Dietary Supplementation of Selenium in Inorganic and Organic Forms Differentially and Commonly Alters Blood and Liver Selenium Concentrations and Liver Gene Expression Profiles of Growing Beef Heifers
- 428 Downloads
In geographic regions where selenium (Se) soil concentrations are naturally low, the addition of Se to animal feed is necessary. Even though it is known that Se in grass and forage crops is primarily present in organic forms (especially as L-selenomethionine, L-selenocystine, and L-selenocystathionine), the feeding of Se in the naturally occurring organic selenium (OSe) compounds produces higher blood and tissue Se levels than the inorganic Se (ISe) salts, and that animal metabolism of OSe and ISe is fundamentally different. Se is commonly added in inorganic form as sodium selenite to cattle feeds because it is a less expensive source of supplemental Se then are OSe forms. A trial was conducted with growing cattle to determine if the addition of OSe versus ISe forms of Se in beef cattle feed produces differences in hepatic gene expression, thereby gaining insight into the metabolic consequence of feeding OSe versus ISe. Thirty maturing Angus heifers (261 ± 6 days) were fed a corn silage-based diet with no Se supplementation for 75 days. Heifers (body weight = 393 ± 9 kg) then were randomly assigned (n = 10) and fed Se supplements that contained none (control) or 3 mg Se/day in ISe (sodium selenite) or OSe (Sel-Plex®) form and enough of a common cracked corn/cottonseed hull-based diet (0.48 mg Se/day) to support 0.5 kg/day growth for 105 or 106 days. More Se was found in jugular whole blood and red blood cells and biopsied liver tissue of ISe and OSe treatment animals than control animals, and OSe animals contained more Se in these tissues than did ISe. Microarray and bioinformatic analyses of liver tissue gene expression revealed that the content of at least 80 mRNA were affected by ISe or OSe treatments, including mRNA associated with nutrient metabolism; cellular growth, proliferation, and immune response; cell communication or signaling; and tissue/organ development and function. Overall, three Se supplement-dependent gene groups were identified: ISe-dependent, OSe-dependent, and Se form-independent. More specifically, both forms of supplementation appeared to upregulate mitochondrial gene expression capacity, whereas gene expression of a protein involved in antiviral capacity was downregulated in ISe-supplemented animals, and OSe-supplemented animals had reduced levels of mRNA encoding proteins known to be upregulated during oxidative stress and cancerous states.
KeywordsBovine Blood Liver Microarray Nutrient–gene interaction Selenium supplementation
analysis of variance
glutamate-cysteine ligase catalytic subunit
glutamate-cysteine ligase modifier subunit
insulin-like growth factor
insulin-like growth factor binding protein
protein kinase C iota
A survey of whole blood Se concentrations to determine the geographic distribution of Se deficiency among beef cows and heifers in the United States found that 42% of cattle in the southeastern United States (including Kentucky) were Se deficient (≤0.080 μg/mL) compared to 18% of deficient animals nationally . For beef cattle production in the low-Se regions of Kentucky, additions of Se to feedstock are necessary [2, 3]. To ruminant feed, selenium is still usually added in inorganic chemical forms (ISe), either as sodium selenite or sodium selenate, as these were the first to be approved as animal feed additives [4, 5]. More recently, Se-enriched yeast, in which L-selenomethionine is the predominant form of Se, has become available and has been approved as a feed additive for beef cattle production [6, 7].
According to the National Research Council  and Food and Drug Administration , Se can be legally supplemented in beef cattle diets to provide 3 mg/head daily or 0.3 mg/kg in complete diets, although the Se requirement of beef cattle can be met by 0.1 mg/kg of diet. The acceptable level of Se in free-choice salt–mineral mixtures is 120 mg/kg .
In terms of bioavailability and biopotency of dietary organic Se (OSe) versus ISe supplementation, many comparisons have been made in the last two decades using a variety of domestic animal species [8, 9, 10]. The amount of total Se versus OSe forms in animal tissues has been compared. In the breast and leg tissues of chickens, selenomethionine and selenocysteine accounted for about 67% and 20% and 56% and 32%, respectively, of the total Se content . In lambs, selenomethionine and selenocysteine accounted for about 57% and 36% and 45 and 36% of the total Se content in the tenderloin and rib eye muscles, respectively. In contrast, selenomethionine and selenocysteine accounted for about 20% and 60%, 18% and 60%, and 8.8% and 88%, of the total Se tissue content in the heart, liver, and kidney tissue, respectively, of lambs . Despite these basal tissue-specific differences in selenomethionine and selenocysteine content, a follow-up dose–effect trial found that more of the diet-supplemented selenomethionine accumulated in all of these lamb tissues than did equivalent amounts of dosed selenocysteine . One plausible explanation for the greater accumulation of selenomethionine versus selenocysteine is that selenocysteine incorporation into polypeptide chains requires a selenocysteine-specific tRNA, whereas selenomethionine can substitute for methionine during protein synthesis .
With regard to the effect of dietary OSe versus ISe on Se bioavailability and biopotency in cattle, the milk of cows fed high selenomethionine-containing yeast contains more Se than cattle fed equal amounts of sodium selenite . In beef cattle, it was found that the bioavailability (defined as blood and tissue Se concentrations) of Se from OSe was higher (approximately 200%) than that from ISe [14, 15, 16, 17], as was biopotency (blood glutathione peroxidase activity) [15, 16, 18].
In terms of production responses of dietary OSe versus ISe, the majority of studies [15, 16, 19] did not detect any difference between ISe and OSe forms in beef production performance measurements, such as average daily gain, average daily feed intake, or gain:feed ratio. However, some improvements have been reported in the average daily gain of OSe-supplemented feedlot steers  and calves born from OSe-supplemented cows . Therefore, OSe was recommended to replace ISe for common beef production practices .
In contrast to the number of studies assessing the effects of OSe versus ISe supplementation of beef cattle diets on whole-animal parameters, little research has been conducted to delineate the effect of dietary Se supplementation on gene expression, despite the understanding that form of Se supplementation is thought to mediate different metabolic pathways and predominant forms of Se in liver tissue [2, 5, 8, 11, 21]. Comparison of hepatic gene expression profiles of ISe- versus OSe-supplemented cattle may provide understandings that have critical commercial consequences, including development of supplementation regimens that contain the optimal forms of Se supplementation for a given physiological challenge.
We hypothesized that dietary supplementation of OSe or ISe to a practical diet for maturing beef heifers, that had adequate hepatic Se levels, would differentially affect both the amount of Se in blood and liver and gene expression profiles of liver tissue. The specific objectives of this study were to test the effect of ISe (sodium selenite) versus OSe (Sel-Plex®) on (1) Se concentrations in blood fractions and liver tissue and (2) liver mRNA expression profiles of mature Angus heifers gaining about 0.5 kg/day in body weight.
Materials and Methods
Animal Trial Procedure
Formulation (As-Fed Basis) and Selected Nutrient Composition (Dry Matter Basis) of the Basal Diet Fed to Growing Heifers
68.00 ± 2.07
1.51 ± 0.09
0.91 ± 0.08
11.95 ± 1.80
47.74 ± 5.96
62.65 ± 7.99
0.05 ± 0.01
After a 75-day adaptation period to the feeding system and diet, shrunk body weights were determined and heifers (393 ± 9 kg) were randomly assigned (n = 10/treatment) to one of three dietary Se supplementation treatments. For the control treatment (control), no exogenous source of Se was supplemented to the basal mineral–vitamin premix. For the ISe and OSe treatments, the basal mineral–vitamin premix was supplemented with sodium selenite (ISe; Prince Se Concentrate; Prince Agri Products, Inc., Quincy, IL, USA) or Se-enriched yeast (OSe; Sel-Plex®; Alltech, Inc., Nicholasville, KY, USA), respectively. The concentrations of Se in the premix (mean ± SD, n = 4, as-fed basis) were 34.9 ± 2.27 mg/kg for ISe treatment and 33.5 ± 4.10 mg/kg for OSe treatment, as determined by chemical analysis (Dairy One Cooperative). The daily supply of Se from either source was calculated to provide 3.0 mg/head.
After 80 days on dietary treatments, the daily supply of the basal diet was adjusted to 8.2 kg/head to compensate for the increased animal maintenance requirement and a continuous growth rate of 0.5 kg/day. However, the daily supply of mineral–vitamin premix and Se was not changed. Water was freely available throughout the trial.
Because of the feed mill capacity, the basal diet was prepared four to five times each month with the same formula and ingredients, and samples of the diet were collected at each preparation. After 105 or 106 days on dietary treatments, samples of blood and liver tissue were collected from each animal. Each animal was temporarily restrained in a squeeze chute and the head restrained by halter immediately before sample collection. For blood samples, the right neck area was clipped, cleansed with 70% ethyl alcohol solution, and dried. Vacutainer tubes with and without lithium heparin were used to collect blood by jugular venipuncture for whole blood, red blood cells, and plasma, respectively. Samples were placed on ice immediately after collection and then transferred to the laboratory. The red blood cells and plasma were separated and recovered after centrifugation at 3,000 × g for 10 min at 4°C. For each animal, whole blood, red blood cell, and plasma samples were aliquoted into 5 mL cryogenic vials and stored frozen (−20°C). Liver tissue was collected by the aspiration biopsy technique modified from Erwin et al. . Briefly, the area from the 10th to the 12th intercostal spaces and 10 to 30 cm from the dorsal median plane on the right side of the animal was clipped free of hair, cleansed with povidone–iodine and two subsequent 70% ethyl alcohol solution washes. The remaining 70% ethyl alcohol solution was dried with medical gauze. Lidocaine (1.6 mL 2% injectible solution per biopsy site; The Butler Company, Dublin, OH, USA) was subcutaneously injected between the 11th and 12th ribs approximately 10 cm from the dorsal medial plane. A topical anesthetic spray (Cetacaine, 300 mg; Cetylite Industries, Pennsauken, NJ, USA) was administered to the skin 20 cm from the dorsal median plane at the 11th intercostal space for 2 s. The remaining spray was dried with gauze and an incision made with a scalpel. A trocar (7 mm diameter) was used to obtain liver tissue. The collected tissue was weighed and separated for RNA extraction (approximately 400 mg wet tissue) and Se analysis (approximately 1 g wet tissue). Samples were placed into aluminum foil packs and snap frozen in liquid nitrogen and transferred to a −80°C freezer for storage.
Animal Performance and Tissue Selenium Analysis
Full and shrunk body weights of individual heifers were taken at the initiation and conclusion of the dietary treatment period (a total of 115 days). During the treatment period, the full body weight of each heifer was measured at the end of each month to monitor animal growth rate. Animals consumed all offered feed. The average daily feed intake and average daily gain of each animal was calculated for the entire dietary treatment period and used to determine the effect of dietary Se supplementation on animal growth performance measurements.
The selected nutrient compositions of the basal diet and the mineral–vitamin premix were analyzed by the Forage Laboratory Services, Dairy One Cooperative Inc. (Ithaca, NY, USA) using commercial standard analytical methods based on Association of Official Analytical Chemists  and National Research Council [24, 25]. Blood hematocrit was determined by the method of capillary centrifugation using the AutocritUltra3 centrifuge (Clay Adams, Parsippany, NJ, USA). Total Se concentrations of blood, red blood cells, plasma, and liver tissue were analyzed  at the animal nutrition laboratory of Dr. D. C. Mahan, Department of Animal Science, The Ohio State University according to the fluorometric method of Association of Official Analytical Chemists  with an ultraviolet detector (Turner filter, fluorometer, model 112; Unipath, Mountain View, CA, USA).
Microarray Analysis of Gene Expression
Six animals were randomly selected from each of three treatment groups for which a complete set of blood, red blood cell, plasma, and liver tissue samples had been taken for Se analysis. For each animal, total RNA was extracted from 300 to 400 mg of frozen liver tissue using TRIzol Reagent (Invitrogen Corporation, Carlsbad, CA, USA) following the manufacturer’s instructions. Briefly, frozen tissue was homogenized in a 15-mL polypropylene centrifuge tube using a Polytron mixer (1 mL TRIzol per 75 mg tissue) and the homogenate transferred to several 1.5-mL microcentrifuge tubes (1 mL/tube). Chloroform (200 µL/tube) was used to separate RNA from proteins and DNA, and then RNA was precipitated with isopropyl alcohol (600 µL/tube) and washed with 1 mL of 75% ethanol. The resulting RNA pellet was air-dried, dissolved in 100-µL RNase-free water, and stored at a −80°C.
The purity and concentration of total RNA samples was analyzed by a NanoDrop ND-1000 Spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA), which revealed that all samples were of high purity with 260/280 absorbance ratios greater than 1.92 and 260/230 absorbance ratios ranging from 1.4 to 1.8. The integrity of total RNA was examined by gel electrophoresis using Agilent 2100 Bioanalyzer System (Agilent Technologies, Santa Clara, CA, USA) at the University of Kentucky Microarray Core Facility. Visualization of gel images and electropherograms showed that all RNA samples had high quality with RNA integrity numbers being greater than 8.2 and 28S/18S rRNA absorbance ratios greater than 1.8.
The GeneChip Bovine Genome Array (Affymetrix, Inc., Santa Clara, CA, USA) was used to test the effect of Se supplementation on the expression of 24,016 bovine gene transcripts (as probe sets on the GeneChip). Microarray analysis was conducted according to manufacturer’s standard protocol (Affymetrix) by the University of Kentucky Microarray Core Facility. Briefly, all the RNA samples were first reverse-transcribed to cDNA, and then from cDNA (double-stranded) to cRNA (single-stranded) which were labeled with biotin. The biotinylated cRNA were further fragmented and then used as probes to hybridize the gene chips in the GeneChip Hybridization Oven 640, using one chip per RNA sample. After hybridization, the chips were washed and stained on a GeneChip Fluidics Station 450. The reaction image and signals were read with a GeneChip Scanner (GCS 3000, 7G) and data collected using the GeneChip Operating Software (GCOS, version 1.2). The raw expression intensity values from the software (i.e., the *.cel files from the raw methylation measurements) were imported into Partek Genomics Suite software (version 6.4; Partek Inc., St. Louis, MO, USA). For GeneChip background correction, the algorithm of Robust Multichip Averaging adjusted with probe length and GC oligo content was implemented [27, 28]. The background-corrected data were further converted into expression values using quantile normalization for each chip and then Median Polish summarization of multiple probes .
All GeneChip transcripts were annotated using NetAffx annotation database for 3’ IVT Expression on Bovine GeneChip Array, provided online by the manufacturer (http://www.affymetrix.com/analysis/index.affx, Affymetrix, Inc.). The dietary treatment-induced effects on the expression of all the transcripts were subjected to statistical analysis (described below). Transcripts showing treatment-induced effects at the statistical significance level of P < 0.01 were selected for Core Analysis using the Ingenuity Pathways Analysis online software (Ingenuity Pathway Analysis; 8.0-2602; http://www.ingenuity.com; Ingenuity Systems, Inc., Redwood City, CA, USA).
Treatment effects on animal performance data (average daily feed intake, average daily gain, and gain:feed ratio) and Se concentrations of whole blood, red blood cells, plasma, and liver tissue were all subjected to ANOVA analysis for a completely randomized experimental design using the General Linear Means procedure of SAS (SAS Inst., Inc., Cary, NC, USA). When a treatment effect was observed (P < 0.01 for ANOVA F test), the least squares means were separated by Fisher’s least square means test, with P < 0.05 considered as statistically different. The unbalanced experimental units (n = 6, 8, or 9) used for statistical analysis resulted from the loss of one animal and one to three samples during the animal trial, sample collection (as in liver biopsy and venous bleeding), and laboratory analysis.
Treatment effects on the relative expression levels of mRNA obtained from the microarray analysis for each gene transcript in liver tissue were also subjected to ANOVA testing using the Partek Genomics Suite software  for a completely randomized experimental design. For a higher degree of confidence (i.e., a conservative approach), when the ANOVA test gave a P value less than 0.01, treatment means were compared by two contrasts (ISe versus Control; OSe versus Control), and the contrast P values less than 0.05 were considered significant.
Results and Discussion
Experiment Model and Animal Performance
Formulation of the practical basal diet fed to the experimental heifers and its nutrient compositions (selected) are presented in Table 1. Dietary concentration of intrinsic Se was 0.05 mg/kg. Therefore, the basal diet plus the basal mineral–vitamin premix (described in “Materials and Methods” section) provided 0.48 mg Se daily per animal. Because animals consumed all offered feed, each control animal received 0.48 mg Se daily, whereas each ISe or OSe-supplemented (3.0 mg Se/day) animal received 3.48 mg Se daily. According to the National Research Council  standard, the Se requirement of beef cattle can be met by providing 0.1 mg/kg in the complete diet although Se can be legally supplemented to provide 0.3 mg/kg. The Se concentration of control treatment animals in the basal diet mixed with the basal mineral–vitamin premix used in this study would be 0.06 mg/kg, which was not adequate to meet the National Research Council  standard. Because this practical diet is common in the southeastern states, the results generated from this study provide relevant Se supplementation information to beef cattle nutritionists and producers.
Effect of Dietary Selenium (Se) Supplementation on Animal Performance
Although the dietary ISe or OSe supplementation had a tendency (P = 0.13) of improvement in gain:feed ratio, animal growth performance was not affected by Se supplementation (Table 2). This “non-affect” observed for slow-growing heifers essentially is in agreement with most other studies that compared effects of OSe and ISe supplementation of beef cattle diets in a variety of cattle models. Nicholson et al. , Gunter et al. , Richards and Loveday , and Davis et al.  did not detect any difference between ISe and OSe forms in beef production performance, such as in average daily gain, average daily feed intake, or gain:feed. Nevertheless, some improvements have been reported in average daily gain of OSe-supplemented feedlot steers  and calves born from OSe-supplemented cows . Given these improvements in average daily gain, plus the tendency of improvement in gain:feed ratio detected in this study, whether the form of dietary supplementation of Se affects animal growth performance is unclear.
Tissue Selenium Concentration
Effect of Dietary Selenium (Se) Supplementation on Tissue Se Concentration
Whole blood (µg/mL)
Red blood cells (µg/mL)
Liver Se content (Table 3) was increased (P < 0.05) 31% and 81% with ISe and OSe supplementation, respectively, with the Se content of OSe heifers being 38% greater (P < 0.0001) than ISe heifers.
The finding of increased Se concentrations in whole blood and liver tissue with Se supplementation (at 3 mg/day) in this study is in agreement with previous reports [15, 16, 17, 31]. For example, Juniper et al.  reported greater Se concentrations in whole blood, liver, kidney, heart, and skeletal muscle in 20-month-old beef cattle supplemented with OSe (Sel-Plex®), compared to the cattle supplemented with ISe (sodium selenite) or no Se supplementation. From our results, as well as those in the literature, it can be concluded that a greater red blood cell and liver Se bioavailability results when Se-deficient cattle diets are supplemented with OSe (Sel-Plex®) versus ISe (sodium selenite).
The most effective way to determine cattle Se status is the analysis of blood or liver Se concentration . According to Dargatz and Ross , whole blood Se concentrations of greater than 0.16 μg/mL are considered to be highly adequate. According to Surai , the liver Se concentration of 0.25 to 0.50 μg/g is considered to be adequate. According to the Food and Drug Administration standard, liver Se concentrations of 0.10 to 1.20 μg/g are considered “normal” (cited in ). In this study, the whole blood and liver Se concentrations of control heifers were 0.17 μg/mL and 0.26 μg/g, respectively (Table 3), and thus, the Se status of these heifers was at an adequate or highly adequate level. Given that none of the heifers were supplemented with Se for 75 days before Se treatments began, the control animals continued to receive no Se supplementation for the entire treatment period (105 or 106 days) and that the basal diet plus the basal mineral–vitamin premix (i.e., the control treatment diet) provided only 0.48 mg Se daily per animal, it is likely that the adequate Se status of the control heifers indicates a carryover effect from prior dietary management.
Genes Affected by Dietary Selenium Supplementation
Genomic microarray analysis has been conducted on a variety of animal tissues affected by diverse pathogenic or nutritional factors in an effort to document treatment effects on tissue transcriptome profiles. Nutrigenomic approaches using microarray technology have been employed to study Se nutrition. It has been established that dietary supplementation of selenomethionine can affect gene expression in mouse intestinal tissues by upregulating genes associated with cellular stress and cell cycling and downregulating genes associated with selenoprotein production, lipid transport, and detoxification mechanisms . Another study comparing the effect of Se source, including sodium selenite (1.0 mg Se/kg diet) and Se-enriched yeast (1.0 mg Se/kg diet), on gene expression profile of mouse intestinal tissue found that nearly 11% of the 23,000 gene transcripts were influenced, with at least 100 being associated with animal reproductive functions (cited in ). However, there is a lack of data regarding the roles of different chemical forms of Se on the regulation of gene expression profile in beef cattle.
In this study, we employed Bovine GeneChip Genome Array (Affymetrix) that contains 24,016 bovine gene transcripts, representing approximately 19,000 UniGene Clusters (UniGene Build 57, March 24, 2004), to investigate the potential changes of bovine liver gene expression upon dietary supplementation of ISe (sodium selenite) and OSe (Sel-Plex®). All the microarray raw data (18 *.cel files) collected with the GCOS software plus the GCRMA-processed data analyzed with the Partek Genomics Suite  have been deposited into the Gene Expression Omnibus (National Center for Biotechnology Information, http://www.ncbi.nlm.nih.gov/geo) as accession number GSE19696.
Genes or Gene Transcripts in Liver of Growing Beef Heifers Affected by Dietary Selenium (Se) Supplementation, Relative to Nonsupplemented Control Animals
Probe set ID
Common description of the gene
ARP8 actin-related protein 8 homolog (yeast)
ADP-ribosylation factor-like 6 interacting protein 2
Antizyme inhibitor 1
Basic helix-loop-helix domain containing, class B, 2
Basic helix-loop-helix domain containing, class B, 2
Calcium-activated nucleotidase 1
Calcium-regulated heat stable protein 1, 24 kDa
Coiled-coil and C2 domain containing 1B
Coiled-coil domain containing 49
Coiled-coil domain containing 88C
Cytochrome c oxidase subunit VIIa polypeptide 2 like
Dehydrogenase/reductase (SDR family) member 4
F-box and WD repeat domain containing 2
Golgi-specific brefeldin A resistant guanine nucleotide exchange factor 1
Glutamate-cysteine ligase, modifier subunit
Histocompatibility (minor) HA-1
Insulin-like growth factor 2 (somatomedin A)
Insulin-like growth factor-binding protein 3
Kruppel-like factor 10
Kruppel-like factor 11
Leucine-rich repeat LGI family, member 4
Leucine-rich repeat LGI family, member 4
Similar to hepatocyte growth factor activator
Similar to cytochrome b5 domain containing 2
Similar to putative eukaryotic translation initiation factor 3 subunit (eIF-3)
Similar to programmed cell death 10 (PDCD10)
Mediator complex subunit 22
Hypothetical protein LOC616423
Fanconi anemia-associated protein, 24 kDa
Mitochondrial transcription termination factor
NCK (Src homology 3 and 2 domain containing) adaptor protein 2
Procollagen-proline, 2-oxoglutarate 4-dioxygenase (proline 4-hydroxylase), alpha
Protein kinase C and casein kinase substrate in neurons 2
Protein kinase C, iota
Protein tyrosine phosphatase, nonreceptor type 23
RNA binding motif (RNP1, RRM) protein 3
Regulator of G-protein signaling 9
RNA pseudouridylate synthase domain containing 3
Sec61 alpha 2 subunit (Saccharomyces cerevisiae)
Suppression of tumorigenicity 5
StAR-related lipid transfer (START) domain containing 5
Serine/threonine kinase 3 (STE20 homolog, yeast)
Vacuolar protein sorting 28 homolog (Saccharomyces cerevisiae)
WD repeat domain 46
WD repeat domain 79
Tryptophan-rich basic protein
Zinc finger, CCHC domain containing 8
Number of Transcripts Affected by Dietary Selenium (Se) Supplementation, Relative to Nonsupplemented Control Animals
Besides these commonly affected gene transcripts, there were 26 transcripts that were solely altered by ISe but not OSe supplementation with 13 being upregulated and 13 being downregulated (Tables 4 and 5). The expression changes of these transcripts appear to be ISe-dependent (sodium selenite-dependent). There were also 23 gene transcripts that were solely altered by OSe but not ISe supplementation with 14 being upregulated and nine being downregulated, and apparently represent OSe-dependent (Sel-Plex-dependent) genes.
Functional analysis of the 80 Se-affected gene transcripts in Table 4 revealed that most transcripts are very important in that they are associated with a broad range of biological processes including (1) nutrient metabolism, (2) cellular growth, proliferation, and immune response, (3) cell communication or signaling, and (4) development and functions of various tissues and organs. For example, the expression of insulin-like growth factor binding protein 3 (IGFBP3) mRNA was increased by 30% and 36% upon ISe and OSe supplementation, respectively. Insulin-like growth factor (IGF) cell signaling pathway comprises a complex set of molecules including three polypeptide hormones (insulin, IGF1, IGF2), three receptors, and six binding proteins . This signaling pathway regulates a diverse array of physiological functions including nutrient metabolism [35, 36, 37], cellular proliferation [38, 39], and tissue/organ development [40, 41]. As a member of IGFBP family, IGFBP3 forms a ternary complex with IGF and acid-labile subunit in the liver (a central organ in the GF-IGF axis). In this form, it circulates in the plasma, prolonging the half-life of IGF and altering their interaction with cell surface receptors . Despite uncertainty about the mechanism involved, enhancement of IGF action by IGFBP3 appears to be a widely accepted phenomenon. The in vivo studies with hypophysectomized rats showed that coadministration of IGF1 with IGFBP3 was more effective than even a larger dose of IGF1 alone in eliciting weight gain and an increase in epiphysial width . Recent IGFBP3 biological studies showed that IGFBP3, as a novel effector molecule, is not just another “binding protein” because it also exerts other IGF-independent actions on animal metabolism and cell growth [44, 45]. The upregulation of IGFBP3 mRNA by ISe and OSe supplementations may explain the tendency of improvement of gain:feed ratio observed in this study.
IGF2 mRNA expression also was increased by 22% by OSe supplementation, but not by ISe supplementation (Table 4). It has been reported that human-IGF2 protein increased IGFBP3 mRNA expression in cultured primary astroglia from cerebral cortex of newborn rat  and lymphoid tissues of transgenic mice . IGF2 also can stimulate the healing of human wounds in at least the presence of recombinant IGFBP1 . The upregulation of both IGFBP3 and IGF2 mRNA by OSe supplementation observed in this study is coincident with the reported enhanced growth performance of beef cattle when fed OSe versus ISe [18, 20].
Mitochondrial transcription termination factor (MTERF) mRNA content was increased by 16% and 30% with ISe and OSe supplementation, respectively (Table 4). In mammals, MTERF gene product has three leucine zipper motifs that participate in intra-molecular interactions to establish the three-dimensional structure required for DNA binding . The binding of mitochondrial DNA at a region immediately downstream of the 16S rRNA gene can terminate the ribosomal transcription of mitochondrial DNA and therefore, provide a system for coordinating the passage of replication and transcription complexes [48, 49]. The upregulation of MTERF mRNA found in this study suggests that dietary supplementation of Se for growing heifers may safeguard the integrity of mitochondrial genome while facilitating its efficient expression.
Besides the 17 upregulated gene transcripts, 13 transcripts were downregulated by both ISe and OSe supplementations, with eight being annotated (Tables 4 and 5). Two annotated transcripts, Bt.22344.1.S1_at and Bt.23123.1.S1_at, are derived from a single gene, BHLHB2 (basic helix-loop-helix domain containing, class B, 2; a.k.a. BHLHE40 or Dec2) and were downregulated 67% to 39% or 100% to 67% by either ISe or OSe supplementation, respectively. As a transcription repressor, BHLHB2 plays a pivotal role in multiple cell signaling pathways that affects many biological processes, including tissue development, cell differentiation/growth, cell death, oncogenesis, immune systems, circadian rhythm, and homeostasis . In terms of circadian rhythms, BHLHB2 functions as one of the clock-related genes  and may mediate synaptic function and neuronal excitability . Fujimoto et al.  suggested that BHLHB2 employs multiple molecular mechanisms including DNA binding and protein–protein interactions to achieve its role via E-box-dependent transcriptional repressions. The physiological significance of the downregulation of BHLHB2 mRNA expression by Se supplementation observed in this study is unknown.
Genes Kruppel-like factor 10 (KLF10) and KLF11 are two of the 17 members in the family of Kruppel-like transcription factors, which are gene regulatory proteins implicated in many biological processes. In particular, KLF10 is known to induce and repress the expression of multiple genes in several cell types and to function as an inhibitor of cell proliferation and an inducer of apoptosis . The correlation between levels of KLF10 protein and stage of breast cancer suggested that KLF10 may play a major role as a tumor suppressor for a variety of cancers . Similar to KLF10 gene, KLF11 gene encodes a nuclear protein which also regulates multiple gene expression and inhibits cell proliferation. In some cases, KLF10 and KLF11 can regulate the same genes . However, KLF10 knockout mice display defects in bone and skeletal muscle development, whereas KLF11 knockout mice appear phenotypically normal, with no apparent impact on viability, development, fertility, or lifespan . The physiological significance of the downregulation of KLF10 and KLF11 mRNA in both ISe- and OSe-supplemented heifers warrants further investigation.
The product of collectin-43 (CL-43) gene is a serum conglutinin protein synthesized mainly in the liver of only Bovidae animals . Involved in the innate immune defense, this conglutinin protein has dual function; CL-43 binds specifically to carbohydrate structures on the surface of a pathogen, and subsequently, recruits other cells and molecules to destroy the pathogen and impede pathogen infectivity [59, 60]. Within the ISe-dependent gene group identified in this study, CL-43 mRNA was downregulated by 60% (Table 4). We do not have evidence from this study to make connection between the downregulation of CL-43 mRNA in the liver and the reduced antiviral capacity of beef cattle. However, further investigation of this gene may be of interest because it would facilitate our understanding of the difference between Bovidae animals and humans in terms of innate anti-microbial immunity and between OSe versus ISe sources of supplemental Se.
As a cofactor of glutathione peroxidases and glutathione S-transferases, glutathione (γ-glutamyl-cysteinyl-glycine) is a cysteine-containing tripeptide with reducing and nucleophilic properties that play a key role in cellular detoxification or protection from oxidative damage of lipids, proteins, and nucleic acids . Glutamate-cysteine ligase (GCL; γ-glutamylcysteine synthetase) is the first rate-limiting enzyme in GSH biosynthesis and is composed of catalytic (GCLC) and modifier (GCLM) subunits which are encoded by different genes. While GCLC catalyzes a unique γ-carboxyl linkage from glutamate to cysteine, GCLM increases the Vmax and Kcat of GCLC, decreases the Km for glutamate, and increases the Ki for GSH-mediated feedback inhibition of GCL . Within the OSe-dependent gene group identified in this study, GCLM mRNA was downregulated by 16% (Table 4). Whether this downregulation results in more free hepatic cysteine, less GSH, or affects basal hepatic selenomethionine:selenocysteine ratio  remains to be determined. However, decreased levels of GCLM mRNA may be indicative of a lesser state of oxidative stress [63, 64].
Protein kinase C iota (PRKCI) is a member of the atypical protein kinase C holoenzyme family, and under normal physiological conditions, is involved in the cell survival and proliferation and the formation of asymmetrical epithelial cell junction structures, which establishes cell polarity . However, PRKCI is a known human oncogene, presumably through the onogenic Ras signaling pathway . Over-expression of PRKCI mRNA and protein is common in cancerous tumors, whereas inhibition of PKRCI expression reduces migration and invasion of cancerous cells . In human liver tissue, expression of PRKCI is positively correlated with development of hepatocellular carcinomas . This pattern of PRKCI expression was accompanied by a parallel increase in β-catenin expression, but decreased expression of E-cadherin. These latter findings are consistent with tumor invasion and metastasis . In the current study, supplementing Se as OSe reduced hepatic PRKCI mRNA expression by 73% relative to control animals, whereas ISe treatment did not affect expression. Because the animals displayed no signs of sickness, all liver biopsy samples appeared normal, and no alteration in β-catenin or E-cadherin was observed, the significance of this finding is unknown. However, given the decreased PRKCI mRNA content in the livers of heifers fed the OSe supplement, and the positive correlation between increased PRKCI expression and hepatic carcinomas, and although speculative, this finding may be of especial value.
This pilot study has identified genes individually or commonly affected by dietary supplementation of Se in the form of ISe (sodium selenite) or OSe (Sel-Plex®) to growing beef heifers that were of adequate Se status. Functional analysis of these genes revealed that a broad range of physiological functions might be affected by Se supplementation. Because there are differences in tissue Se concentrations and liver gene expression profiles between ISe and OSe supplementations, any dietary recommendation for the same level of supplementation of both forms should be questioned. Identification of key genes or biomarkers for different forms of Se can facilitate our designing of form-specific supplementation regimen for improved beef cattle health and productivity. However, further studies are required to resolve the key metabolic and signaling pathways that specifically underlie ISe and OSe functions. Only with a thorough understanding of the relationship between gene expression profile and Se form and concentration in diets, recommendations that delineate the concentration and length of time for ISe and OSe forms in diets can be given to beef cattle producers. In short, analysis of the dataset generated from this study provides the proof-of-concept that supplementing cattle diets with ISe (sodium selenite) or OSe (Sel-Plex®) at a same level (3 mg/day) results in different hepatic gene expression profiles, albeit with a common change in a core set of genes commonly affected by both forms of supplemental Se.
The data generated from this study demonstrated that more Se was retained in whole blood, red blood cells, and liver tissues, but not in plasma, of growing beef heifers upon dietary Se supplementation. Moreover, more Se was retained in these measured tissues (except for plasma) when Se was supplemented in OSe than ISe form. The study of liver gene expression profile by microarray analysis indicated the chemical form (inorganic versus organic) of supplemental Se matters. Functional analysis of these affected genes showed that they are associated with a broad range of physiological functions including (1) nutrient metabolism, (2) cellular growth, proliferation, and immune response, (3) cell communication or signaling, and (4) tissue/organ development and function. Regarding the effect of the supplemental Se forms, three distinct groups of genes were identified: those (1) commonly affected by ISe and OSe supplementation, (2) solely affected by ISe supplementation, and (3) solely affected by OSe supplementation. More specifically, both forms of supplementation appeared to upregulate mitochondrial gene expression capacity, whereas gene expression of a protein involved in antiviral capacity was downregulated in ISe-supplemented animals, and OSe-supplemented animals had reduced levels of mRNA encoding proteins known to be upregulated during oxidative stress and cancerous states.
- 2.Pehrson BG (1993) Selenium in nutrition with special reference to the biopotency of organic and inorganic selenium compounds. In: Lyons TP (ed) Biotechnology in the Feed Industry: Proc Alltech's 9th Annu Symp. Nottingham University Press, Nottingham, pp 71–89Google Scholar
- 3.NRC (1996) Energy. In: Nutrient requirements of beef cattle, 7th rev edn. National Academy Press, Washington, DC, pp 3–15Google Scholar
- 4.FDA (2003) Code of Federal Regulations: Title 21, Food and Drugs. Available via USGPO. http://www.access.gpo.gov/cgi-bin/cfrassemble.cgi?title=200321. Accessed 12 Nov 2009
- 5.Weiss WP (2003) Selenium nutrition of dairy cows: comparing responses to organic and inorganic selenium forms. In: Lyons TP, Jacques KA (eds) Nutritional biotechnology in the feed and food industries: Proc Alltech's 19th Annu Symp. Nottingham University Press, Nottingham, pp 333–343Google Scholar
- 6.FDA (2007) Food additives permitted in feed and drinking water of animals; Selenium yeast (FR Doc. E7-13954). Fed Regist 72:39560–39562Google Scholar
- 8.Surai PF (2006) Selenium in ruminant nutrition. In: Surai PF (ed) Selenium in nutrition and health. Nottingham University Press, Nottingham, pp 487–587Google Scholar
- 12.Surai PF (2006) Molecular mechanisms of Se action: selenoproteins. In: Surai PF (ed) Selenium in nutrition and health. Nottingham University Press, Nottingham, pp 47–49Google Scholar
- 14.Pehrson B, Knutsson M, Gyllensward M (1989) Glutathione peroxidase activity in heifers fed diets supplemented with organic and inorganic selenium compounds. Swed J Agric Res 19:53–56Google Scholar
- 18.Guyot et al (2006) Comparative responses to sodium selenite and Sel-Plex® organic selenium supplements in Belgian Blue cows and calves. In: Lyons TP, Jacques KA, Hower JM (eds) Nutritional Biotechnology in the Feed and Food Industries: Proc Alltech’s 22nd Annu Symp. Nottingham University Press, Nottingham, pp 331–340Google Scholar
- 19.Davis PA, McDowell LR, Van Alstyne R et al (2008) Effects of form of parenteral or dietary selenium supplementation on body weight and blood, liver, and milk concentrations in beef cows. Prof Anim Sci 24:52–59Google Scholar
- 20.Clyburn BS, Richardson CR, Montgomery JL et al (2001) Effect of selenium source and vitamin E level on performance and meat quality of feedlot steers. In: Lyons TP, Jacques KA (eds) Science and technology in the feed industry: Proc. Alltech's 17th Annu Symp. Nottingham University Press, Nottingham, pp 377–392Google Scholar
- 21.Schrauzer GN, Surai PF (2009) Selenium in human and animal nutrition: resolved and unresolved issues. A partly historical treatise in commemoration of the fiftieth anniversary of the discovery of the biological essentiality of selenium, dedicated to the memory of Klaus Schwarz (1914–1978) on the occasion of the thirtieth anniversary of his death. Crit Rev Biotechnol 29:2–9PubMedCrossRefGoogle Scholar
- 22.Erwin ES, Dyer IA, Meyer TO et al (1956) Uses of aspiration biopsy technique. J Anim Sci 15:428–434Google Scholar
- 23.AOAC (2000) Official methods of analysis, 17th edn. Association of Official Analytical Chemists, ArlingtonGoogle Scholar
- 24.NRC (1996) Energy. In: Nutrient requirements of beef cattle, 7th rev. ed. National Academy Press, Washington, DC, pp 3–15Google Scholar
- 25.NRC (2001) Energy. In: Nutrient requirements of dairy cattle, 7th rev. ed. National Academy Press, Washington, DC, pp 13–27Google Scholar
- 29.Partek D (2009) Partek documentation: turning data into discovery. Partek Incorporated, St. LouisGoogle Scholar
- 30.Howell JA, Matthews AD, Welbourne TC et al (2003) Content of ileal EAAC1 and hepatic GLT-1 high-affinity glutamate transporters is increased in growing versus non-growing lambs, paralleling increased tissue concentrations of D- and L-glutamate and plasma glutamine and alanine. J Anim Sci 81:1030–1039PubMedGoogle Scholar
- 31.Richards CJ, Loveday HD (2004) Redefining selenium nutrition using organic selenium (Sel-Plex®): defining maximal acceptable tissue residues in beef. In: Lyons TP, Jacques KA (eds) Nutritional biotechnology in the feed and food industries: Proc. Alltech's 20th Annu Symp. Nottingham University Press, Nottingham, pp 211–220Google Scholar
- 61.Gérard-Monnier D, Chaudiere J (1996) Metabolism and antioxidant function of glutathione [in French]. Pathol Biol (Paris) 44:77–85Google Scholar
- 65.Suzuki A, Akimoto K, Ohno S (2003) Protein kinase Cλ/ι (PKCλ/ιa PKC isotype essential for the development of multicellular organisms. J Biol Chem 133:9–16Google Scholar