg/L)
      K
      Quantity element.
      (mg/L)
      0.45 (0.192)0.025
      Ca
      Quantity element.
      (mg/L)
      Mg
      Quantity element.
      (mg/L)
      0.54 (0.210)0.014
      V
      Trace element.
      (μg/L)
      Ni
      Trace element.
      (μg/L)
      −0.98 (0.354)0.008
      Mn
      Trace element.
      (μg/L)
      Cd
      Heavy metal.
      (μg/L)
      0.77 (0.219)<0.001
      Mn
      Trace element.
      (μg/L)
      Mg
      Quantity element.
      (mg/L)
      0.57 (0.247)0.028
      Co
      Trace element.
      (μg/L)
      Ni
      Trace element.
      (μg/L)
      −0.86 (0.319)0.011
      Se
      Trace element.
      (μg/L)
      Cd
      Heavy metal.
      (μg/L)
      −0.46 (0.210)0.036
      Mo
      Trace element.
      (μg/L)
      Fe
      Trace element.
      (μg/L)
      0.68 (0.302)0.031
      1 Only correlations significantly different from zero at P < 0.05 are presented.
      2 Quantity element.
      3 Trace element.
      4 Heavy metal.
      Table 9Phenotypic correlations (r) between milk and serum elements and vitamin B12 with corresponding SE and P-values
      Only correlations significantly different from zero at P < 0.05 are presented.
      MilkSerumr (SE)P-value
      Na
      Quantity element.
      (mg/L)
      Cd
      Heavy metal.
      (μg/L)
      0.39 (0.070)<0.001
      Na
      Quantity element.
      (mg/L)
      Pb
      Heavy metal.
      (μg/L)
      0.36 (0.074)<0.001
      Na
      Quantity element.
      (mg/L)
      Na
      Quantity element.
      (mg/L)
      −0.79 (0.175)<0.001
      Mg
      Quantity element.
      (mg/L)
      Fe
      Trace element.
      (μg/L)
      0.19 (0.064)0.004
      Mg
      Quantity element.
      (mg/L)
      Na
      Quantity element.
      (mg/L)
      0.49 (0.208)0.024
      P
      Quantity element.
      (mg/L)
      Cd
      Heavy metal.
      (μg/L)
      0.32 (0.072)<0.001
      P
      Quantity element.
      (mg/L)
      Ca
      Quantity element.
      (mg/L)
      0.15 (0.071)0.043
      P
      Quantity element.
      (mg/L)
      Fe
      Trace element.
      (μg/L)
      0.15 (0.069)0.037
      P
      Quantity element.
      (mg/L)
      Mg
      Quantity element.
      (mg/L)
      0.15 (0.070)0.037
      K
      Quantity element.
      (mg/L)
      Cd
      Heavy metal.
      (μg/L)
      0.30 (0.074)<0.001
      K
      Quantity element.
      (mg/L)
      Ca
      Quantity element.
      (mg/L)
      0.18 (0.069)0.014
      K
      Quantity element.
      (mg/L)
      Co
      Trace element.
      (μg/L)
      0.21 (0.066)0.002
      K
      Quantity element.
      (mg/L)
      Fe
      Trace element.
      (μg/L)
      0.15 (0.068)0.037
      K
      Quantity element.
      (mg/L)
      Mg
      Quantity element.
      (mg/L)
      0.20 (0.067)0.005
      K
      Quantity element.
      (mg/L)
      Mo
      Trace element.
      (μg/L)
      0.18 (0.076)0.027
      K
      Quantity element.
      (mg/L)
      P
      Quantity element.
      (mg/L)
      0.16 (0.064)0.019
      K
      Quantity element.
      (mg/L)
      K
      Quantity element.
      (mg/L)
      0.19 (0.067)0.006
      K
      Quantity element.
      (mg/L)
      Na
      Quantity element.
      (mg/L)
      0.64 (0.173)<0.001
      Ca
      Quantity element.
      (mg/L)
      Cd
      Heavy metal.
      (μg/L)
      0.22 (0.078)0.006
      Ca
      Quantity element.
      (mg/L)
      Ca
      Quantity element.
      (mg/L)
      0.17 (0.070)0.019
      Ca
      Quantity element.
      (mg/L)
      Fe
      Trace element.
      (μg/L)
      0.16 (0.068)0.024
      Ca
      Quantity element.
      (mg/L)
      Mg
      Quantity element.
      (mg/L)
      0.14 (0.069)0.047
      Ca
      Quantity element.
      (mg/L)
      Na
      Quantity element.
      (mg/L)
      0.81 (0.118)<0.001
      V
      Trace element.
      (μg/L)
      Pb
      Heavy metal.
      (μg/L)
      0.53 (0.107)<0.001
      Cr
      Trace element.
      (μg/L)
      Se
      Trace element.
      (μg/L)
      −0.16 (0.070)0.029
      Fe
      Trace element.
      (μg/L)
      Cd
      Heavy metal.
      (μg/L)
      −0.28 (0.086)0.002
      Fe
      Trace element.
      (μg/L)
      Pb
      Heavy metal.
      (μg/L)
      −0.62 (0.046)<0.001
      Co
      Trace element.
      (μg/L)
      Cd
      Heavy metal.
      (μg/L)
      0.56 (0.063)<0.001
      Zn
      Trace element.
      (μg/L)
      Co
      Trace element.
      (μg/L)
      −0.15 (0.066)0.033
      Zn
      Trace element.
      (μg/L)
      Na
      Quantity element.
      (mg/L)
      0.74 (0.142)<0.001
      Se
      Trace element.
      (μg/L)
      Ca
      Quantity element.
      (mg/L)
      0.15 (0.068)0.037
      Mo
      Trace element.
      (μg/L)
      Cd
      Heavy metal.
      (μg/L)
      0.22 (0.072)0.004
      Mo
      Trace element.
      (μg/L)
      Fe
      Trace element.
      (μg/L)
      0.17 (0.063)0.011
      Mo
      Trace element.
      (μg/L)
      Mo
      Trace element.
      (μg/L)
      0.19 (0.070)0.009
      Hg
      Heavy metal.
      (μg/L)
      Cu
      Trace element.
      (μg/L)
      −0.23 (0.074)0.003
      Hg
      Heavy metal.
      (μg/L)
      Mo
      Trace element.
      (μg/L)
      −0.20 (0.089)0.034
      Vitamin B12 (μg/L)Cd
      Heavy metal.
      (μg/L)
      0.21 (0.092)0.028
      Vitamin B12 (μg/L)Ni
      Trace element.
      (μg/L)
      −0.38 (0.096)<0.001
      1 Only correlations significantly different from zero at P < 0.05 are presented.
      2 Quantity element.
      3 Trace element.
      4 Heavy metal.
      All phenotypic correlations obtained between milk and serum element concentrations are presented in Table 9. Statistically significant correlations were obtained for Ca (r = 0.17, P = 0.019), Mo (r = 0.19, P = 0.009), K (r = 0.19, P = 0.006), and Na (r = 0.79, P < 0.001). Serum Na was also found to be strongly positively correlated with milk Ca (r = 0.81, P < 0.001), Zn (r = 0.74, P < 0.001), K (r = 0.64, P < 0.001), and Mg (r = 0.49, P = 0.024). The majority of associations observed were positive, but negative correlations were noted for milk Cr with serum Se (r = −0.16, P = 0.029), milk Fe with serum Pb (r = −0.62, P < 0.001) and serum Cd (r = −0.288, P = 0.002), milk Hg with serum Cu (r = −0.23, P = 0.003) and serum Mo (r = −0.20, P = 0.034), milk Zn with serum Co (r = −0.15, P = 0.033), and milk B12 with serum Ni (r = −0.38, P < 0.001; Table 9). It was noted that Cd and Pb in milk as well as Cr in serum showed no associations with any other nutrient, whether in milk or serum.

      DISCUSSION

      The main aim of this study was to estimate (co)variance components of important dairy cattle milk and serum elements (minerals, heavy metals), as well as milk vitamins B12, to explore potential selection strategies for optimizing concentrations for the benefit of both the cow and the human dairy product consumer. Significant heritability estimates were obtained for 6 milk and 4 serum minerals in addition to repeatability estimates for 10 milk and 15 serum elements (Table 4, Table 5). From the literature, the majority of genetic analyses previously reported correspond to quantity elements in milk (a summary of h2 values in the literature can be found in Supplemental Table S4, https://doi.org/10.3168/jds.2019-16960). Milk Ca, Mg, P, K, and Na have been shown to have heritabilities ranging from 0.10 (
      • Toffanin V.
      • Penasa M.
      • McParland S.
      • Berry D.P.
      • Cassandro M.
      • De Marchi M.
      Genetic parameters for milk mineral content and acidity predicted by mid-infrared spectroscopy in Holstein-Friesian cows.
      ) to 0.72 (
      • Buitenhuis B.
      • Poulsen N.A.
      • Larsen L.B.
      • Sehested J.
      Estimation of genetic parameters and detection of quantitative trait loci for minerals in Danish Holstein and Danish Jersey milk.
      ), 0.08 (
      • Buitenhuis B.
      • Poulsen N.A.
      • Larsen L.B.
      • Sehested J.
      Estimation of genetic parameters and detection of quantitative trait loci for minerals in Danish Holstein and Danish Jersey milk.
      ) to 0.60 (
      • van Hulzen K.J.E.
      • Sprong R.C.
      • van der Meer R.
      • van Arendonk J.A.M.
      Genetic and nongenetic variation in concentration of selenium, calcium, potassium, zinc, magnesium, and phosphorus in milk of Dutch Holstein-Friesian cows.
      ), 0.12 (
      • Toffanin V.
      • Penasa M.
      • McParland S.
      • Berry D.P.
      • Cassandro M.
      • De Marchi M.
      Genetic parameters for milk mineral content and acidity predicted by mid-infrared spectroscopy in Holstein-Friesian cows.
      ) to 0.62 (
      • van Hulzen K.J.E.
      • Sprong R.C.
      • van der Meer R.
      • van Arendonk J.A.M.
      Genetic and nongenetic variation in concentration of selenium, calcium, potassium, zinc, magnesium, and phosphorus in milk of Dutch Holstein-Friesian cows.
      ), 0.19 (
      • Visentin G.
      • Niero G.
      • Berry D.P.
      • Costa A.
      • Cassandro M.
      • De Marchi M.
      • Penasa M.
      Genetic (co)variances between milk mineral concentration and chemical composition in lactating Holstein-Friesian dairy cows.
      ) to 0.46 (
      • van Hulzen K.J.E.
      • Sprong R.C.
      • van der Meer R.
      • van Arendonk J.A.M.
      Genetic and nongenetic variation in concentration of selenium, calcium, potassium, zinc, magnesium, and phosphorus in milk of Dutch Holstein-Friesian cows.
      ), and 0.20 (
      • Buitenhuis B.
      • Poulsen N.A.
      • Larsen L.B.
      • Sehested J.
      Estimation of genetic parameters and detection of quantitative trait loci for minerals in Danish Holstein and Danish Jersey milk.
      ) to 0.24 (
      • Visentin G.
      • Niero G.
      • Berry D.P.
      • Costa A.
      • Cassandro M.
      • De Marchi M.
      • Penasa M.
      Genetic (co)variances between milk mineral concentration and chemical composition in lactating Holstein-Friesian dairy cows.
      ), respectively. Heritability estimates for some milk trace elements have also been reported, including Cu (0.28,
      • Buitenhuis B.
      • Poulsen N.A.
      • Larsen L.B.
      • Sehested J.
      Estimation of genetic parameters and detection of quantitative trait loci for minerals in Danish Holstein and Danish Jersey milk.
      ), Fe (0.15,
      • Buitenhuis B.
      • Poulsen N.A.
      • Larsen L.B.
      • Sehested J.
      Estimation of genetic parameters and detection of quantitative trait loci for minerals in Danish Holstein and Danish Jersey milk.
      ), Mn (0.13,
      • Buitenhuis B.
      • Poulsen N.A.
      • Larsen L.B.
      • Sehested J.
      Estimation of genetic parameters and detection of quantitative trait loci for minerals in Danish Holstein and Danish Jersey milk.
      ), Se (0.20,
      • van Hulzen K.J.E.
      • Sprong R.C.
      • van der Meer R.
      • van Arendonk J.A.M.
      Genetic and nongenetic variation in concentration of selenium, calcium, potassium, zinc, magnesium, and phosphorus in milk of Dutch Holstein-Friesian cows.
      ; 0.20,
      • Buitenhuis B.
      • Poulsen N.A.
      • Larsen L.B.
      • Sehested J.
      Estimation of genetic parameters and detection of quantitative trait loci for minerals in Danish Holstein and Danish Jersey milk.
      ), and Zn (0.41,
      • van Hulzen K.J.E.
      • Sprong R.C.
      • van der Meer R.
      • van Arendonk J.A.M.
      Genetic and nongenetic variation in concentration of selenium, calcium, potassium, zinc, magnesium, and phosphorus in milk of Dutch Holstein-Friesian cows.
      ; 0.57,
      • Buitenhuis B.
      • Poulsen N.A.
      • Larsen L.B.
      • Sehested J.
      Estimation of genetic parameters and detection of quantitative trait loci for minerals in Danish Holstein and Danish Jersey milk.
      ). Regarding serum, a genetic analysis of Ca, Mg, P, and K carried out by
      • Tsiamadis V.
      • Banos G.
      • Panousis N.
      • Kritsepi-Konstantinou M.
      • Arsenos G.
      • Valergakis G.E.
      Genetic parameters of calcium, phosphorus, magnesium, and potassium serum concentrations during the first 8 days after calving in Holstein cows.
      reported heritabilities of 0.20, 0.21, 0.25, and 0.10, respectively. Furthermore, heritabilities of serum Cu and Zn have been reported as 0.22 (
      • Morris C.A.
      • Amyes N.C.
      • Hickey S.M.
      Genetic variation in serum copper concentration in Angus cattle.
      ). The results from the present study are within these ranges for these nutrients. We also investigated several milk and serum elements (including heavy metals) that we believe have not yet been reported in dairy cows. As such, we believe that this is the first study to estimate heritability of the milk trace element Mo (0.19) as well as repeatability estimates for milk and serum trace elements and heavy metals.
      Concentrations of I in milk are known to be affected by several factors, including dietary I level, the presence of I antagonists (e.g., glucosinolates) in the feed, farm management practices, teat dipping with I-containing substances, and milk processing (
      • Flachowsky G.
      • Franke K.
      • Meyer U.
      • Leiterer M.
      • Schöne F.
      Influencing factors on iodine content of cow milk.
      ). In the present study, milk I was influenced by diet type and was significantly repeatable (0.24, P = 0.005); this should be important given the importance of UK intakes of I for milk and dairy products (
      • Kliem K.E.
      • Givens D.I.
      Dairy products in the food chain: Their impact on health.
      ). Although mean milk I concentrations were much higher than those listed in the current UK Food Composition Database (
      • Food Standards Agency
      McCance and Widdowson's The Composition of Foods.
      ), it is important to note that the current study analyzed raw milk and that pasteurization is known to substantially reduce I concentrations in milk (
      • Nazeri P.
      • Norouzian M.A.
      • Mirmiran P.
      • Hedayati M.
      • Azizi F.
      Heating process in pasteurization and not in sterilization decreases the iodine concentration of milk.
      ).
      Sodium is another essential quantity element that has been shown to be an important factor in milk production (
      • Derrig R.G.
      • Clark J.H.
      • Davis C.L.
      Effect of abomasal infusion of sodium caseinate on milk yield, nitrogen utilization and amino acid nutrition of the dairy cow.
      ;
      • Spek J.W.
      • Bannink A.
      • Gort G.
      • Hendriks W.H.
      • Dijkstra J.
      Interaction between dietary content of protein and sodium chloride on milk urea concentration, urinary urea excretion, renal recycling of urea, and urea transfer to the gastrointestinal tract in dairy cows.
      ) and is lost though milk, urine, saliva, and feces (
      • Renkema J.A.
      • Senshu T.
      • Gaillard B.D.E.
      • Brouwer E.
      Regulation of sodium excretion and retention by the intestine in cows.
      ). We observed a strong negative phenotypic association between concentrations of Na in milk and serum (r = −0.79, P < 0.001), suggesting that increased milk Na concentrations correspond to a decrease in serum concentrations. During lactation, Na concentrations of milk have been shown to increase (
      • Gueguen L.
      • Journet M.
      • Langlois M.
      Les variations de la composition minérale du lait de vache.
      ;
      • Safwate A.
      • Davicco M.-J.
      • Barlet J.P.
      • Delost P.
      Sodium and potassium in blood and milk and plasma aldosterone levels in high-yield dairy cows.
      ), whereas in blood large variations (dependent on physiological condition or age) have been observed (
      • Skrzypczak W.
      • Kurpinska A.
      • Stanski L.
      • Jarosz A.
      Sodium, potassium and chloride homeostasis in cows during pregnancy and first months of lactation.
      ). Moreover, it has been hypothesized that decreased Na concentrations in early lactation may be due to decreased plasma rennin activity postcalving (

      Ożgo, M., W. F. Skrzypczak, K. Michałek, A. Lepczyński, A. Herosimczyk, and A. Dratwa. 2008. Regu- Maternal and Newborn Water and Electrolyte Balance. Monography, Wrocław, Poland.

      ).
      Milk is also an excellent source of vitamin B12, and milk and dairy products contribute significantly to vitamin B12 intakes in humans (150% of RNI;
      • Henderson L.
      • Gregory J.
      • Irving K.
      • Swan G.
      The National Diet and Nutrition Survey: Adults Aged 19–64 Years. Vol. 2: Energy, Protein, Carbohydrate, Fat and Alcohol Intake.
      ,
      • Henderson L.
      • Irving K.
      • Gregory J.
      • Bates C.J.
      • Prentice A.
      • Perks J.
      • Swan G.
      • Farron M.
      The National Diet and Nutrition Survey: Adults Aged 19–64 Years. Vol. 3: Vitamin and Mineral Intake and Urinary Analytes.
      ;
      • Kliem K.E.
      • Givens D.I.
      Dairy products in the food chain: Their impact on health.
      ), making it an attractive breeding target in terms of enhancing nutrient quality for the consumer. Vitamin B12 contains Co, and Co is required in the diet of cattle in order for this vitamin to be synthesized endogenously by rumen bacteria (
      • Stemme K.
      • Lebzien P.
      • Flachowsky G.
      • Scholz H.
      The influence of an increased cobalt supply on ruminal parameters and microbial vitamin B12 synthesis in the rumen of dairy cows.
      ). Although the concentration of Co in serum was repeatable (0.14, P = 0.019), we did not observe significant repeatability in milk Co or vitamin B12. The estimated heritability of vitamin B12 in milk was not significant (h2 = 0.12, P = 0.18); this was also true for milk and serum Co (h2 = 0.04, P = 0.31 and h2 = 0.07, P = 0.30, respectively). Furthermore, we found no significant associations between milk vitamin B12 and Co in either milk or serum.
      Genetic line (average or highest genetic merit for milk fat plus protein yield) had no significant effect on circulating element or vitamin B12 concentrations in either milk or blood serum with the exception of the heavy metal Hg, which showed higher concentrations in the milk of select-line cows. Moreover, because the cows were part of an experimental research herd, any biases in management between the genetic lines were nonexistent such that cows within the same line but on diverging diets were consequently unaffected by management decisions (
      • Pryce J.E.
      • Nielsen B.L.
      • Veerkamp R.F.
      • Simm G.
      Genotype and feeding system effects and interactions for health and fertility traits in dairy cattle.
      ). Concentrations of elements in both milk and serum were affected depending on whether the cow was fed the homegrown or by-product ration. This has potential benefits for manipulation of nutrient content through changes in management alone, a benefit that could be complemented or improved through selection and breeding. It also suggests that selection for higher milk fat and protein is independent of blood or milk micronutrient concentrations.
      Given the mostly positive genetic correlations among the milk minerals examined in the present study, selection alone for a single milk mineral might be expected to also increase the concentrations of other minerals. For example, selection for milk Ca would likely boost P, Zn, and Se concentrations, leading to multiple improvements in milk mineral concentrations for the benefit of the human dairy consumer.
      Furthermore, because heavy metals have adverse effects on health, any breeding objectives should also be directed toward minimizing concentrations of these metals or at least not increasing concentrations. The findings of this study identified few significant genetic associations of heavy metals with micronutrient concentrations, and in cases where a significant association was found, these tended to be negative. This suggests that genetic selection programs aimed at increasing micronutrient concentrations should not inadvertently increase concentrations of toxic heavy metals. The minimum risk level has not been established for Cd or Hg in milk, but the minimum risk level for Pb in EU milk is 20 µg/kg (

      European Union. 2001. CE regulation no. 2001/466. March 8, 2001. GUCE L77/1 March 16, 2001.

      ). The Pb concentrations as found in milk in this study were below levels of food safety concern in the European Union.
      It is interesting to note that although significant phenotypic relationships were observed between some milk and corresponding serum element measurements, only 1 genetic association was identified (between milk and serum K). Moreover, we observed stronger and additional relationships between differing nutrients between milk and serum. Results from the present study agree with those of
      • Wang H.
      • Liu Z.
      • Liu Y.
      • Qi Z.
      • Wang S.
      • Liu S.
      • Dong S.
      • Xia X.
      • Li S.
      Levels of Cu, Mn, Fe and Zn in cow serum and cow milk: Relationship with trace elements contents and chemical composition in milk.
      in that concentrations of Cu, Fe, and Zn in milk do not reflect corresponding serum concentrations. Additionally, our findings suggest that the same is true of all elements examined in the present study with the exception of Na, K, Ca, and Mo.

      CONCLUSIONS

      The present study established that circulating concentrations of elements in both the milk and serum of dairy cows are significantly influenced by genetics and feeding system. As expected, diet had a significant effect on mineral concentrations, especially in milk, and as such provided a potential route for manipulation via changes in rations. The results presented provide clear evidence that many of such traits are heritable, indicating that selection for desired element concentrations in both milk and serum is possible. This work will help inform industry solutions to better improve both genetics and management practices for the benefit of not only the cow but also the healthfulness of the milk for the consumer.

      ACKNOWLEDGMENTS

      This research, including the Langhill experiment at Crichton Dairy Research Centre and all authors, was funded by the Scottish Government Rural Affairs, Food and the Environment (RAFE) Strategic Research Portfolio 2016-2021. Samples collected before 2016 were collected as part of a Biotechnology and Biological Sciences Research Council project awarded to EW (grant no. BB/K002260/1) and TNM (BB/K002171/1). The authors gratefully acknowledge the high standard of work by all staff at Crichton farm (Scotland's Rural College, Dumfries, Scotland) in the collection of samples and management of animals and Ian Archibald (Scotland's Rural College, Edinburgh, Scotland) for managing the Langhill database and assisting with data extraction.

      Supplementary Materials

      REFERENCES

        • Alpert P.T.
        The role of vitamins and minerals on the immune system.
        Home Health Care Manage. Pract. 2017; 29: 199-202
        • Buitenhuis B.
        • Poulsen N.A.
        • Larsen L.B.
        • Sehested J.
        Estimation of genetic parameters and detection of quantitative trait loci for minerals in Danish Holstein and Danish Jersey milk.
        BMC Genet. 2015; 16 (25989905): 52
        • AHDB Dairy
        Average UK milk yields.
        • Denholm S.J.
        • McNeilly T.N.
        • Banos G.
        • Coffey M.P.
        • Russell G.C.
        • Bagnall A.
        • Mitchell M.C.
        • Wall E.
        Estimating genetic and phenotypic parameters of cellular immune-associated traits in dairy cows.
        J. Dairy Sci. 2017; 100 (28131586): 2850-2862
        • Denholm S.J.
        • McNeilly T.N.
        • Banos G.
        • Coffey M.P.
        • Russell G.C.
        • Bagnall A.
        • Mitchell M.C.
        • Wall E.
        Immune-associated traits measured in milk of Holstein-Friesian cows as proxies for blood serum measurements.
        J. Dairy Sci. 2018; 101 (30172405): 10248-10258
        • Derrig R.G.
        • Clark J.H.
        • Davis C.L.
        Effect of abomasal infusion of sodium caseinate on milk yield, nitrogen utilization and amino acid nutrition of the dairy cow.
        J. Nutr. 1974; 104 (4810975): 151-159
        • Doherty C.P.
        Host-pathogen interactions: The role of iron.
        J. Nutr. 2007; 137 (17449603): 1341-1344
      1. European Union. 2001. CE regulation no. 2001/466. March 8, 2001. GUCE L77/1 March 16, 2001.

        • FAO/WHO (Food and Agriculture Organization of the United Nations/World Health Organization)
        Vitamin and Mineral Requirements in Human Nutrition.
        2nd ed. World Health Organization, Geneva, Switzerland2004
        • Flachowsky G.
        • Franke K.
        • Meyer U.
        • Leiterer M.
        • Schöne F.
        Influencing factors on iodine content of cow milk.
        Eur. J. Nutr. 2014; 53 (24185833): 351-365
        • Food Standards Agency
        McCance and Widdowson's The Composition of Foods.
        7th ed. The Royal Society of Chemistry, Cambridge, UK2015
        • Gernand A.D.
        • Schulze K.J.
        • Stewart C.P.
        • West K.P.
        • Christian P.
        Micronutrient deficiencies in pregnancy worldwide: Health effects and prevention.
        Nat. Rev. Endocrinol. 2016; 12 (27032981): 274-289
        • Gilmour A.R.
        • Gogel B.
        • Cullis B.
        • Thompson R.
        ASReml User Guide. Release 3.0.
        VSN Int. Ltd., Hemel Hempstead, UK2009
        • Givens D.I.
        • Livingstone K.M.
        • Pickering J.E.
        • Fekete A.
        • Dougkas A.
        • Elwood P.C.
        Milk: White elixir or white poison? An examination of the associations between dairy consumption and disease in human subjects.
        Anim. Front. 2014; 4: 8-15
        • Gueguen L.
        • Journet M.
        • Langlois M.
        Les variations de la composition minérale du lait de vache.
        Ann. Biol. Anim. Biochim. Biophys. 1961; 1: 305-310
        • Henderson L.
        • Gregory J.
        • Irving K.
        • Swan G.
        The National Diet and Nutrition Survey: Adults Aged 19–64 Years. Vol. 2: Energy, Protein, Carbohydrate, Fat and Alcohol Intake.
        Office for National Statistics, London, UK2003
        • Henderson L.
        • Irving K.
        • Gregory J.
        • Bates C.J.
        • Prentice A.
        • Perks J.
        • Swan G.
        • Farron M.
        The National Diet and Nutrition Survey: Adults Aged 19–64 Years. Vol. 3: Vitamin and Mineral Intake and Urinary Analytes.
        Office for National Statistics, London, UK2003
        • Hoffmann P.R.
        • Berry M.J.
        The influence of selenium on immune responses.
        Mol. Nutr. Food Res. 2008; 52 (18384097): 1273-1280
        • Kliem K.E.
        • Givens D.I.
        Dairy products in the food chain: Their impact on health.
        Annu. Rev. Food Sci. Technol. 2011; 2 (22129373): 21-36
        • Maggini S.
        • Wintergerst E.S.
        • Beveridge S.
        • Hornig D.H.
        Selected vitamins and trace elements support immune function by strengthening epithelial barriers and cellular and humoral immune responses.
        Br. J. Nutr. 2007; 98 (17922955): S29-S35
        • Morris C.A.
        • Amyes N.C.
        • Hickey S.M.
        Genetic variation in serum copper concentration in Angus cattle.
        Anim. Sci. 2006; 82: 798-803
        • Muth O.H.
        • Oldfield J.E.
        • Remmert L.F.
        • Schubert J.R.
        Effects of selenium and vitamin E on white muscle disease.
        Science. 1958; 128: 1090
        • Nazeri P.
        • Norouzian M.A.
        • Mirmiran P.
        • Hedayati M.
        • Azizi F.
        Heating process in pasteurization and not in sterilization decreases the iodine concentration of milk.
        Int. J. Endocrinol. Metab. 2015; 13e27995
      2. Ożgo, M., W. F. Skrzypczak, K. Michałek, A. Lepczyński, A. Herosimczyk, and A. Dratwa. 2008. Regu- Maternal and Newborn Water and Electrolyte Balance. Monography, Wrocław, Poland.

        • Percival S.S.
        Copper and immunity.
        Am. J. Clin. Nutr. 1998; 67 (9587153): 1064S-1068S
        • Prasad A.S.
        Zinc in human health: Effect of zinc on immune cells.
        Mol. Med. 2008; 14 (18385818): 353-357
        • Pryce J.E.
        • Nielsen B.L.
        • Veerkamp R.F.
        • Simm G.
        Genotype and feeding system effects and interactions for health and fertility traits in dairy cattle.
        Livest. Prod. Sci. 1999; 57: 193-201
        • Renkema J.A.
        • Senshu T.
        • Gaillard B.D.E.
        • Brouwer E.
        Regulation of sodium excretion and retention by the intestine in cows.
        Nature. 1962; 195 (14491472): 389-390
        • Rey-Crespo F.
        • Miranda M.
        • López-Alonso M.
        Essential trace and toxic element concentrations in organic and conventional milk in NW Spain.
        Food Chem. Toxicol. 2013; 55 (23391598): 513-518
        • Roberts D.J.
        • March M.D.
        Feeding systems for dairy cows: Homegrown versus by-products feeds.
        in: Garnsworthy P.C. Wiseman J. Recent Advances in Animal Nutrition. Context Products Ltd., Leicestershire, UK2013: 61-70
        • Rooke J.
        • Flockhart J.
        • Sparks N.
        The potential for increasing the concentrations of micro-nutrients relevant to human nutrition in meat, milk and eggs.
        J. Agric. Sci. 2010; 148: 603-614
        • Safwate A.
        • Davicco M.-J.
        • Barlet J.P.
        • Delost P.
        Sodium and potassium in blood and milk and plasma aldosterone levels in high-yield dairy cows.
        Reprod. Nutr. Dev. 1981; 21 (7349545): 601-610
        • Skrzypczak W.
        • Kurpinska A.
        • Stanski L.
        • Jarosz A.
        Sodium, potassium and chloride homeostasis in cows during pregnancy and first months of lactation.
        Acta Biol. Cracoviensia. Ser. Zool. 2013; 55–56: 58-64
        • Spek J.W.
        • Bannink A.
        • Gort G.
        • Hendriks W.H.
        • Dijkstra J.
        Interaction between dietary content of protein and sodium chloride on milk urea concentration, urinary urea excretion, renal recycling of urea, and urea transfer to the gastrointestinal tract in dairy cows.
        J. Dairy Sci. 2013; 96 (23871366): 5734-5745
        • Stemme K.
        • Lebzien P.
        • Flachowsky G.
        • Scholz H.
        The influence of an increased cobalt supply on ruminal parameters and microbial vitamin B12 synthesis in the rumen of dairy cows.
        Arch. Anim. Nutr. 2008; 62 (18610536): 207-218
        • Toffanin V.
        • Penasa M.
        • McParland S.
        • Berry D.P.
        • Cassandro M.
        • De Marchi M.
        Genetic parameters for milk mineral content and acidity predicted by mid-infrared spectroscopy in Holstein-Friesian cows.
        Animal. 2015; 9 (25584638): 775-780
        • Tsiamadis V.
        • Banos G.
        • Panousis N.
        • Kritsepi-Konstantinou M.
        • Arsenos G.
        • Valergakis G.E.
        Genetic parameters of calcium, phosphorus, magnesium, and potassium serum concentrations during the first 8 days after calving in Holstein cows.
        J. Dairy Sci. 2016; 99 (27179858): 5535-5544
        • van Hulzen K.J.E.
        • Sprong R.C.
        • van der Meer R.
        • van Arendonk J.A.M.
        Genetic and nongenetic variation in concentration of selenium, calcium, potassium, zinc, magnesium, and phosphorus in milk of Dutch Holstein-Friesian cows.
        J. Dairy Sci. 2009; 92 (19841235): 5754-5759
        • Veerkamp R.F.
        • Simm G.
        • Oldham J.D.
        Effects of interaction between genotype and feeding system on milk production, feed intake, efficiency and body tissue mobilization in dairy cows.
        Livest. Prod. Sci. 1994; 39: 229-241
        • Visentin G.
        • Niero G.
        • Berry D.P.
        • Costa A.
        • Cassandro M.
        • De Marchi M.
        • Penasa M.
        Genetic (co)variances between milk mineral concentration and chemical composition in lactating Holstein-Friesian dairy cows.
        Animal. 2019; 13 (29976269): 477-486
        • Wang H.
        • Liu Z.
        • Liu Y.
        • Qi Z.
        • Wang S.
        • Liu S.
        • Dong S.
        • Xia X.
        • Li S.
        Levels of Cu, Mn, Fe and Zn in cow serum and cow milk: Relationship with trace elements contents and chemical composition in milk.
        Acta Sci. Vet. 2014; 42: 1-14
        • Yu W.H.
        A study of nutritional and bio-geochemical factors in the occurrence and development of Keshan disease.
        Jpn. Circ. J. 1982; 46 (7131711): 1201-1207