ABSTRACT
Selenium is an essential trace mineral that serves as an antioxidant, and aids in both immune function as well as thyroid hormone metabolism. The objective of this research was to evaluate the effects of Se depletion and repletion on whole blood Se concentrations and erythrocyte glutathione peroxidase (RBC GSH-Px) activity. Ten geldings received 23% of the NRC’s recommended daily Se intake during the 112-d depletion phase. After depletion, horses were stratified by whole blood Se concentrations and evenly divided into 2 groups of 5, and assigned to 1 of 2 treatments: 0.1 ppm organic Se (SE1) and 0.3 ppm organic Se (SE3). During repletion, horses were fed their respective diets for 112 d. Venous blood was collected at d 0, 28, 56, 84, and 112 of depletion, and d 14, 28, 56, 84, 96, and 112 of repletion. Whole blood Se concentrations and RBC GSH-Px activity were analyzed. Non-linear regression curves for whole blood Se concentrations were developed for the depletion phase as well as both treatments during the repletion phase. The curve of the regression equation, during the repletion phase, were compared and were not significantly different. Whole blood Se concentrations were compared using t-tests, and were significantly greater in horses receiving SE3 at d 14, 28, 56, 84, 96, and 112 as compared to horses consuming SE1. Due to the large variation in RBC GSH-Px activity, non-linear regression curves could not be developed, and there were no significant differences between treatments within ii time throughout the repletion phase. A possible explanation for the wide variation observed in RBC GSH-Px activity is the handling and storage of blood samples, as this enzyme is very sensitive to temperature, especially during centrifugation. Results from this study indicate that feeding Se above that of the NRC recommendation to previously depleted horses may be beneficial, however never reached original values in moderatelyexercised horses. Key words: Selenium, whole blood, glutathione peroxidase activity, depletion, repletion, horse iii ACKNOWLEDGMENTS Numerous people have helped me in my pursuit of receiving a Doctorate of Philosophy degree. First, I would like to thank my mom and dad. They have shown me the value a strong work ethic has, furthermore they have instilled the importance and value a quality education has in assisting me to achieve my life goals. Throughout this journey, they have been a constant source of support. Without them, I would not have the bright future which lays ahead of me. To Roger Nonella, my husband, you kept me calm and was a faithful listener throughout this experience. You never complained about feeding for me or assisting me during my collections. I am very thankful for your support, and that I was able to share this experience with you. Thank you to my committee members; Dr. Lance Baker, Dr. John Pipkin, Dr. David Parker, Dr. Mallory Vestal, and Dr. Marty Rhoades. Dr. Baker, you helped me to gain a deeper understanding of my research, and helped to ease my nerves when my research did not go as anticipated. Dr. Pipkin, you have helped me grow into a better person, people manager, and have been a consistent line of communication. Dr. Parker, thank you for allowing me to use the CORE laboratory to prepare and store samples. Dr. Vestal, you have helped to guide me to look at the business side. Dr. Rhoades, thank you iv for aiding in the statistical analysis of my research. This project would not have been possible without the funding from Horse Guard, Inc. and Killgore Research Grant. v Approved: __________________________________________ ____________ Chair, Thesis Committee Date __________________________________________ ____________ Member, Thesis Committee Date __________________________________________ ____________ Member, Thesis Committee Date _________________________________________ ____________ Member, Thesis Committee Date __________________________________________ ____________ Member, Thesis Committee Date ______________________________ ____________ Head, Major Department Date ______________________________ ____________ Dean, Graduate School Date vi TABLE OF CONTENTS Page ABSTRACT i ACKNOWLEDGMENTS iii LIST OF TABLES x LIST OF FIGURES xi Chapter I. INTRODUCTION 1 II. LITERATURE REVIEW 4 Selenium Functions 4 Selenium in Soil and Forage 6 Selenium Absorption, Metabolism, and Storage 7 Selenium Absorption- Organic versus Inorganic 10 Selenium- Injectable 11 Selenium and Glutathione Peroxidase in Blood 12 Selenium Deficiency 14 Selenium Deficiency Economic Impact 15 Selenium Toxicity 16 Selenium Supplementation Environmental Impact 18 vii Selenium in Cattle 19 Selenium in Sheep 21 Selenium in Swine 22 Selenium in Horses 23 Statement of the Problem 47 III. MATERIALS AND METHODS 48 Experimental Design 48 Diets 49 Sample Collections, Preparation, and Handling 50 Laboratory Analysis 51 Inductively-Coupled Plasma Mass Spectrometry 51 Glutathione Peroxidase Activity Assay 51 Statistical Analysis 52 IV. RESULTS AND DISCUSSION 54 Depletion Phase Whole Blood Selenium Concentrations Regression 54 Repletion Phase Whole Blood Selenium Concentrations Regression 58 Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 0 of Repletion 63 Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 14 of Repletion 65 Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 28 of Repletion 67 viii Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 56 of Repletion 69 Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 84 of Repletion 72 Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 96 of Repletion 74 Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 112 of Repletion 76 Depletion Phase Erythrocyte Glutathione Peroxidase Activity Regression 78 Repletion Phase Erythrocyte Glutathione Peroxidase Activity Regression 79 Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 0 of Repletion 80 Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 14 of Repletion 80 Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 28 of Repletion 83 Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 56 of Repletion 85 Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 84 of Repletion 87 ix Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 96 of Repletion 89 Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 112 of Repletion 89 Possible Explanation for Differences in Erythrocyte Glutathione Peroxidase Activity between Studies 92 V. CONCLUSIONS AND IMPLICATIONS 94 LITERATURE CITED 96 APPENDIX FIGURES A, WHOLE BLOOD SELENIUM CONCENTRATIONS GRAPHS 101 APPENDIX FIGURES B, ERYTHROCYTE GLUTATHIONE PEROXIDASE ACTIVITY GRAPHS 105 APPENDIX TABLES A 108 x LIST OF TABLES Page 1. Feed Analysis for Orchard Grass Hay (DM) 50 2. Selenium Analysis for Supplements and Hay (DM) 50 3. Mean Selenium Intake (mg/kg DM) 54 A-1 Individual whole blood selenium concentrations 109 A-2 Individual erythrocyte glutathione peroxidase activity 110 A-3 Individual body weights 111 xi LIST OF FIGURES Page 1. Non-linear regression equation throughout 112-d selenium depletion period (d 0, 28, 56, 84, and 112) 56 2. Non-linear regression equation of selenium depletion (means at d 0, 28, 56, 84, and 112) and forecasted to d 250 57 3. Adjusted whole blood selenium concentrations in horses consuming 0.1 ppm selenium at d 0, 14, 28, 56, 84, 96, and 112 59 4. Adjusted whole blood selenium concentrations in horses consuming 0.3 ppm selenium at d 0, 14, 28, 56, 84, 96, and 112 60 5. Non-linear regressions over 112-d repletion of treatments (0.1 and 0.3 ppm selenium) on adjusted whole blood selenium concentrations 62 6. Overall mean whole blood selenium concentrations at d 0 of selenium repletion 64 7. Overall mean whole blood selenium concentrations at d 14 of selenium repletion 66 xii 8. Overall mean whole blood selenium concentrations at d 28 of selenium repletion 68 9. Overall mean whole blood selenium concentrations at d 56 of selenium repletion 70 10. Overall mean whole blood selenium concentrations at d 84 of selenium repletion 73 11. Overall mean whole blood selenium concentrations at d 96 of selenium repletion 75 12. Overall mean whole blood selenium concentrations at d 112 of selenium repletion 77 13. Overall mean erythrocyte glutathione peroxidase activity at d 0 of selenium repletion 81 14. Overall mean erythrocyte glutathione peroxidase activity at d 14 of selenium repletion 82 15. Overall mean erythrocyte glutathione peroxidase activity at d 28 of selenium repletion 84 16. Overall mean erythrocyte glutathione peroxidase activity at d 56 of selenium repletion 86 17. Overall mean erythrocyte glutathione peroxidase activity at d 84 of selenium repletion 88 18. Overall mean erythrocyte glutathione peroxidase activity at d 96 of selenium repletion 90 xiii 19. Overall mean erythrocyte glutathione peroxidase activity at d 112 of selenium repletion 91 A-1 Individual whole blood selenium concentrations throughout selenium depletion phase 102 A-2 Individual whole blood selenium concentrations throughout selenium repletion phase 103 A-3 Non-linear regression equations throughout 112-d selenium depletion and 112-d selenium repletion 104 B-1 Individual erythrocyte glutathione peroxidase activity throughout selenium depletion phase 106 B-2 Individual erythrocyte glutathione peroxidase activity throughout selenium repletion phase 107 1 Chapter 1 INTRODUCTION Selenium (Se) is an essential trace mineral found in varying amounts in soil, and subsequently plants grown in that soil. Several geographical areas of the United States are notoriously Se deficient, including the Pacific Northwest, Great Lakes Region, and Eastern Seaboard. Therefore, horses consuming feed grown in these areas are subject to becoming Se deficient. The primary function of Se in the body is to serve as an antioxidant, and is a rate-limiting component of the enzyme glutathione peroxidase (GSH-Px). Glutathione peroxidase activity is greatest in erythrocytes (Ullrey, 1987). Selenium is also a vital component of the immune system and thyroid hormone metabolism (Koller and Exon; 1986; Daniels, 1996; Mayer, 2009). Very few studies have reported the effects of Se depletion in horses. The current recommended dietary intake set by the NRC(2007) is 0.1 ppm Se. Brummer et al. (2013) reported horses receiving 0.06 ppm Se had significantly lower whole blood Se concentrations at d 84 of depletion as compared to d 0. Whole blood Se concentrations were significantly lower at d 140 of depletion as compared to d 84. However, there were no significant differences at d 168 or 196 as compared to d 140. Furthermore, the researchers reported significantly lower whole blood GSH-Px activity at d 84 as compared to d 0. Whole blood GSH-Px activity was also significantly lower at d 168 and 196 of depletion as compared to d 84. 2 Previous studies have reported conflicting results about the possible benefits of Se supplementation above the NRC Se recommendation (0.1 ppm), particularly in previously Se depleted horses. Brummer et al. (2013) reported horses consuming 0.3 ppm Se had higher whole blood Se concentrations as compared to horses consuming 0.12 ppm Se at d 154. Calamari et al. (2009) reported greater whole blood and plasma Se concentrations in horses consuming 0.39 ppm Se as compared to horses consuming 0.18 ppm Se at d 28. Richardson et al. (2006) reported significantly greater plasma Se concentrations in horses consuming 0.45 ppm Se as compared to horses consuming 0.15 ppm Se at d 28. Shellow et al. (1985) reported no significant differences in whole blood Se concentrations in horses consuming 0.11 and 0.26 ppm Se in an 84-d trial. However, the authors reported plasma Se concentrations were greater in horses consuming 0.26 ppm Se as compared to horses consuming 0.11 at d 35. Janicki et al. (2001) reported significantly greater serum Se concentrations in mares receiving 3 mg organic Se/d as compared to mares receiving 1 mg inorganic Se/d at d 55. Richardson et al. (2003) reported plasma Se concentrations were significantly greater at d 28 in horses consuming 0.6 ppm Se as compared to horses consuming 0.15 ppm Se. Previous studies have also reported conflicting results on the effects of Se supplementation on GSH-Px activity. Brummer et al. (2013) reported horses receiving 0.3 mg Se/kg DM had significantly greater whole blood GSH-Px activity as compared to horses consuming 0.12 mg Se/kg DM at d 154. Calamari et al. (2009) reported significantly greater plasma GSH-Px activity in horses consuming 0.29 and 0.39 mg Se/kg DM as compared to horses consuming 0.18 mg Se/kg DM at d 84. Richardson et 3 al. (2006) reported RBC GSH-Px activity was significantly greater in horses consuming 0.45 ppm organic Se as compared to horses receiving 0.12 ppm Se. The researchers, however, reported no significant differences in plasma and muscle GSH-Px activity over a 56-d trial. Shellow et al. (1985) reported no significant differences in plasma GSH-Px activity in horses consuming 0.11, 0.16, and 0.26 ppm Se over a 12-wk trial. Richardson et al. (2003) reported no significant differences in plasma GSH-Px activity between horses consuming 0.15 and 0.5 ppm Se throughout a 56-d study. However, the researchers reported horses consuming 0.6 ppm Se had significantly higher RBC GSH-Px activity as compared to horses consuming 0.15 ppm Se at d 28. The objective of the current study was to 1) determine the depletion rate of Se in horses consuming a Se-deficient diet and 2) compare the effects of two different levels of organic Se supplementations on Se repletion as indicated by whole blood Se concentrations and erythrocyte GSH-Px activity in moderately-exercised horses. 4 Chapter II LITERATURE REVIEW Selenium (Se), atomic number 34, is an essential trace mineral that is found in varying amounts in feed. It is a non-metal mineral with an atomic weight of 78.96 that exists in two forms; inorganic species, selenate and selenite, and organic varieties, selenomethionine and selenocysteine. The inorganic forms are present in soil, which plants accumulate and convert to organic forms (NIH, 2013). Selenium Functions The primary function of Se is to serve as an antioxidant. It is a rate-limiting component of the enzyme glutathione peroxidase (GSH-Px). Glutathione peroxidase contains 4 g of Se atoms/mol. The greatest activity of GSH-Px occurs in erythrocytes(RBC) and liver tissue in animals (Ullrey, 1987). Glutathione peroxidase protects cellular membranes and organelles by inhibition and destruction of endogenous peroxides, furthermore it works in conjunction with Vitamin E to maintain the integrity of these membranes. The enzyme catalyzes the breakdown of hydrogen peroxide and certain organic hydroperoxides produced by glutathione during the process of redox cycling (Koller and Exon, 1986). Selenium also counteracts the toxicity of As, Cd, Hg, Cu, Pb, and Ag (Koller and Exon, 1986; Charlton and Ewing, 2007). 5 There are 5 other proteins that incorporate or require Se in order to be produced. Selenoprotein is present in striated muscle (Lescure et al., 2009). Selenoflagellin is a Sebinding polypeptide in sperm (Selenium in Nutrition, 1983). Selenocysteine-containing protein has been reported to be involved in the transport of Se (Reddy and Massaro, 1983). The bacterial enzymes that require Se are formate dehydrogenase and glycin reductase, which are classified as redox enzyme systems (Reddy and Massaro, 1983). Selenium is an essential component of selenoenzymes, nicotinic acid hydroxylase, xanthine dehydrogenase, and a bacterial thiolase, which participate in electron transfer processes and acts as redox catalysts (Stadtman, 1983). Other selenoproteins and selenoamino acid transfer nucleic acids have been identified, but remain undefined (Koller and Exon, 1986). Selenium is a vital component of the immune system. Selenium stimulates production of Immunoglobulin M antibody-producing cells, and enhances Immunoglobulin G production. Immunoglobulins are glycoprotein molecules that are produced by plasma cells in response to an immunogen and function as antibodies (Mayer, 2009). Selenium is also involved in oxidative bursts of phagocytes. Koller and Exon (1986) reported that neutrophils, peritoneal macrophages, and pulmonary alveolar macrophages from Se-deficient animals had low amounts of GSH-Px activity, decreased microcidal activity, and their ability to destroy phagocytized bacteria was compromised. In thyroid hormone production, the enzyme Type-I Iodothyronine 5’-Deiodinase contains Se, which converts the prohormone thyroxine to triidothyronine. Triidothyronine affects growth and development, metabolism, body temperature and heart rate. Within 4 6 to 5 wk of Se depletion in rats, activity of Type-I Iodothyronine 5’- Deiodinase was dramatically reduced, and the ratio of thyroxine: triidothyronine changed with an increase in thyroxine of 50 to 100% (Daniels, 1996). The authors stated that changes in plasma thyroid hormone status are Se specific, and occur as rapidly as the changes in GSH-Px activity (Daniels, 1996). Selenium in Soil and Forage Plant uptake of Se is variable, depending on the chemical form of Se in soil, soil acidity, the climate, and the plant species (Lewis, 1995). Selenium has similar chemical and physical properties to S, and both share common metabolic pathways. Selenium and S compete in biochemical processes that affect uptake throughout plant development (Sors et al., 2005). Intensive farming with S-containing fertilizers has created many crops that are deficient in Se (Charlton and Ewing, 2007). Rapidly growing plants and legumes tend to be low in Se. Plants grown in poorly aerated, acidic soils, soils originated from volcanic rock, and soils with a high content of Fe or S typically have low Se concentrations (Aleman, 2008). Regions of the United States that are generally extremely Se deficient are the Pacific Northwest, Northeast, Great Lake States, Atlantic Seaboard, and Florida. The Plains States and Southwest commonly have adequate Se in soils and plants. Around the world, Australia, New Zealand, and China have extremely low Se content in the soil, and consequently, forage (Koller and Exon, 1986). Soils usually have adequate levels of Se in areas with low rainfall, where minimal leaching of Se from the soil occurs. All but four states (DE, RI, WV, and WY) have 7 reported areas of Se deficiency. Eight states (CA, CO, ID, MT, OR, SD, UT, and WY) have reported excess Se in certain species of plants. Selenium is more readily taken up by plants grown in more alkaline soils (Lewis, 1995). Three types of plants have been identified that are capable of accumulating Se. The categories are: 1) obligate Se accumulator, 2) facultative Se accumulator plants, and 3) crop plants, alfalfa, and grasses. Crops and alfalfa normally contain non-toxic concentrations of Se, however, if they are grown in Se-rich soils, they may contain 1 to 30 ppm Se. Obligate Se accumulator plants have an unpleasant garlic-sulfur odor, which makes them relatively unpalatable and assists grazing animals in identifying them. Horses and other livestock will avoid eating these plants if other feed is available. Obligate Se accumulator plants only grow in soils high in Se. These plants are capable of accumulating up to 10-times the amount of Se present in soil and may contain up to 10,000 ppm Se. Obligate Se accumulator plants include Milkvetches, Golden weeds, Woody asters, Prince’s plume, Astragalus, Haploppus spp., Xylorrhiza glabriuscula, and Stanleya pinnta. Facultative Se accumulator plants do not require Se for growth, but may accumulate up to several hundred ppm of Se when grown in soils high in available Se. This groups of plants includes Asters, Saltbrush, Indian paintbrush, Broomweed, Beard tongue, Gumweed, Ironweed, Bastard toadflax, Aster spp., Machaeranthera spp. Atriplex spp., Castilleja spp., Gutierrezia spp., Penstemon spp., Grindelia squarrosa, Sideranthus grindelioides, and Comandra pallid (Lewis, 1995). 8 Selenium Absorption, Metabolism, and Storage There is no known homeostatic control of Se absorption (Charlton and Ewing, 2007). Absorption takes place primarily in the duodenum of monogastrics. Selenomethionine absorption rate in the duodenum is 98 to 100% (Charlton and Ewing, 2007; EXRX, 2013). Selenomethionine and selenocysteine are actively absorbed by the same mechanism as the amino acid transporters for methionine and cysteine (Daniels, 1996). Absorption rates of the inorganic forms of Se vary between 30 to 100% due to luminal factors. Selenite and selenate are passively, but rapidly, absorbed. Selenate has an apparent absorption of 95%, compared with 62% for selenite (Daniels, 1996). Ruminants absorb 35 to 65% of Se from forages and concentrates. Sodium selenite is oxidized in the rumen, and then metabolized by rumen microorganisms. Organic Se can be metabolized by rumen microorganisms, or absorbed in the small intestine utilizing amino acid pathways (Charlton and Ewing, 2007). Many factors affect Se absorption. Selenium absorption is higher when animals consume a high protein diet (Daniels, 1996). Adequate dietary supplementation of vitamins A, C, and E, and reduced glutathione result in enhanced intestinal absorption of Se. Heavy metals, such as Pb, Fe, Hg, and Cu inhibit Se absorption via precipitation and chelation (EXRX, 2013). Once Se is absorbed, it is bound to a protein and transported in blood to tissues. Plasma Se is primarily present as selenoprotein P. Selenoprotein P accounts for 60 to 70% of plasma Se and is also found in liver. Plasma selenoprotein P concentration is directly dependent on dietary Se. Selenoprotein P in Se-deficient rats was decreased to 5 9 to 10% of that in control rats (Daniels, 1996). However, selenoprotein P declines less rapidly than plasma GSH-Px when the exogenous Se supply is limited (Daniels, 1996; Charlton and Ewing, 2007). In tissues, Se is incorporated into tissue protein as selenocysteine and selenomethionine (Charlton and Ewing, 2007). Daniels (1996) observed that albumin was the main plasma acceptor of Se over the first 4 h post-ingestion, but by 8 h, Se was primarily incorporated into selenoprotein P after processing by the liver. Animals can endogenously synthesize selenocysteine from selenomethionine via the methionine transamination and transsulfuration pathways with adequate concentrations of methionine available, but cannot synthesize selenomethionine. Proteins such as those in skeletal muscle, which nonspecifically incorporate exogenous and preformed selenomethionine or selenocysteine, have been defined as Se-containing proteins. Proteins containing endogenously synthesized selenocysteine are referred to as selenoproteins and are metabolically active (Daniels, 1996). Adipose tissue has very low concentrations of Se. Selenium is more commonly associated with protein tissue. Research in steers and lambs has indicated that diets adequate in natural Se produced liver and skeletal muscle Se concentrations that were higher than those resulting from equal intakes of Se principally from sodium selenite. Naturally occuring Se, such as selenomethione, produced relatively higher milk Se levels as compared to inorganic Se compounds such as sodium selenite (Ullrey, 1987). Selenium is stored in the kidney, liver, spleen, pancreas, and muscle. Kidneys have the highest Se concentration followed by liver, spleen, pancreas, testes, heart, 10 skeletal muscle, lungs, and brain (Ullrey, 1987; Stowe and Herdt, 1992). Normal liver Se concentrations range between 1.2 and 2.0 µg/g of dry weight for all species regardless of age (Stowe and Herdt, 1992). However, skeletal muscle is the major site of Se storage, accounting for approximately 28 to 46% of the total Se pool (http://ods.od.nih.gov, 2013). Selenium homeostasis is primarily regulated by excretion. The primary routes of excretion for monogastrics are urine and feces, and when toxic levels are consumed excretion also occurs via lungs through exhalation. In ruminants, unabsorbed dietary Se is excreted through the feces, and injected Se is excreted through urine. Se retention was reported to be influenced by animals’ Se status, and the amount and chemical form of Se fed (Charlton et al., 2007). Much of tissue Se is labile, and following transition from seleniferous diets to low Se diets, losses from the body are rapid initially and then decrease (Ullrey, 1987). Selenium Absorption- Organic versus Inorganic Plant forms of Se are the same as organic forms in yeast, which is the form that horses naturally consume. The inorganic forms of Se are a by-product of Cu mining. Organic Se, predominantly selenoamino acids and related compounds, are more easily digested, metabolized, and retained in tissues. Organic Se is much safer to feed to livestock than inorganic Se because selenoamino acids are absorbed from the gut via amino acid pathways, which aids in limiting excessive absorption of Se. Selenite Se is passively absorbed, which allows rapid and unregulated uptake of possibly toxic levels of 11 Se. Organic Se is also safer to handle because it is not absorbed through human skin like sodium selenite (Equine Nutrition, 2005). In order for Se to be incorporated into selenoproteins, dietary sources of Se must be inserted into cysteine where Se replaces the thiol (-SH) side chain, thus forming the amino acid residue selenocysteine. Inorganic species of Se (selenite and selenate) must first be reduced to selenide before being incorporated into selenocysteine residues. Sodium selenite is the most common inorganic form of Se supplemented to horses. Apparent absorption of selenite in mature horses was reported to be 51.1% (Pagan et al., 2007). Selenomethionine is the most common organic form of Se fed to horses, and is most prevalent in plants and yeast. Apparent absorption of selenomethionine was shown to be 57.3% in horses (Pagan et al, 2007). Selenomethionine is actively transferred through the intestinal membrane and can replace methionine during protein synthesis. Selenium is not catalytically active in selenomethionine form and must be converted to selenocysteine. Dietary sodium selenite is more rapidly incorporated into GSH-Px in serum than selenomethionine, but is not stored in tissues as much as selenomethionine (White, 2010). Selenium- Injectable Injectable Se products administered immediately before competition have been gaining in popularity because of their possible performance-enhancing qualities in equine. In April 2009, 21 polo ponies in South Florida died after receiving an injectable Se supplement containing an acutely toxic concentration of Se. The compounding pharmacy responsible for creating the supplement miscalculated the amount of Se to be 12 added to the injection (Desta et al., 2011). This example highlights the narrow margin between the Se requirement of animals and Se toxicity, especially with injectable Se (White, 2010). In addition, adverse responses to injectable Se/vitamin E products have been observed in several species and animal owners should be advised of the potential fatal effects of these products, even when used at recommended doses. The response is an immediate, usually fatal, anaphylactic reaction (Stowe, 1998). The reaction is not to the Se or vitamin E in the product but apparently to an emulsifying agent or preservative present in the product. None of these untoward reactions is observed from oral administration of Se at appropriate rates (Stowe, 1998). Selenium and Glutathione Peroxidase in Blood Whole blood Se is a good measure of Se intake because it represents both serum and RBC Se, and appears to be a more preferable indicator of Se status than serum (NRC, 2007). However, whole blood Se responds more slowly than serum or plasma to changes in dietary Se intake because a majority of the glutathione peroxidase in whole blood is incorporated into the RBC at the time of erythropoiesis, and changes very little over the life of the cell. A measurable response in whole blood Se to a Se supplement, therefore, requires a time span equal to the average life span of RBC. In cattle, the life span of a RBC is about 90 to 120 d. (Stowe and Herdt, 1992). Carter et al. (1974) reported the lifespan of erythrocytes in light horses to be 145 to 165 d. The whole blood Se: serum Se ratio is approximately 1:1 in swine, 1.4:1 to 1.5:1 in horses, 2.5:1 in dairy cattle, and 4:1 in sheep, particularly neonates (Stowe and Herdt, 1992). These ratios would initially 13 narrow after an increase in oral Se intake and initially widen on cessation of Se supplementation. In swine, the correlation coefficients of Se concentrations of these tissues related to plasma Se are 0.95, 0.91, and 0.71 for skeletal muscle, liver, and kidney, respectively, on a wet weight basis (Stowe and Herdt, 1992). Whole blood GSH-Px activities are consistently measurable. Complete GSH-Px activity response to Se supplementation requires about 80 to 90 d, equal to the life span of equine erythrocytes, as Se is only incorporated into RBC during erythropoiesis. Whole blood GSH-Px concentrations range from 40 to 160 units of enzyme activity(mU)/mg (hemoglobin) Hb in horses (Stowe, 1998). There is a high correlation between erythrocyte GSH-Px activity and Se concentrations in whole blood of humans, cattle, sheep, horses, and swine with a low Se status; however, these correlations became much weaker as Se status increased (Ullrey, 1987). Whole blood has a 10 to 50% higher Se concentration due to the significantly higher concentration of Se in erythrocytes than in plasma. In sheep blood, GSH-Px activity in erythrocytes was 99-times that of plasma (Ullrey, 1987). In erythrocytes, plasma GSH-Px activity ratio in cattle is 49:1, and 26:1 in swine (Ullrey, 1987). Animals are considered to be sub-clinically deficient when whole blood Se concentration and GSH-Px activity is less than 0.05 ppm and 30 mU/mg hemoglobin, respectively. Selenium and GSH-Px statuses are considered marginal between 0.05 to 0.1 ppm and 30 to 60 mU/mg hemoglobin. Blood Se concentrations and GSH-Px activity greater than 0.1 ppm and 60 mU/mg hemoglobin, respectively, are considered adequate (Koller and Exon, 1986). 14 Selenium Deficiency Signs and symptoms of Se deficiency are similar for both animals and humans. Severe Se deficiency is characterized by cardiomyopathy. Moderate deficiency is characterized in less severe, myodegenerative symptoms such as muscular weakness and pain. Symptoms range from the well-recognized, ominous, severe condition of nutritional muscular dystrophy i.e., “white muscle disease” to numerous, less explicit conditions often referred to as Se-associated or Se-responsive diseases. Selenium-associated diseases are characterized by muscular weakness, unthriftness, reduced weight gain, diarrhea, stillbirths, abortions, and diminished fertility (Koller and Exon, 1986; Charlton and Ewing, 2007). White muscle disease is a myodegenerative disorder most commonly associated with neonates. The disease occurs most often in lambs, but has also been observed in calves and horses. Young animals may die suddenly due to myocardial dystrophy. Subacute symptoms are stiffness, weakness, and trembling of the limbs, frequently followed by the inability to stand, and swollen muscles that feel hard to the touch. Affected animals also exhibit dyspnea and labored breathing from involvement of diaphragm and intercostal muscles (Koller and Exon, 1986; Charlton and Ewing, 2007). In horses, Se-deficiency symptoms are more ambiguous than in other livestock species. Selenium-deficient horses can experience myopathies such as myositis, polymyositis, and azoturia. Infertility is also commonly observed in deficient horses. 15 Finally, muscular weakness in foals and reduced performance during exercise are common symptoms of Se deficiency (Koller and Exon, 1986; Charlton and Ewing, 2007). A muscular disorder associated with Se/vitamin E deficiency in horses is a nonexertional myopathy with rhabdomyolysis. It is a peracute to subacute myodegenerative disease of cardiac and skeletal muscle caused by a dietary deficiency of Se, and to a lesser extent vitamin E (Aleman, 2008). This disease occurs primarily in young growing foals, but has also occurred in older horses. Peracute clinical signs in foals include recumbency, tachypnea, myalgia, arrhythmias, and sudden death. Subacute signs include severe weakness, inability to stand, muscle fasciculations, firm muscles on palpation, stiffness, stilted gait, myalgia, lethargy, dysphagia, trismus, ptyalism, and a weak suckle reflex. Physiological alterations in horses with low whole-blood Se and GSH-Px activity include high serum creatine kinase and aspartate aminotransferase activities, hyperprotienemia, azotemia, hyponatremia, hypochloremia, hyperkalemia, hyperphosphatemia, respiratory acidosis, and myoglobinuria (Aleman, 2008). Also, muscles are pale with white streaks, representing coagulative necrosis and edema. The muscles most affected are the myocardium, thoracic, pelvic and cervical muscles, diaphragm, tongue, pharynx, intercostals and masticatory muscles (Aleman, 2008). Clinical manifestations of many of these disorders require contributory factors, such as stress, to precipitate symptoms (Koller and Exon, 1986). Selenium Deficiency Economic Impact Selenium deficiency caused enormous yearly economic loss on producers before Se supplementation was permitted (Koller and Exon, 1986). Prior to the allowance of Se 16 supplementation, 15 to 20% mortality, and at least 25% morbidity was observed in growing pigs of swine herds. In addition, reproductive efficiency was lower and resistance to environmental stress and infectious disease was diminished. Comparable death losses and declines in productivity were observed in poultry and other livestock species (Ullrey, 1992). The USDA approved Se supplementation at a rate of 0.1 ppm in 1974. Prior to this, the inability to supplement deficient poultry and swine caused U.S. producers annual losses of over $82,000,000 (Ullrey, 1992). Dietary supplementation at 0.1 ppm was approved for beef cattle, dairy cattle, and all ages and gender of sheep in 1979. Prior to this approval, estimated annual loss for U.S. producers of beef cattle, dairy cattle, and sheep was approximately $545,000,000 in 1976 (Ullrey, 1992). Subsequent research provided evidence that additional Se could be beneficial. In 1987, a maximum level of supplemental Se from 0.1 to 0.3 ppm in complete feeds for all major food-producing animals was approved by the FDA (Ullrey, 1992). Selenium Toxicity Selenium was first identified as a toxic element that induced hair loss, lameness, hoof sloughing, and death in grazing livestock in SD and WY in 1934 (Ullrey, 1992). Marco Polo and T.C. Madison observed similar signs in horses in China in the 13th century, and at Fort Randall, NE Territory in 1860, respectively. T.C. Madison called the toxicity “alkali disease” (Ullrey, 1992). Toxic concentrations of Se inhibit cellular enzyme oxidation-reduction reactions, especially those involving sulfate or S-containing amino acids methionine and cysteine, which affect cell division and growth. Hoof and 17 hair are especially susceptible to the effects on cell division (Lewis, 1995). Selenium toxicity is usually associated with incorrect feed levels or eating Se-accumulating plants. Feeds containing more than 5 ppm Se are considered Se toxic. The Maximum Dietary Tolerable Level has been established at 2 mg/kg DM (Charlton and Ewing, 2007). Acute signs of Se toxicity include garlicky breath, vomiting, dyspnea, titanic spasms, apparent blindness, head pressing, perspiration, abdominal pain, colic, diarrhea, increased heart and respiration rates, and death from respiratory failure (Koller and Exon, 1986; NRC, 1989). Death due to pulmonary congestion and edema occurs from acute Se toxicity with 25 to 50 mg/kg (Lewis, 1995). Chronic toxicity occurs when an animal consumes 5 mg/kg or more of Se (Koller and Exon, 1986). Chronic poisoning symptoms are abnormal hoof and hair growth, alopecia (especially mane and tail), fetal abnormalities, and cracking of hooves around the coronary band (NRC, 1989; Lewis, 1995). In horses, chronic Se toxicity was reported in 25 horses used in a feedlot in NE (Stowe and Herdt, 1992). The horses developed hair loss and lesions around the coronary band. One horse had 928 ng Se/mL in serum as compared to an expected normal range of 140 to 160 ng Se/mL. Feed analysis determined that these horses were fed hay containing 20 ppm Se for more than 3 wk. After 2 wk of consuming hay with non-toxic Se concentrations, mean serum Se was 525 ng Se/mL, and 6 wk after the diet change, mean Se serum was 285 ng Se/mL (Stowe and Herdt, 1992). 18 Selenium Supplementation Environmental Impact Some environmental groups have raised concerns about the impact that Se supplementation may have on the environment. For example, the Kesterson Reservoir was essentially a wastewater dump that received irrigation drainage water from the San Luis United of the United States of Reclamation’s Central Valley Project in western Fresno County (Ullrey, 1992). Selenium was proposed as a cause of death and deformities in aquatic birds and other organisms. Selenium, from seleniferous marine rocks of Oligocene, may have been one of the factors involved. However, there is no evidence of undesirable amounts of Se entering this ecosystem from the legal use of Se supplements in animal diets (Ullrey, 1992). No significant differences in upstream and downstream contents of Se in water, stream sediment, algae, invertebrates, and fish was observed at ranches on which Se supplementation of beef cattle had been practiced for 3 to 8 yr (Oldfield, 1998). Primary domestic production of Se in 1989 was about 250 metric tons and 450 metric tons imported, of which 40% was used for electronic and photocopier components, 20% glass manufacturing, 20% chemicals and pigments, and 20% other applications. Less than 6.8% of 47.5 metric tons was used for supplementing animal diets. Fuel combustion, refuse combustion, metal mining and refining, and industrial production were identified as anthropogenic contributions of 4,670 metric tons (Ullrey, 1992). If all the Se incorporated into animal feeds were to enter the environment, it would account for less than 0.5% of the Se that originated from natural and other identified anthropogenic sources. Unabsorbed inorganic Se in animal feces is largely insoluble elemental Se and 19 metal selenides and urinary trimethyl selenonium is poorly available for absorption by wildlife and aquatic life. Therefore, the environmental threat from legal use of Se supplementation seems very small (Ullrey, 1992). Selenium in Cattle Podoll et al. (1992) used 18 lactating Holstein cows, split into 2 treatments; 0.3 ppm sodium selenite and 0.3 ppm sodium selenate to determine if sodium selenate had superior bioavailability to Na selenite as a source of supplemented Se. Serum was collected on d 0, 3, 7, 10, 14, 28, 42, and 49 for Se and GSH-Px assays. The authors reported serum Se concentrations rose significantly during the study in cattle consuming both treatments. Response to supplemental Se was cubic. Sodium selenate supplementation produced significantly greater serum Se concentrations than selenite. Serum GSH-Px activities were unaffected by the form of Se, but were significantly different over time (Podoll et al., 1992). In order to determine the response of calves fed low-Se diets, then supplemented with either Se-enriched yeast or inorganic Se, Nicholson et al. (1991a) fed 50 crossbred beef calves (6 to 7 mo of age) and 20 yearling Holstein heifers. There were 5 treatments; Control (no supplemental Se or yeast), Inorganic Se (Sodium selenite to supply 1 mg Se/kg of supplement), organic Se (Alkosel yeast to supply 1 mg Se/kg of supplement), live yeast, and autoclaved yeast (commercial yeast culture was autoclaved for 8 min at 15 psi). The experimental period was 112 d. Blood samples were collected at 4 wk intervals and whole blood Se and GSH-Px activity was measured. The authors reported animals fed organic Se supplement had significantly higher blood Se concentrations than those 20 not receiving supplemental Se at d 28. Differences continued to be observed for the duration of the trial. By d 84, whole blood Se concentrations for animals fed organic Se were numerically higher as compared to those fed inorganic Se, and the differences became significant by d 112. There were no significant differences in whole blood Se concentrations among the groups that did not receive supplemental Se. Whole blood GSH-Px activity was not significantly different at d 28 among treatments, but became significantly different at d 56 and continued remained significantly different throughout the remainder of the trial. Differences in blood GSH-Px activities due to 2 sources of Se became significant at d 112. Whole blood Se and GSH-Px values for cattle fed inorganic Se appeared to plateau between d 84 and 112, while those fed organic Se appeared continued to increase (Nicholson et al., 1991a). Nicholson et al. (1991b) used 48 growing beef cattle and 20 yearling Holstein heifers to compare rates of change in Se concentrations of whole blood or blood plasma due to altered dietary Se source. Animals received 2 kg concentrate with or without addition of Se enriched yeast to supply 1 ug Se/d. Blood samples were collected for Se analysis at approximately 28-d intervals over a 163-d period. The authors reported increased Se levels in whole blood over 8 wk even though levels in all animals were within the normal range at the start of the experiment. Increases were greater for animals that received Se-enriched yeast in their supplement than for the non-supplemented animals, but the slopes of the linear regressions did not differ significantly. When animals were changed to low Se diets, there was a significant decline in whole blood Se for those animals not fed Se-enriched yeast in their concentrate in the slope of the regression line 21 over the final 16 wk of the trial for whole blood concentrations. However, this significant decline was not observed in plasma. The authors concluded that changes in whole blood Se concentrations are a more accurate measure than plasma Se concentrations of the adequacy of current Se intake as the magnitude of change was greater and values did not plateau at as low a level of intake (Nicholson et al., 1991b) Selenium in Sheep In sheep, Wright and Bell (1966) fed 10 wethers 0.35 ppm Se 2 wk prior to, and during the experimental period to determine Se retention. Five wethers were given a single oral dose via a gelatin capsule, and 5 were given a single intravenous dose of radioactively-labeled selenium. The authors observed the retention of Se after 120 hr was 29% when radioactively-labeled selenium was administered in a single oral dose. Retention of intravenous dose of Se was 70% after 120 hr with the major route of excretion via urine (Wright and Bell, 1966) Podoll et al. (1992) fed 20 crossbred wethers 1 of 2 treatments; 0.3 ppm sodium selenite and 0.3 ppm sodium selenate to determine bioavailability of each Se source. Serum was collected on d 0, 3, 7, 10, 14, 28, 42, and 49 for Se and GSH-Px assays. The authors observed serum Se concentrations rose significantly during the study. The response of wethers to supplemental Se was quadratic. However, there were no differences observed in serum Se due to treatment. Serum GSH-Px activities were unaffected by the form of Se but were significantly different over time (Podoll et al, 1992). 22 Selenium in Swine Wright and Bell (1966) fed 10 barrows 0.5 ppm Se daily for 2 wk prior to and during an experimental period, consisting of a 4-d preliminary and a 5-d collection period. Five barrows were given a single oral dose via stomach tube, and 5 were given a single intravenous dose of radioactively-labeled selenium. Retention of oral Se after 120 hr was 77%. Retention of intravenous dose of Se was 70% after 120 hr with the major route of excretion via urine (Wright and Bell, 1966). Chavez (1979) weaned 16 piglets at 14 d to evaluate the biodynamic relationship between blood Se concentrations and the activity of GSH-Px in the plasma, and Se concentration changes in different body tissues during Se depletion and repletion. Piglets were randomly assigned into 2 dietary treatments for 4 wk: basal diet containing 0.02 ppm Se, or basal diet supplemented with 0.1 ppm Se as Na selenite. After 4 wk, half of the piglets from each dietary regime were changed over to the other diet for 5 wk. The change in diet represented the depletion or repletion period for piglets fed the previous respective dietary treatment. Blood samples were collected at weaning, and weekly throughout the trial. The authors reported after 1 wk of receiving the experimental diets, there was a significant difference in blood Se concentrations between the 2 groups of piglets. At wk 4, piglets receiving Se supplement had 132 ug Se/L as compared to 27 ug Se/L in non-supplemented piglets. Piglets receiving Se supplement had increased blood Se concentration (a maximum of 165 ug/L in wk 9), although no significant difference was observed during the last 3 wk of the trial. Blood Se concentration of piglets continuously fed the Se deficient diet decreased to an average minimum value of 23 approximately 17 ug/L after 8 wk, with significant variation observed during the last 4 wk of the trial. In piglets changed from a Se deficient to Se supplemented diet, blood Se concentration increased steadily for 5 wk, although a much faster repletion rate took place during the first 2 wk after the change in diet. Piglets receiving Se supplementation during the entire trial had a significant increase in plasma GSH-Px activity, while piglets fed the Se deficient diet during the entire trial had a significant decrease. Piglets changing from Se supplementation to a Se deficient diet had a significant decrease in plasma GSHPx activity during wk 1 of depletion, and this activity continued to decrease thereafter, but at a slower rate. Piglets changed from a Se-deficient diet to Se supplementation at wk 4 had a significant and steady increase in GSH-Px activity for 4 wk after the change. Further, this activity peaked at a value higher than that observed in the plasma of control animals receiving Se supplementation continuously. The repletion rate of blood Se was about 12% faster in piglets changed to supplemental Se treatment as compared to the depletion rate of piglets changed to the basal diet (Chavez, 1979). Selenium in Horses Horses, zoo animals, llamas, and other pets have never been included in the FDA regulations on Se supplementation. However, reference values for Se in mature horses have been established; serum between 130 to 160 ng Se/mL, whole blood between 182 to 240 ng Se/mL, and liver between 1.2 to 2.0 ug Se/g DM (Stowe, 1998). Maximum tolerable concentration of dietary Se for horses is reported to be about 2 ppm (Stowe, 1998). Therefore, a considerable margin of safety exists between the practiced 0.1 to 0.3 ppm rate of supplementation and maximum tolerable level (Stowe, 1998). 24 Historically, inorganic Se sources have been used in equine feeds, but the margin of error for inorganic Se supplementation is narrow, and efficacy has been questioned. A growing body of research suggests organic Se sources enhance Se incorporation into tissues, both at rest and during exercise (Dunnett and Dunnett, 2008). Further, organic Se supplementation has been observed to increase Se status, enzyme activities, antioxidant capacity and immune function in mature horses and foals. Dietary organic Se appears to cause a greater relative increase in plasma Se over 28 d as compared to selenite, although comparative effects were similar over 56 d for skeletal muscle Se and plasma GSH-Px activity. Dietary organic Se also produced a greater numerical, but statistically insignificant, increase in plasma GSH-Px activity than selenite during supplementation to horses over 112 d. Post-supplementation decline in plasma GSH-Px activity was also reduced in horses receiving organic Se. Data from other studies have indicated that preand post-partum organic Se supplementation in mares has subsequent benefits in the foals through improved Se status (Dunnett and Dunnett, 2008). The Se requirement for sedentary horses was estimated at 0.1 mg/kg of diet by Stowe in 1967. Exercise increases oxidative metabolism markedly, which results in mobilization of tissue Se to meet increased antioxidant demand, explaining why performance horses have greater Se requirements than non-athletes (Equine Nutrition, 2005). Unlike other livestock species, the FDA only makes dietary recommendation for Se in equine feeds. In horses, nutritional myopathy involving skeletal and cardiac muscles is associated with GSH-Px values lower than 25 mU/mg and serum Se values lower than 60 ng/mL. Selenium deficiency results in weakness, impaired locomotion, 25 difficulty in suckling and swallowing, respiratory distress, and impaired cardiac function. In deficient horses, serum concentrations of creatine kinase, aspartate aminotransferase, K, aspartic-pyruvic transaminase and gamma-glutamyltransferase are increased (NRC, 1989). Serum Se in foals from Se-adequate mares is typically much lower than their dams, and ranges from 70 to 80 ng/mL. Serum Se values lower than 65 ng/mL are indicative of deficiency (NRC, 1989). Stowe (1967) obtained 12 orphaned foals initially fed a commercial milk replacer. Foals were used to evaluate the effect of Se on growth rate before clinical evidence of deficiency occurs. Half of the foals were fed a semi-purified diet, and other half were fed semi-purified diet supplemented with 2 ppm Se in the form of Na selenite. The author reported a tendency for the Se-supplemented foals to gain more rapidly than the Sedeficient foals (Stowe, 1967). Carmel et al. (1989) surveyed a randomly selected horse population from 4 contiguous counties in northern MD to determine the Se status of resident horses. From the MD horse population, 203 horses from 74 farms were sampled from January through May, 1988. Information on signalment, duration of residence, use, housing, medical history, and feeding program was collected. Whole blood Se concentrations greater than or equal to 0.1 ppm were considered adequate. The authors reported average whole blood Se concentrations were 0.137 ppm, and ranged from 0.05 to 0.266 ppm. Of the horses sampled, 18.7% were considered deficient. There was a significant negative correlation observed between whole blood Se concentration and amount of time horses had access to 26 pastures. Horses used daily and those fed daily supplement were significantly more likely to have adequate Se concentrations (Carmel et al., 1989). Ludvikova et al. (2005) collected blood samples from 159 horses from 35 different farms to determine the relationship between Se concentration and activity of GSH-Px in whole blood of horses, reference ranges for the activity of GSH-Px, and to evaluate Se status of horses in the Czech Republic. The authors observed a highly significant linear relationship between Se concentration and GSH-Px activity. Whole blood Se concentrations of 75 ug/mL were considered the threshold of Se deficiency. There was a high prevalence of selenium deficiency in horses examined. Selenium status and GSH-Px activity was considered deficient in 47 and 48% of horses examined, respectively (Ludvikova et al., 2005). Blackmore et al. (1982) measured Se concentrations and GSH-Px activity in 84 Thoroughbreds to assess the relationship between Se status, and muscle and hepatic disorders. Whole blood was collected and analyzed for Se, and any muscle or hepatic disorders were assessed. Researchers reported a significant linear (r = 0.843) and quadratic (r = 0.976) relationship between whole blood Se and RBC GSH-Px activity (Blackmore et al, 1982). Knight and Tyznik (1990) evaluated the effects of supplemental Se on equine humoral antibody production. The authors utilized 15 Shetland ponies; five 2-yr old, four 3-yr old, and 6 yearlings. During the depletion phase, 2 and 3-yr old ponies were fed a low-Se diet for 1 yr, and yearlings were fed similarly for 9 mo. During the depletion period, the average GSH-Px activity decreased from 150 mU/mg hemoglobin to 20 27 mU/mg Hb. During the 7-wk repletion period, horses were assigned to 1 of 2 treatments; low (0.02 ppm), or high (0.22 ppm). The authors stated Se supplementation had a positive effect on the immune response. Serum Immunoglobulin G concentrations of ponies receiving Se-supplementation were significantly higher as compared to those receiving no Se. Older ponies had significantly higher serum Immunoglobulin G concentrations than did yearlings. A significant interaction between Se and time was observed for serum Immunoglobulin G concentration and hemagglutination titers. Serum Immunoglobulin G concentrations were significantly higher during wk 2, 3, 4, and 5 in Se-supplemented ponies. Hemagglutination titers during wk 2, 3, 4, and 5 also were significantly greater in ponies receiving supplemental Se. Horses consuming Se supplementation had significantly higher whole blood Se concentrations and glutathione peroxidase activity. Whole blood Se concentrations and glutathione peroxidase activities significant increased during the 6-wk trial in supplemented ponies. Selenium-supplemented ponies had significant higher whole blood Se concentrations during wk 4, 5, and 6 and glutathione peroxidase activities during wk 6. A significant positive correlation (r = 0.79) between whole blood Se and glutathione peroxidase activity within treatment was observed (Knight and Tyznik, 1990). Shelle et al. (1985) conducted a study using 8 Arabian mares in a 2 x 2 double split-plot design with repeated measures to determine the effects of conditioning, exercise, and daily Se supplementation. Treatments consisted of 2 levels of dietary Se, no added Se or 2.5 mg added Se/d. Exercise treatments were non-conditioned, conditioned 6 d per wk for 45 d, or conditioned for 3 d followed by 2 d stall rest for 45 d. Blood 28 samples were taken before, during and after exercise. The authors reported plasma Se concentrations significantly increased with Se supplementation above pre-feeding levels. Horses consuming the basal ration had a significant decline in plasma Se over the length of the trial. Plasma Se concentrations were significantly elevated during exercise as compared to 1 hr post-exercise. The authors hypothesized increased plasma Se concentrations during exercise resulted from changes in plasma volume rather than mobilization of Se from body stores. Mean plasma GSH-Px activities were 5.3 and 7.7 mU/mg of plasma protein and significantly different in non-supplemented and supplemented mares, respectively. Furthermore, the authors stated that erythrocyte GSHPx activities increased as a result of conditioning. Selenium supplementation appeared to augment the effect of conditioning and resulted in significant treatment by conditioning interaction. Glutathione peroxidase activity in whole blood was significantly elevated during exercise (Shelle et al., 1985). Brummer et al. (2009) used 24 horses to establish the correlation between Se status, as measured by serum Se concentrations, and GSH-Px activity, along with several immune-related variables. Sixteen horses received no dietary supplementation except for access to a salt block, while 8 horses were supplemented with a commercial grain-based concentrate containing at least 0.3 ppm Se. Horses were fed their respective diets for 4 mo prior to blood sampling. Blood was drawn over a period of 6 wk. Each horse was sampled once. Authors reported serum Se concentrations ranged from 69 to 193 ng/mL. Mean serum Se concentrations were significantly different between supplemented (165 ng/mL) and unsupplemented (91 ng/mL) horses. A positive correlation of medium 29 strength was observed between serum Se and whole blood GSH-Px activity (r = 0.710). A weak correlation was observed for serum Se and IL-10 gene expression (r = 0.419). A trend was observed for a weak correlation between serum Se and serum GSH-Px (r = - 0.0359). There was also a trend for weak correlations between serum Se and tumor necrosis factor expression (r = 0.381), and serum Se and tumor necrosis factor production (r = 0.446; Brummer et al., 2009). Chiaradia et al. (1998) studied the possible relationships between physical exercise, lipid peroxidation and muscle fiber damage in trained horses. Researchers fed ten 3-yr old Maremmana stallions a minimum of 12 mg Se and 1000 IU of vitamin E/d. Stallions underwent physical training for 3 mo, 30 min/d, 6 d/wk, and intensity gradually increased. At the end of the trial, horses performed an exercise test consisting of an 8 min warm-up period followed by two 200-m gallops. Blood samples were collected before exercise, immediately after warm-up, after the gallops, and 18-hr post exercise. Total plasma glutathione, reduced glutathione and glutathione disulphide were measured. Results indicated that the pattern of glutathione content in the plasma was similar before exercise and immediately after warm-up, increased after the gallops, and decreased to pre-exercise concentrations 18-hr post exercise. The authors stated that after an oxidative stress, glutathione is released in the blood. The oxidized form of glutathione is transferred from the cells to the liver to be reduced, and the reduced form of glutathione is then released by the liver to support increased requirement of cells for this substrate, which is necessary for the activity of GSH-Px (Chiaradia et al., 1998). 30 In a companion study to Chiaradia et al. (1998), Avellini et al. (1999) sought a better understanding of the effect of dietary supplements and a 70-d training period on the peroxidation phenomena induced by rigorous programmed physical exercise trials of increasing intensity. The authors reported the activity of GSH-Px was significantly lower at the beginning of the trial as compared to d 70. No significant modifications in enzyme activity were observed after physical exercise. Glutathione peroxidase activity significantly increased over the 70-d period of training and diet supplements. The authors concluded training and diet supplements increased antioxidant defenses in extracellular fluids and blood cells of the horses (Avellini et al., 1999). Greiwe-Crandell et al. (1993) split 45 pregnant Thoroughbred mares into 3 groups to determine mineral status of mares and foals during Se depletion. Mares were divided into the following treatment groups; 15 fed mixed grass/legume pasture and supplemented with hay only, 15 fed similar pasture and supplemented with hay plus 3.2 kg/d of a concentrate, and 15 dry-lotted and fed 2-yr old mixed grass hay and 4.6 kg/d of concentrate. After foaling, weanlings remained on the same regimens as their dams. The concentrate contained 0.6 mg/kg Se, and the pastures and hay contained 0.08 mg/kg Se. The authors reported whole blood Se concentrations of horses fed pasture and hay only were significantly lower than horses fed 3.2 kg/d of concentrate, and horses fed 4.6 concentrate. Whole blood Se concentrations tended to be different in November between horses fed 3.2 and 4.6 kg/d concentrate, however, there was no difference between the 2 groups in January (Greiwe-Crandell et al., 1993). 31 Karren et al. (2010) used 28 pregnant Quarter Horse mares in a 2 x 2 factorial, randomized complete block design to investigate the maternal plane of nutrition and the role of Se-yeast on muscle Se concentration, plasma GSH-Px activity, and colostrum Se concentration in mares and their foals. There were 4 treatments. Seven mares were allowed access to pasture and received no Se supplementation, receiving a total of 0.19 mg Se/kg DM. Eight mares were allowed access to pasture and Se supplementation, receiving 0.49 mg Se/kg DM. Five mares fed pasture and grain with no Se supplementation receiving 0.35 mg Se/kg DM. Eight mares were fed pasture and grain with Se supplementation receiving 0.65 mg Se/kg DM. The treatments were initiated 45 d prior to third trimester. Selenomethionine supplementation was initiated at the beginning of the third trimester (approximately 110 d before estimated foaling date). Blood samples were collected every 14 d until parturition. Foal blood samples were taken beginning at birth and every 14 d until 56 d of age. Colostrum samples were obtained after parturition and before nursing. Mare muscle biopsies were collected every 28 d until parturition. Foal muscle biopsies were collected at birth and on d 28 and 56. The authors reported that mare plasma Se concentrations were significantly greater in mares consuming pasture and grain with no Se supplementation, and pasture and grain with Se supplementation than in mares consuming pasture with no Se supplementation and pasture with Se supplementation. Mares receiving Se supplement had significantly greater plasma Se concentrations than mares receiving no supplement. Muscle and colostrum Se concentrations were significantly greater in mares consuming pasture and grain with Se supplementation and pasture with Se supplementation than in mares 32 consuming pasture with no Se supplementation, and pasture and grain with no Se supplementation. No effect of treatment was reported on mare muscle or colostrum Se concentrations. Mare plasma GSH-Px activities were not affected by nutrition or selenomethionine supplementation. Foals of mares consuming pasture and grain with no Se supplementation and pasture and grain with Se supplementation had significantly greater plasma and muscle Se concentrations as compared to foals of mares consuming pasture with no Se supplementation and pasture with Se supplementation. Foal plasma GSH-Px activities were not affected by maternal plane of nutrition or selenomethionine supplementation of the dam (Karren et al., 2010). Brummer et al. (2011a) examined the effect of low Se status on the ability of both the humoral and cell mediated components of the immune system to respond to a vaccine challenge. Of the 28 horses used in the study, 7 received 0.14 ppm Se from sodium selenite, and 21 horses received 0.07 ppm Se from sodium selenite. Blood samples were taken at d 0 and thereafter every 4 wk for 28 wk, and analyzed for whole blood Se and GSH-Px activity. Authors reported whole blood Se was significantly lower in horses consuming 0.07 ppm Se (164.7 ng/mL) than horses consuming 0.14 ppm Se (211.1 ng/mL) after 28 wk of treatment. The authors also reported GSH-Px activity was significantly lower in horses consuming 0.07 ppm Se (42.72 mU/mg Hb) than horses consuming 0.14 ppm Se (55.00 mU/mg Hb). In response to vaccination, KLH-specific Immunoglobulin G concentrations increased over time in both groups, but horses consuming 0.14 ppm Se responded significantly quicker with significantly higher antibody concentrations at 3 wk than horses consuming 0.07 ppm. Expression of the 33 transcription factor T-bet was significantly greater at 5 wk in horses consuming 0.14 ppm Se horses consuming 0.07 ppm Se (Brummer et al., 2011a). In a sister study, Brummer et al. (2011b) hypothesized that Se depletion would result in a decrease in GSH-Px activity, serum total antioxidant capacity (TAC) and an increase in oxidative stress, as measured as serum malondialdehyde concentration (MDA), and T3:T4 ratio changes. The data indicated that whole blood Se concentration significantly decreased over time in both groups but the decrease was greater in horses consuming 0.07 ppm Se. At the end of 28 wk, T3:T4 significantly increased in horses 0.07 ppm Se, while it remained similar to initial levels in horses consuming 0.14 ppm Se. Whole blood GSH-Px activity decreased during the study in both groups; however, final whole blood GSH-Px activity was significantly different. Total antioxidant capacity did not change during study. Malondialdehyde concentration was not different between horses consuming 0.14 and 0.07 ppm Se at the initial or final draw, however MDA did significantly increase over time in horses consuming 0.07 ppm Se while it remained similar in horses consuming 0.14 ppm. The increased ratio of T3:T4 in 0.07 ppm Se horses along with the changes in whole blood Se and GSH-Px suggested that horses consuming 0.07 ppm Se were at, or approaching, deficient Se status (Brummer et al., 2011b). Podoll et al. (1992) fed 12 adult Arabian horses 1 of 2 treatments; 0.3 ppm sodium selenite and 0.3 ppm sodium selenate to determine differences in supplemental Se source. Serum was collected on d 0, 3, 7, 10, 14, 28, 42, and 49 for Se and GSH-Px assays. The authors reported serum Se concentrations rose significantly during the trial in 34 all horses regardless of treatment. The overall response to supplemental Se was quadratic. There were no significant differences observed in serum Se concentrations between those consuming selenite or selenate. Serum GSH-Px activities were unaffected by the form of Se but were significantly different over time (Podoll et al., 1992). Montgomery et al. (2012a) assigned 15 Standardbred horses to 1 of 3 treatments; a control receiving no supplementation, inorganic Se (sodium selenite), and organic Se (Se yeast). Three months prior to trial, the study horses were turned out on a pasture containing less than 0.05 ppm Se. The objective of the study was to examine the effects of oral Se supplementation and Se source on aspects of innate and adaptive immunity in horses. Immune function tests were performed that measured lymphocyte proliferation in response to mitogen concanavalin A, and neutrophil phagocytosis, and antibody production after rabies vaccination. Relative cytokine gene expression in stimulated lymphocytes (interferon gamma, IL-2, IL-5, IL-10, tumor necrosis factor-alpha) and neutrophils (IL-1, IL-6, IL-8, IL-12, tumor necrosis factor-alpha) was also examined. Plasma and RBC Se, and blood GSH-Px activity were analyzed. Plasma and RBC Se were significantly highest in horses in the organic Se group as compared to those in the inorganic and control groups. Organic Se supplementation increased the relative lymphocyte expression of IL-5, as compared to inorganic Se or no Se. Selenium supplementation increased relative neutrophil expression of IL-1 and IL-8. Other measures of immune function were unaffected (Montgomery et al., 2012a). Montgomery et al. (2012b) also investigated the effect of dietary Se source on Se status of mares and, consequently, the Se status and immune function of their foals. The 35 authors used 20 pregnant Standardbred mares. Mares were assigned to 1 of 2 treatments; a complete pelleted feed containing 0.3 ppm organic Se, or a complete pelleted feed containing 0.3 ppm inorganic Se. Mares were fed their respective diets for 2 mo prior to estimated parturition. The authors reported that mean plasma Se concentrations prior to the beginning of treatment were reflective of low Se intake, falling into a range considered inadequate. Mare plasma and RBC Se concentrations increased in both groups following onset of treatment. There was no significant effect of Se source on plasma or RBC Se concentration. Se concentration in mammary secretion significantly declined over time with the highest concentrations found in colostrum. No effect of Se source was observed on colostrum or milk Se concentration measured at foaling, or during first mo of lactation. In foals born to mares in the organic group, RBC Se concentration was 170% compared to that of foals born to mares in inorganic group. However, source of maternal Se did not influence IgG concentration in foals. At 1-d of age, foals in the organic group had higher relative gene expression for interferon gamma. However, no significant difference in relative gene expression of the neutrophil cytokines IL-1 and IL-8 at d 1 of age was observed. Foals at 1-mo of age in the organic group had higher relative gene expression for IL-2 when compared with foals in the inorganic group (Montgomery et al., 2012b). Pagan et al. (2007) used 4 mature trained Thoroughbred geldings in a 2-period switch back design trial to evaluate how exercised Thoroughbreds digest and retain 2 forms of Se. Two horses were fed 2.90 mg inorganic Se (averaged 0.41 ppm Se with about 77% selenite). The other two horses were fed 2.76 mg of organic Se yeast 36 (averaged 0.40 ppm Se with about 75% of total Se provided from yeast). In period 1, respective diets were fed for 5 wk. The horses were exercised 6 d/wk in first 4 wk. In the fifth wk, a 5-d digestion trial was conducted. On d 3 of collection, horses completed a standardized exercise test. Feed, feces, urine and blood were analyzed for Se. In period 2, Se supplementation was switched for 3 wk, horses received the same exercise in the first 2 wk, and in the third wk of total collections and an exercise test were conducted. The authors reported horses consuming inorganic Se excreted significantly more fecal Se than those consuming organic Se. Apparent absorption of dietary sodium selenite and organic Se averaged 51.1and 57.3%, respectively. Selenium retention was increased when organic Se was fed. The authors concluded most of the difference in Se retention was the result of increased Se absorption, since there was no difference in average daily urinary Se excretion between treatments. After the exercise test, horses consuming inorganic Se had higher Se excretion as compared to d 1 or 2 of the collection period (Pagan et al., 2007). Richardson et al. (2006) fed 1 of 3 treatments to 18 sedentary 18-mo old stock type horses; 11 geldings, and 7 mares. Treatments consisted of Control with no supplemental Se, totaling 0.15 mg/kg Se, inorganic Se with control in addition to 0.45 mg/kg Se from sodium selenite, or organic Se with control in addition to 0.45 mg/kg Se from Zn-L-selenomethionine to determine the effect of organic and inorganic Se sources on the Se status of horses. Plasma and skeletal muscle Se concentrations and GSH-Px activities in plasma, erythrocytes, and skeletal muscle were determined. Blood was drawn on d 0, 28, and 56. Muscle samples were taken on d 0 and 56. The researchers reported 37 mean plasma and middle gluteal muscle Se concentrations on d 0 were not different among treatments, and significantly increased over the experimental period. Plasma Se concentrations were significantly greater on d 28 and 56 for Se-supplemented horses as compared to control horses. There was a tendency for greater plasma Se concentrations in horses consuming the organic treatment as compared to those consuming the inorganic treatment on d 28. Mean muscle Se concentration was unaffected by treatment. Mean plasma GSH-Px activity increased in horses consuming all treatments throughout the trial. However, this activity was not affected by Se supplementation or source. Mean erythrocyte GSH-Px activity also tended to increase over the experimental period for horses fed all diets. There was a tendency for horses consuming the organic treatment to have greater erythrocyte GSH-Px activity on d 28 as compared with those consuming both the control and inorganic treatments. The authors hypothesized the rapid (less than 4 wk) increase in the erythrocyte GSH-Px activity of horses consuming the organic diet may indicate greater incorporation into erythrocyte GSH-Px. Mean erythrocyte GSH-Px activity of horses consuming inorganic and organic treatments were not different as compared to those consuming control on d 56. Mean skeletal GSH-Px activity significantly decreased over the experimental period for all horses (Richardson et al., 2006). Richardson et al. (2003) sought to determine the effects of Se source and Se status in horses. These researchers compared Se concentrations and GSH-Px activities in blood and skeletal muscle of horses receiving organic and inorganic Se supplementation. Twenty-four 16-mo-old horses were fed 1 of 4 treatments: control containing 0.15 mg 38 Se/kg, inorganic containing 0.6 mg/kg sodium selenite, organic treatment 1 containing 0.6 mg Se/kg, organic treatment 2 containing 0.6 mg Se/kg. All horses received the basal diet during a 28-d acclimation period, and were placed on their respective treatments for a 56-d supplementation period. Blood was drawn on d 0, 28, and 56 of the supplementation period. Plasma was harvested and the RBC fraction was washed and lysed. Muscle biopsies were taken from the middle gluteal muscle on d 0 and 56. The authors reported plasma Se concentrations of the supplemented groups significantly increased from d 0 to 28, plateaued by d 56, and were significantly greater than control on d 28 and 56. Mean plasma Se concentration of those consuming organic treatment 1 was greater than those consuming organic treatment 2 and inorganic on d 28, and continued to be greater than those consuming organic treatment 2 on d 56, and tended to be different from inorganic on d 56. Muscle Se concentrations increased in horses on all treatments from d 0 to 56. Plasma GSH-Px activity fluctuated over time, but was not affected by treatment. Erythrocyte GSH-Px activity significantly increased between d 0 and 28 in organic treatment 1 and was significantly greater than the other 3 treatments on d 28 (Richardson et al., 2003). Janicki et al. (2001) used 15 mares to determine if Se form or level had an effect on mare and foal Se status, GSH-Px activity, and antibody titer to influenza. The mares were blocked by expected foaling date and assigned to 1 of 3 treatments; 1 mg sodium selenite, 3 mg sodium selenite, or 3 mg Se-yeast. The respective diets were fed for 55 d pre-foaling, to 56 d post-foaling. Mare blood samples were taken prior to supplementation, every 2 wk until foaling, immediately post-foaling, and every wk for 56 39 d. Colostrum samples were taken post-foaling, and milk samples every 2 wk for 56 d. Foal blood samples were obtained prior to suckling, at 12 h, and 2, 4, 6, and 8 wk of age. Selenium concentration, GSH-Px activity and serum influenza antibody titers were analyzed. Serum Se was significantly greater in mares receiving 3 mg organic Se as compared to other treatments at post-foaling, wk 4 and 8. Selenium in colostrum and milk was greater in mares receiving 3 mg organic Se as compared to other treatments. At 12 h, serum Se in foals from mares receiving organic Se was significantly greater than foals from mares receiving 1 mg inorganic Se. At 2, 4, 6, and 8 wk, serum Se in foals from mares receiving organic was significantly greater than those receiving other treatments. At 2 wk, serum Se was also significantly greater in foals from mares receiving 0.3 ppm inorganic as compared to foals from mares receiving 0.1 ppm inorganic. At 6 wk, GSHPx activity in foals from mares receiving 0.3 ppm was significantly greater than foals from mares receiving 0.1 ppm. At 8 wk, GSH-Px activity was significantly greater in foals from mares receiving 0.3 ppm organic Se compared to foals from mares receiving 0.1 ppm inorganic Se. At 6 wk, influenza antibodies to A2/KY/92 were significantly greater in foals from mares receiving 0.3 ppm than foals from mares receiving 0.1 ppm. At 2, 4, and 8 wk, influenza antibodies to A2/KY/92 tended to be greater in foals from mares receiving 0.3 ppm. At 8 wk, influenza antibodies to A1/Prague were significantly greater in foals from mares receiving 0.3 ppm than in foals from mares receiving 0.1 ppm (Janicki et al. 2001). White et al. (2011) hypothesized that Se supplementation above NRC recommendations would enhance selenoprotein activity and reduce oxidative damage in 40 horses following a prolonged exercise bout. Twelve mature, untrained Thoroughbreds were fed 1 of 2 respective diets for 36 d; 0.1 mg/kg sodium selenite or 0.3 mg/kg sodium selenite. On d 35, horses were subjected to 120 min of submaximal exercise with a mean heart rate of 135 bpm. Blood samples were taken at d 0, after 34 d of Se supplementation, and on d 35 immediately after exercise; and 6 and 24 h post-exercise. Samples were analyzed for serum Se, plasma and RBC lysate GSH-Px activity, serum creatine kinase, and total lipid hydroperoxides. Muscle biopsies were taken d 0 and after 34 d of Se supplementation; and at 6 and 24 hr post-exercise on d 35 and 36 for determination of GSH-Px and thioredoxin reductase activities. The authors reported supplementation with 0.3 ppm significantly increased serum Se, but had no effect on GSH-Px activity in plasma, RBC lysate or muscle in horses at rest. Serum creatine kinase was not different between horses consuming 0.3 ppm and 0.1 ppm, but significantly increased in response to prolonged exercise, indicating excessive reactive oxygen species generation and tissue damage. Serum lipid hydroperoxidase was significantly affected in horses fed 0.3 ppm, indicating these horses were possibly better equipped to combat the oxidative load. Glutathione peroxidase activity significantly increased in plasma and significantly decreased in RBC lysate after prolonged exercise in all treatments. A significant treatment by time interaction was observed for RBC lysate and muscle GSH-Px activity. Compared to enzyme activity before exercise, RBC GSH-Px activity was significantly lower immediately after exercise in horses fed 0.3 ppm, whereas a similar decline wasn’t significantly observed until 6 h post-exercise in those consuming other treatments. Muscle GSH-Px activity was significantly elevated over pre-exercise levels at 6 h post- 41 exercise in those consuming 0.3 ppm and remained unchanged in horses consuming 0.1 ppm (White et al., 2011). Shellow et al. (1985) fed 20 mature geldings; 10 Quarter Horses and 10 Thoroughbreds to determine the influence of dietary Se on whole blood and plasma Se levels, and GSH-Px activity. All horses were fed a basal diet of 50% concentrate containing 0.077 mg/kg of naturally occurring Se, 50% Timothy hay containing 0.43 mg/kg of naturally occurring Se for a total of 0.060 mg/kg of Se for at least a 4-wk preliminary period. At the beginning of the repletion phase, horses were supplemented with 0, 0.05, 0.1, or 0.2 ppm Se as sodium selenite with final Se concentrations of 0.06, 0.11, 0.16, and 0.26 ppm. Blood was drawn at weekly intervals for 2 wk before supplementation, and at 12 wk following inclusion of supplemental Se in diet. The authors reported a significant increasing linear trend in plasma Se concentration over time. At wk 0, there was no significant difference observed among treatment groups. Supplementation of the diets with Se significantly increased plasma Se above that of the control group by the wk 2 of the trial. By wk 5, there were significant differences in plasma Se concentration between horses in the control group, and those receiving 0.05, 0.10, or 0.2 ppm supplemental Se in their diet. Plasma Se concentrations for horses receiving the 2 highest levels of Se were significantly greater than those receiving 0.05 ppm supplemental Se. There were no significant differences in plasma Se concentrations between those receiving 0.10 and 0.20 supplemental Se. Little change in plasma Se concentration was observed in Se-supplemented horses after 5 wk. Plasma Se reached plateaus of 0.1 to 0.11, 0.12 to 0.14, and 0.13 to 14 µg/mL in horses supplemented with 42 0.5, 0.1, and 0.2 ppm Se, respectively. Maximum response in whole blood Se concentration occurred by wk 6 with no further significant changes throughout the remainder of the trial. Whole blood Se reached plateaus of 0.16 to 0.18, 0.19 to 0.21, and 0.17 to 0.18 µg/mL in groups supplemented with 0.05, 0.1, and 0.2 ppm Se, respectively. Plasma GSH-Px activity was not significantly affected by dietary treatment, although an increasing trend in activity over time was observed (Shellow et al., 1985). Calamari et al. (2010) compared the effects of organic and inorganic Se supplements on hematological profiles, enzyme activities, plasma oxidative status, and inflammatory status. Twenty-five slightly exercised, mature Italian Saddle Horses were used in the trial. All horses were fed the control diet for 56 d to allow for diet adaption. The trial utilized 5 treatments; negative control, 0.2 mg organic Se/kg, 0.3 mg organic Se/kg, 0.4 mg organic Se/kg, or positive control containing 0.3 mg inorganic Se/kg. Blood was drawn d 0, 28, 56, 84, and 112. Authors reported plasma metabolites related to energy and protein metabolism and mineral metabolism were not affected by Se source or dose. Inflammatory status did not appear to be affected by Se source and dose. Horses consuming 0.3 mg organic Se and 0.4 mg organic Se had significantly lower total plasma antioxidants than horses consuming control, and horses consuming Se yeast supplement had significantly lower total plasma antioxidants as compared to those consuming comparable dose of selenite. Total plasma antioxidants decreased linearly as Se yeast supplementation increased. Total white blood cells was not affected by treatment. Number of lymphocytes tended to increase slightly as Se-yeast supplementation increased. Greater numbers of lymphocytes were observed in those consuming 0.3 mg 43 organic Se and 0.4 organic Se as compared to those consuming 0.3 mg inorganic Se (Calamari et al., 2010). In a companion study to Calamari et al. (2010), Calamari et al. (2009) evaluated the effects of dietary Se sources on Se status, GSH-Px activity, and thyroid hormone status. Blood was analyzed for RBC GSH-PX activity, whole blood Se, packed cell volume, and plasma Se concentrations. The authors reported horses consuming all treatments supplemented with Se had significantly greater total Se concentrations in whole blood and plasma when compared with those consuming the negative control. A linear dose effect and source effect were observed for total Se in blood. Whole blood Se concentrations were significantly higher in treatments supplemented with greater doses of Se, and in those horses supplemented with Se yeast at d 84 and 112 as compared to those receiving a comparable dose of selenite. Total Se in blood in horses consuming all treatments supplemented with Se was greater as compared to those consuming the control from d 28 to the end of the study. The 16-wk experimental trial was not sufficient for horses consuming all treatments to achieve asymptotic, steady-state Se concentrations in whole blood. There was a significant linear dose effect for plasma Se, with greater values in those consuming treatments supplemented with greater doses of Se. Selenium source did not affect plasma Se concentrations. Plasma Se concentrations achieved asymptotic steady state within 75 to 90 d of the beginning of the study in all supplemented groups. Plasma Se in all treatments appeared to increase by 50 to 60% at d 28, 85 to 93% at d 56, and almost 100% at d 84. Correlations were observed between whole blood Se and plasma Se (r = 0.83). Horses consuming all treatments supplemented with Se had 44 significantly greater GSH-Px activity when compared with those consuming the negative control. Linear and quadratic dose effects were observed on GSH-Px activity between those consuming low and intermediate doses of Se yeast, or between those consuming the least and greatest dose of Se yeast. Horses consuming all supplemented treatments had significantly greater GSH-Px activity as compared to those consuming negative control from d 56 to study completion. Asymptotic GSH-Px activity did not appear to have been achieved in any of the horses consuming Se-supplemented treatments after completion of the 16-wk experimental period. There was a correlation observed between GSH-Px activity and whole blood Se (r = 0.86), and GSH-Px activity and plasma Se (r = 0.75). Plasma GSH-Px activity was significantly greater in those consuming Se yeast and sodium selenite when compared with those consuming the negative control. The rate of increase in the proportion of total Se as selenomethionine over time was significantly greater in whole blood and plasma in those horses consuming 0.3 mg organic Se as compared with those consuming a comparable dose of selenite. Selenocysteine was the predominant form of Se in blood and accounted for 79.1 and 71.4% of total Se in whole blood and plasma, respectively, whereas selenomethionine only accounted for 15.2 and 10.0% (Calamari et al., 2009). Brummer et al. (2013) evaluated the impact of change in Se status on measures of antioxidant status and oxidative stress in adult horses during Se depletion and repletion. Twenty-eight horses were divided into 4 treatment groups. During the 196-d depletion period, 3 treatments provided 0.06 mg Se/kg DM and 1 treatment provided 0.12 mg Se/.kg DM. During the 189 d repletion period, horses were assigned to 1 of 4 treatments; 45 7 horses continued to consume 0.06 mg Se/kg, 7 horses continued to consume 0.12 mg Se/kg, 7 horses were fed Se-yeast at 0.3mg/kg, and 7 horses were fed sodium selenite 0.3mg/kg. Horses were not exercised during this trial. The 196-d depletion period was selected on the basis of the Se status of the horses as determined by the monthly blood samples obtained for whole blood Se and GSH-Px activity evaluation and compared with published adequate reference range for whole blood Se between 180 to 240 ng/mL, and whole blood GSH-Px activity between 40 to 160 enzyme units/g Hb (Stowe, 1998). Blood samples were taken at the start of each phase and on d 84, 140, 168, and 196 of depletion and d 28, 56, 154, and 189 of repletion. The authors reported whole blood Se concentrations were affected by the interaction of treatment and time. The authors also appeared to erroneously report significant main effects of treatment and time during the depletion phase. Whole blood Se concentrations in horses consuming 0.06 mg Se decreased until d 140 then stabilized and were significantly less than those consuming 0.12 mg Se. Selenium concentrations in horses consuming 0.12 mg Se stabilized within first 84 d of depletion. At the end of depletion, there was a significant difference in whole blood Se between those consuming the 2 treatments. Whole blood GSH-Px activity in those consuming 0.12 mg Se decreased during first 84 d and then stabilized. Glutathione peroxidase activity in those consuming 0.06 mg Se decreased throughout the depletion period. Mean GSH-Px activity was less in those consuming 0.06 mg Se as compared to those consuming 0.12 mg Se at d 196. A positive correlation existed between whole blood Se and GSH-Px activity (r = 0.63). During repletion, there was a significant treatment by time interaction. The authors also appeared to erroneously report significant 46 main effects of treatment and time. Within 28 d of starting the repletion phase, whole blood Se was similar in those consuming 0.12 mg/kg Se, 0.3 mg/kg organic Se, and 0.3 mg/kg inorganic Se, but greater than those consuming 0.06 mg Se/kg DM. At 154 d, whole blood Se concentrations in those consuming 0.3 mg organic Se/kg DM and 0.3 mg inorganic Se/kg DM were significantly greater than those consuming 0.12 mg Se/kg DM. Whole blood Se did not increase from d 154 to 189 in either those consuming 0.3 mg inorganic Se or 0.3 mg organic Se. Whole blood GSH-Px activity during the repletion phase was affected by the interaction of treatment by time. The authors also appeared to erroneously report main effects of treatment and time. At d 154, GSH-Px activity in those consuming 0.3 mg inorganic Se were comparable to those consuming 0.3 mg organic Se, and appeared greater than those consuming 0.12 mg Se. A strong positive correlation existed between whole blood Se and GSH-Px activity (r = 0.82). The authors theorized the current Se recommendation of 0.1 mg Se/kg DM must be close to the minimum Se requirement for mature idle horses. The authors stated that whole blood GSH-Px is responsive to dietary Se intakes above 0.1 mg/kg and supplementation of 0.1 ppm Se may not allow for maximum GSH-Px activity in the horse. An increase in whole blood GSHPx activity required 56 d of repletion in comparison with a response time of 28 d of repletion for whole blood Se. These differences in response times are most likely due to the incorporation of GSH-Px in recently formed red blood cells. In this study, both the depletion and repletion phases exceeded the period needed for the complete turnover of the RBC population, thus allowing enough time for the incorporation of GSH-Px into recently formed RBC. The authors suggested that the lack of detectable change in GSH- 47 Px activity in response to supplementation levels above 0.1 mg/kg DM could be due to the shorter experimental periods used in some studies compared with the length of time required for complete RBC turnover in the horse (Brummer et al., 2013). Statement of the Problem Selenium deficiency in horses is a prominent problem in the Pacific Northwest, and other areas with extremely deficient soils. Although the current NRC recommendation is 0.1 ppm (NRC, 2007), studies have reported conflicting results on the possible benefits of 0.3 ppm Se supplementation. Data from previous studies indicate that in order to see a benefit of 0.3 Se supplementation, trials need to be conducted for at least 112 d, and whole blood Se concentrations and erythrocyte GSH-Px activity should be used to evaluate Se status. Little data exists as to the rate of Se depletion in horses consuming Se-deficient diets. Brummer et al. (2013) drew blood on d 0, 84, 140, 168, and 196 d of a depletion period. It is difficult to determine a precise depletion curve with so much time in between blood sampling. The objectives of the current study was to 1) determine the depletion rate of Se in horses consuming a Se-deficient diet and 2) compare the effects of two different levels of organic Se supplementation on Se repletion as indicated by whole blood Se concentrations and erythrocyte GSH-Px activity in moderately exercised horses. 48 Chapter III MATERIALS AND METHODS Experimental Design Twelve mature, stock-type geldings were used in a 2-part study. First, to determine the effects of feeding a low-Se diet, containing 23% of the NRC recommended amount of dietary Se, on whole blood Se concentrations and erythrocyte glutathione peroxidase (RBC GSH-Px) activity over a 112-d depletion phase. Secondly, the geldings previously depleted to an average whole blood Se concentration of 109 ng Se/mL received 1 of 2 levels of Se organic supplementation, in an effort to compare the rate of repletion between horses consuming a supplement containing 0.1 vs. 0.3 ppm organic Se. Horses were divided into 4 groups of 3 and housed in 6 x 20 m pens at the West Texas A&M University Horse Center. Throughout the trial, horses were classified as moderately exercised (NRC, 2007), as they were used in horsemanship classes and equestrian team practices 3 to 5 times/ wk. Horses were fed individually in 2 x 5 m stalls twice daily at 0600 and 1700, and were allowed 3 h to consume rations before being turned out in 6 x 20 m pens. Supplement and hay was weighed out prior to feeding. Intakes and orts were weighed and recorded throughout the trial. Routine farrier work, vaccinations, and deworming were consistent with West Texas A & M University protocols. Salt blocks were provided ad libitum throughout the study. Body condition 49 scores were assigned, and BW was measured at 0500 in 28-d intervals on a platform scale (LBS Inc. Garden City, KS). Jugular venous blood was drawn on d 0, 28, 56, 84 and 112 of the depletion phase, and on d 14, 28, 56, 84, 96 and 112 of the repletion phase at 0500. On day 2 of the depletion phase, one gelding died due to natural causes unrelated to the study. On d 8 of the depletion phase, another gelding was removed from the trial due to refusal to consume the supplement. During the repletion phase, 10 horses were stratified by whole blood Se concentrations at d 84 of depletion, and evenly assigned to 1 of 2 repletion treatments. The 10 remaining horses ranged in age from 9 to 19 yr with a mean age of 14 yr. Trial protocol was approved by West Texas A & M University Institutional Animal Care and Use Committee. Diets At the onset of the trial (d 0), venous blood was drawn, weights recorded, and BCS assigned. Horses were fed the depletion diet for 112 d. During the repletion phase, horses were fed their respective treatments for 112 d. Diets were fed in amounts to attempt maintenance of BCS of 5.0. All horses consumed a basal diet of Orchard Grass Hay fed at 1.25 to 2.34% BW/d. Hay was grown in extremely Se-deficient soils (< 0.5 ppm soil Se; Koller and Exon, 1986) in Central Oregon. In addition the hay was fertilized with (NH4)2SO4, a commonly used fertilizer, which decreases Se uptake into the plant due to the antagonistic relationship between S and Se. During the depletion phase, diets consisted of the Orchard Grass Hay top dressed with 57 g of vitamin/mineral supplement with no added Se. At d 0 of the repletion phase, horses were stratified by whole blood Se 50 concentrations at d 84 of depletion and then evenly divided and assigned one of two supplemental Se treatments. Diets consisted of the Orchard Grass Hay along with supplemental Se contained in the vitamin/mineral supplement that was top-dressed at 2 concentrations; 0.1 ppm Se (SE1; n = 5); or 0.3 ppm Se (SE3; n = 5). Hay was analyzed for DE, CP, ADF, and NDF. Feed analysis for Orchard Grass Hay is presented in Table 1. Samples of all supplements (No Se, SE1 and SE3) and hay were analyzed for Se concentration at the Michigan State University Diagnostic Center for Population and Animal Health (DCPAH; Lansing, MI; Table 2)). Table 1. Feed Analysis for Orchard Grass Hay (DM) Crude Protein, % 13.1 Acid Detergent Fiber, % 39.1 Neutral Detergent Fiber, % 58.2 DE Mcal/kg 2.18 Table 2. Selenium Analysis for Supplements and Hay (DM) Orchard Grass Hay (mg/kg) 0.01 No Se Added Supplement (mg/1.9 oz) 0.14 1 ppm Se Supplement (mg/1.9 oz) 1.05 3 ppm Se Supplement (mg/1.9 oz) 3.52 Sample Collections, Preparation, and Handling Venous blood samples were collected at 0500 prior to the morning feeding on d 0, 28, 56, 84, 112 of the depletion phase, and d 14, 28, 56, 84, 96 and 112 of the repletion phase via jugular veni-puncture using two 3-ml lavender-top Vacutainer™ tubes containing EDTA. After blood collection, sample tubes were slowly inverted 8 times, and then placed on ice. One tube of whole blood from each horse was shipped on ice at 0900 51 in specialized insulated containers purchased from DCPAH (Lansing, MI) to DCPAH (Lansing, MI) for Se analysis. The remaining tubes were transported to the West Texas A & M University CORE laboratory (Canyon, TX) to be prepared for GSH-Px analysis and stored. For GSH-Px analytical preparation, four 500 uL whole blood samples from each tube were transferred into 2 mL micro-centrifuge tubes. The 2 mL tubes were centrifuged at 2500 x g for 5 min. After separation, plasma was discarded. Remaining RBC were washed with 500 uL of 0.9% NaCl solution, vortexed, and centrifuged again at 2500 x g for 5 min. Saline supernatant was removed and discarded. Erythrocytes were lysed with 1 mL of ice-cold distilled, deionized water, vortexed, closed and stored upright at -80◦C until RBC GSH-Px activity analysis. Laboratory Analysis Inductively-Coupled Plasma Mass Spectrometry. An inductively-coupled plasma mass spectrometry (ICP-MS) 7500ce (Agilent Technologies, Santa Clara, CA) was used to determine concentrations of Se in whole blood and feed samples at DCPAH (Lansing, MI). For preparation of whole blood Se analysis, 200 uL of whole blood was mixed with 5 mL of diluent containing; NH4OH, butanol, EDTA, Triton-x 100, and 5 internal standards. Samples were analyzed for Se using the ICP-MS on “non-gas” mode, and Se concentration were reported in ng Se/mL whole blood. Glutathione Peroxidase Activity Assay. For the analysis of GSH-Px activity, RBC samples were thawed in the West Texas A&M University CORE Laboratory (Canyon, TX) and analyses performed in the West Texas A&M University RHIL Laboratory (Canyon, TX) using an EPIC spectrophotometer (Palmyra, WI). The spectrophotometer 52 was set to the “kinetic option”, and GSH-Px activity was determined at a wavelength of 340 nm. The reading were collected every 30 s for 3 min. Each well of the assay kit contained 75 uL assay buffer, 75 uL NADPH reagent, and 15 uL diluted sample. Erythrocyte samples were diluted using 7 uL sample and 64 uL assay buffer, and were then plated in their respective wells, running all blood draw from each respective horse on the same plate twice. Two controls were created, a high (225 mU/ mL), and low control (112.5 mU/ mL). Blank standard, low control and high control were then plated using 15 uL of each. Using the multi-channel pipette, 75 uL tert-butyl was added to each column (12 columns per plate), and the respective column analyzed. Once samples were analyzed, the rate of decrease in absorbance at 340 nm per min was calculated. The net rate for the sample was calculated by subtracting the rate observed for the water blank. The net absorbance/min was calculated as: 1 mU/mg Hb = 1 nmol NADPH/mL = (A340/min)/ 0.00622 The concentrations were then corrected for the dilution of the sample (10:90 dilution), and the dilution of the RBC and deionized water (1:5). The units of activity in original sample are expressed mU/mg Hb. Statistical Analysis Data for depletion phase whole blood Se concentrations and RBC GSH-Px activity was analyzed using non-linear regression analysis (SPSS Version 21, 2012). Data for repletion phase whole blood Se concentrations were adjusted by subtracting d 112 of 53 depletion values from all other days of repletion for each horse to determine changes in whole blood Se concentrations, and was analyzed using non-linear regression analysis (SPSS Version 21, 2012). The slope of the non-linear regression curves were compared using the t-test. Erythrocyte GSH-Px activity was also analyzed using non-linear regression analysis (SPSS Version 21, 2012). Data for whole blood Se concentrations was also analyzed using the t-test assuming equal variances (Excel, 2010) to determine differences between treatments within time. Data for repletion phase RBC GSH-Px activity was analyzed using t-tests assuming unequal variances (Excel, 2010). Significant differences between treatments were declared at P ≤ 0.05. Trends for differences between treatments were declared at P ≤ 0.10. 54 Chapter IV RESULTS AND DISCUSSION Mean Se intakes for horses consuming Se depletion diet, 0.1 ppm supplemental Se treatment (SE1), and 0.3 ppm supplemental Se treatment (SE3) are presented in Table 3. During the depletion phase, horses consuming overall mean of 23% of NRC recommendations for Se. Horses consuming SE1 consumed an overall mean of 131% of NRC recommendations for Se. Horses consuming SE3 consumed an overall mean 421% of NRC recommendations for Se. Table 3. Mean Selenium Intake (mg/kg DM) Se depletion 0.023 SE1 0.131 SE3 0.421 Depletion Phase Whole Blood Selenium Concentrations Regression At initiation of the depletion phase (d 0), overall mean whole blood Se concentrations were 187.4 ± 7.36 ng Se/mL. Individual whole blood Se concentrations during the depletion phase can be observed in Figure A-1 in the Appendix. Overall whole blood Se concentrations in horses consuming 23% of NRC Se recommendations depleted at a non-linear rate. A non-linear regression equation was developed {predicted whole 55 blood Se concentration = 184.95 * (1 * EXP (-0.005 * day))} and can be observed in Figure 1. The corrected r-squared of Se depletion was 0.863. Using the non-linear equation, forecasting of whole blood Se concentrations was estimated to d 250 (Figure 2). If the predicted equation was proven, horses consuming 23% of NRC recommendations for Se would become clinically deficient within 209 d from initiation of the depletion period. Non-linear regression equations were developed because of biological reasons in the body. The rate of depletion slows over time. Linear regression equations would possibly predict clinical Se deficiency too quickly. There are no published studies reporting the non-linear regression of a Se depletion phase. However, data for Se depletion in horses has been reported. Brummer et al. (2013) reported horses fed 60% of the NRC recommendation of Se for 196 d had significantly lower whole blood Se concentrations at d 140 and 196 as compared to horses fed 0.12 mg Se/kg DM. Furthermore, Brummer et al. (2013) reported significantly lower whole blood Se concentrations in horses fed 0.06 mg Se/kg DM at d 84, 140, 168, and 196 as compared to d 0. Whole blood Se concentrations in horses receiving 0.06 mg Se/kg DM was significantly lower at d 140, 168, and 196 as compared to d 84. However, the authors reported no significant differences in whole blood Se concentrations between d 140, 168 and 196. In the current study, overall mean whole blood Se concentrations at the initiation of depletion were 187.4 ng Se/mL, as compared to Brummer et al. (2013), who reported overall mean whole blood Se concentrations of 251.7 ng Se/mL. In addition, at the end of the depletion phase (d 112) in the current study, overall mean whole blood Se concentrations in horses consuming 23% of NRC recommendations were 109 ng Se/mL. 56 57 58 Brummer et al. (2013) reported whole blood Se concentrations in horses consuming 60% of NRC recommendations of 173.5 ng Se/mL at d 140, and 165.1 ng Se/mL at d 196. Repletion Phase Whole Blood Selenium Concentrations Regression At the initiation of the repletion phase (d 112 of depletion), horses were stratified according to whole blood Se concentrations at d 84 of depletion, and assigned to 1 of 2 Se supplement treatments. Horses assigned to SE1 had overall mean whole blood Se concentrations of 108.2 ± 12.2 ng Se/mL. Horses assigned to SE3 had overall mean whole blood Se concentrations of 109.8 ± 11.2 ng Se/mL. Individual whole blood Se concentrations during the repletion phase can be observed in Figure A-2 in the Appendix. Non-linear regression equations were developed using adjusted whole blood Se concentrations, calculated by subtracting d 0 of repletion values from d 14, 28, 56, 84, 96 and 112. Adjusted whole blood Se concentrations in horses consuming SE1 repleted at a non-linear rate. A non-linear regression equation was developed {predicted change in whole blood Se concentration = 20.911 * (1 - EXP (-0.062 * day))} and is shown graphically in Figure 3. The corrected r-squared of Se repletion in horses consuming SE1 was 0.550. Adjusted whole blood Se concentrations in horses consuming SE3 also repleted at a non-linear rate. A non-linear regression equation was developed {predicted change in whole blood Se concentration = 38.249 * (1 - EXP (-0.070 * day))} and is shown graphically in Figure 4. The corrected r-squared of Se repletion in horses consuming SE3 was 0.779. 59 60 61 When comparing the non-linear regression equations of the change in whole blood Se concentrations in horses consuming SE1 and SE3, it appeared that horses consuming SE3 repleted at a faster rate, and maintained higher whole blood Se concentrations. Non-linear regressions over 112-d Se repletion are shown in Figure 5. There are no studies reporting non-linear regression analysis of whole blood Se concentrations during a repletion phase. However, Calamari et al. (2009) reported linear regressions of whole blood Se concentrations {total blood Se, ng/g = 1.472 ± 0.278 x time (d) + 179.8 ± 19.1} in horses consuming 0.18 mg Se yeast/kg DM, {total blood Se, ng/g = 2.186 ± 0.267 x time (d) + 195 ± 19.7} in horses consuming 0.29 mg Se yeast/kg DM, { total blood Se, ng/g = 2.167 ± 0.301 x time (d) + 232.6 ± 23.3} in horses consuming 0.39 mg Se yeast/kg DM, and {total blood Se, ng/g = 2.186 ± 0.267 x time (d) + 195 ± 19.7} in horses consuming 0.29 mg Na selenite/kg DM. Calamari et al. (2009) also reported a quadratic regression for plasma Se concentrations {plasma Se, ng/g = - 0.00697 ± 0.00537 x time (d)2 + 1.2991 ± 0.6269 x time (d) + 97.1 ± 14.8} in horses consuming 0.18 mg Se yeast/kg DM, { plasma Se, ng/g = -0.01727 ± 0.00356 x time (d)2 + 2.5768 ± 0.4157 x time (d) + 89.2 ± 9.8} in horses consuming 0.29 mg Se yeast/kg DM, { plasma Se, ng/g = -0.01556 ± 0.00417 x time (d)2 + 2.4917 ± 0.4865 x time (d) + 104.2 ± 11.5} in horses consuming 0.39 mg Se yeast/kg DM, and { plasma Se, ng/g = - 0.01478 ± 0.00443 x time (d)2 + 2.2985 ± 0.5178 x time (d) + 81.2 ± 12.2} in horses consuming 0.29 mg Na selenite/kg DM. 62 63 In the current study, non-linear regression equations were developed because of biological reasons in the body. The rate of repletion slows over time. Linear regression equations would possibly predict clinical Se deficiency too quickly. Upon analysis of the repletion data, a decrease in whole blood Se concentrations appears to occur at d 84 in both SE1 and SE3, before returning to expected values at d 96 and 112. Overall mean whole blood Se concentrations at d 84 were below concentrations at d 28 and 56. The reason for this dramatic, and unexpected, decrease in Se concentrations at d 84 of repletion is unknown. Possible causes include a difference in sample handling during shipment of samples to DCPAH, or differences, however slight, in laboratory analysis of the samples. However, d 84 data from this study partially agree with Shellow et al. (1985) who reported horses consuming 0.16 ppm Se had whole blood Se concentrations of 0.140 ug Se/mL at d 56, and had decreased concentrations of 0.138 ug Se/mL at d 63, although this decrease was not statistically significant. Further, the authors reported horses consuming 0.26 ppm Se had whole blood Se concentrations of 0.142 ug Se/mL at d 63, and these values declined slightly to 0.135 ug Se/mL at d 70, although again, the decrease was not statistically significant. Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 0 of Repletion There was no significant effect of treatment observed on overall mean whole blood Se concentrations in horses consuming SE1 (mean = 108.2 ng Se/mL) and SE3 (mean = 109.8; at d 0 of repletion (P = 0.417; Figure 6). 64 65 Data for whole blood Se concentrations at d 0 of repletion in the current study agree with that of Brummer et al. (2013), who reported no differences between horses previously fed 0.06 mg Se/kg DM for 196 d at d 0 of repletion. The results also agree with Calamari et al. (2009), who reported no differences in whole blood Se concentration in horses fed 0.085 mg Se/kg DM for 2 mo at d 0 of repletion. Richardson et al. (2003) and Richardson et al. (2006) reported no significant differences at d 0 of repletion in plasma Se concentrations in horses fed 0.15 mg Se/kg DM for 28 d. The d 0 of repletion results of this study disagree with Shellow et al. (1985), who reported significantly lower whole blood Se concentrations in horses fed 0.06 ppm Se for at least 4 wk, in horses consuming 0.06 and 0.26 ppm Se as compared to horses consuming 0.16 ppm Se, and significantly higher whole blood Se concentrations in horses consuming 0.11 ppm Se as compared to horses consuming 0.26 ppm Se. Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 14 of Repletion A significant effect of treatment was observed on overall whole blood Se concentrations in horses consuming SE1 and SE3 at d 14 (Figure 7). Horses consuming SE3 (mean = 129.8 ng Se/mL) had significantly greater (P = 0.032) whole blood Se concentrations as compared to horses consuming SE1 (x bar = 115.6 ng Se/mL). Data for whole blood Se concentrations at d 14 of the current study partially agree with that of Shellow et al. (1985), who reported significantly lower whole blood Se concentrations at d 14 in horses consuming 0.06 and 0.11 ppm Se as compared to horses consuming 0.16 ppm Se. However, these authors also reported significantly higher whole 66 67 blood Se concentrations in horses consuming 0.16 ppm Se as compared to horses consuming 0.26 ppm Se. The results of the current study disagree with Janicki et al. (2001), who reported no difference in serum Se concentrations at d 14 of supplementation in pregnant mares supplemented with 1 or 3 mg Se/d. Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 28 of Repletion A significant effect of treatment was observed on overall whole blood Se concentrations in horses consuming SE1 and SE3 at d 28 (Figure 8). Horses consuming SE3 (mean = 147.0 ng Se/mL) had significantly greater (P = 0.007) whole blood Se concentrations as compared to horses consuming SE1 (mean = 129.4 ng Se/mL). Data for whole blood Se concentrations at d 28 of the current study agree with Richardson et al. (2006), who reported significantly higher plasma Se concentrations in horses consuming 0.45 mg organic and inorganic Se/kg DM as compared to horses consuming 0.15 mg Se/kg DM. Additionally, these results agree with Richardson et al. (2003), who reported plasma Se concentrations were significantly greater at d 28 in horses consuming 0.6 mg organic Se/kg DM as compared to horses consuming 0.15 mg Se/kg DM. The results of this study partially agree with that of Shellow et al. (1985), who reported significantly higher whole blood Se concentrations at d 28 in horses consuming 0.11 ppm Se, 0.16 ppm Se, and 0.26 ppm Se as compared to horses consuming 0.06 ppm Se. Data from this study also partially agrees with Calamari et al. (2009) who reported horses consuming 0.18 mg organic Se/kg DM, 0.29 mg organic Se/kg DM, 0.39 mg organic Se/kg DM, and 0.29 mg inorganic Se/kg DM had significantly greater whole 68 69 blood Se concentrations as compared to horses consuming 0.085 mg Se/kg DM. Furthermore, the researchers reported horses consuming 0.39 mg organic Se/kg DM, and 0.29 mg inorganic Se/kg DM had significantly greater whole blood Se concentrations as compared to horses 0.18 mg organic Se/kg DM. Horses consuming 0.39 mg organic Se/kg DM had significantly greater whole blood Se concentrations as compared to horses consuming 0.29 mg inorganic Se/kg DM at d 28 (Calamari et al., 2009). The results of the current study disagree with Janicki et al. (2001), who reported no difference in serum Se concentrations at d 28 of supplementation in pregnant mares supplemented with 1 or 3 mg Se/d. The results of this study disagree with Brummer et al. (2013), who reported no significant differences in whole blood Se concentrations in horses consuming 0.12 mg Se/kg DM, 0.3 mg inorganic Se/kg DM, or 0.3 mg organic Se/kg DM at d 28 of repletion. A possible explanation for the differences in results observed in the current study and that of Brummer et al. (2013) is the horses in Brummer’s study were consuming 60% of NRC requirements and mean whole blood Se was much higher (165.1 ng/mL) at the end of depletion as compared to the horses used in the current study (109 ng Se/mL). Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 56 of Repletion A significant effect of treatment was observed on overall whole blood Se concentrations in horses consuming SE1 and SE3 at d 56 (Figure 9). Horses consuming SE3 (mean = 153.2 ng Se/mL) had significantly greater (P = 0.004) whole blood Se concentrations as compared to horses consuming SE1 (mean = 135.0 ng Se/mL). 70 71 Data for whole blood Se concentrations at d 56 of the current study agree with Richardson et al. (2006), who reported significantly higher plasma Se concentrations in horses consuming 0.45 mg organic and inorganic Se/kg DM as compared to horses consuming 0.15 mg Se/kg DM. Additionally, these results agree with Richardson et al. (2003), who reported plasma Se concentrations were significantly greater at d 56 in horses consuming 0.6 mg organic Se/kg DM as compared to horses consuming 0.15 mg Se/kg DM. Data from this study also agree with Janicki et al. (2001), who reported significantly greater serum Se concentrations at d 56 of supplementation in mares supplemented with 3 mg organic Se/d as compared to mares consuming 3 mg inorganic Se/d and 1 mg inorganic Se/d. The results of this study partially agree with that of Shellow et al. (1985), who reported significantly higher whole blood Se concentrations at d 56 in horses consuming the 0.11, 0.16, and 0.26 ppm Se as compared to horses consuming 0.06 ppm Se. Shellow et al. (1985) also reported significantly greater whole blood Se concentrations in horses consuming 0.16 and 0.26 ppm Se as compared to horses consuming 0.11 ppm Se. Data from this study also partially agree with Calamari et al. (2009), who reported at d 56, horses consuming 0.18 mg organic Se/kg DM, 0.29 mg organic Se/kg DM, 0.39 mg organic Se/kg DM, and 0.29 mg inorganic Se/kg DM had significantly greater whole blood Se concentrations as compared to horses consuming 0.085 mg Se/kg DM. Furthermore, the researchers reported horses consuming 0.29 mg organic Se/kg DM and 0.39 mg organic Se/kg DM had significantly greater whole blood Se concentrations as compared to horses 0.18 organic Se/kg DM. Horses consuming 0.39 organic Se/kg DM had significantly greater whole blood Se concentrations as compared 72 to horses consuming 0.29 mg inorganic Se/kg DM at d 56 (Calamari et al., 2009). The results of this study disagree with Brummer et al. (2013), who reported no significant differences in whole blood Se concentrations in horses consuming 0.12 mg Se/kg DM, 0.3 mg inorganic Se/kg DM, and 0.3 mg organic Se/kg DM at d 56. Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 84 of Repletion A significant effect of treatment was observed on overall mean whole blood Se concentrations in horses consuming SE1 and SE3 at d 84 (Figure 10). Horses consuming SE3 (mean = 136.4 ng Se/mL) had greater (P = 0.001) whole blood Se concentrations as compared to horses consuming SE1 (mean = 120.0 ng Se/mL). Data for whole blood Se concentrations at d 84 of the current study agree with Janicki et al. (2001), who reported significantly greater serum Se concentrations at d 84 of supplementation in mares supplemented with 3 mg organic Se/d as compared to mares consuming 3 mg inorganic Se/d and 1 mg inorganic Se/d. The results of this study partially agree with that of Shellow et al. (1985), who reported significantly higher whole blood Se concentrations at d 84 in horses consuming 0.11, 0.16, and 0.26 ppm Se as compared to horses consuming 0.06 ppm Se. Shellow et al. (1985) also reported significantly greater whole blood Se concentrations in horses consuming 0.16 and 0.26 ppm Se as compared to horses consuming 0.11 ppm Se. Data from this study also partially agree with Calamari et al. (2009), who reported at d 84, horses consuming 0.18 mg organic Se/kg DM, 0.29 mg organic Se/kg DM, 0.39 mg organic Se/kg DM, and 0.29 mg inorganic Se/kg DM had significantly greater whole blood Se concentrations as 73 74 compared to horses consuming 0.085 mg Se/kg DM. Furthermore, the researchers reported horses consuming 0.29 mg organic Se/kg DM and 0.39 mg organic Se/kg DM had significantly greater whole blood Se concentrations as compared to horses 0.29 mg inorganic Se/kg DM at d 84 (Calamari et al., 2009). Once again, overall mean whole blood Se concentrations at d 84 were much lower than expected, and led to an unplanned additional blood draw 12 days later on d 96. Upon analysis of the raw data at d 84, whole blood Se concentrations decreased in all horses regardless of treatment, indicating that the means at d 84 were not statistical outliers, but rather a “real” biological event. The biological explanation for the decrease at d 84 is unknown. A possible explanation is the extent of depletion in all horses, and a possible RBC lifecycle effecting Se incorporation, at d 84. Stowe (1998) reported the life span of equine RBC about 80 to 90 d. However, Carter et al. (1974) reported the lifespan of erythrocytes in light horses to be 145 to 165 d. Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 96 of Repletion A significant effect of treatment was observed on overall whole blood Se concentrations in horses consuming SE1 and SE3 at d 96 (Figure 11). Horses consuming SE3 (mean = 151.0 ng Se/mL) had greater (P = 0.001) whole blood Se concentrations as compared to horses consuming SE1 (mean = 132.0 ng Se/mL). There are no studies reporting the effects of Se supplementations on whole blood Se concentrations at d 96. As previously stated, the unexpectedly low values observed at d 84 led to an additional blood sampling at d 96. Overall mean whole blood Se 75 76 concentrations appeared to return to expected values based on the regression curve previously developed and data from previous published studies. Whole Blood Selenium Concentrations in Horses Consuming 0.1 and 0.3 ppm Selenium at d 112 of Repletion A significant effect of treatment was observed on overall whole blood Se concentrations in horses consuming SE1 and SE3 at d 112 (Figure 12). Horses consuming SE3 (mean = 149.6 ng Se/mL) had significantly greater (P < 0.001) whole blood Se concentrations as compared to horses consuming SE1 (mean = 128.2 ng Se/mL). Data for whole blood Se concentrations at d 112 of the current study agree with that of Janicki et al. (2001), who reported significantly greater serum Se concentrations at d 84 of supplementation in mares supplemented with 3 mg organic Se/d as compared to mares consuming 3 mg inorganic Se/d and 1 mg inorganic Se/d. Data from this study partially agrees with Calamari et al. (2009), who reported at d 112, horses consuming 0.18 mg organic Se/kg DM, 0.29 mg organic Se/kg DM, 0.39 mg organic Se/kg DM, and 0.29 mg inorganic Se/kg DM had significantly greater whole blood Se concentrations as compared to horses consuming 0.085 mg Se/kg DM. Furthermore, Calamari et al. (2009) reported horses consuming 0.29 mg organic Se/kg DM and 0.39 mg organic Se/kg DM had significantly greater whole blood Se concentrations as compared to horses 0.18 mg organic Se/kg DM, 0.29 mg inorganic Se/kg DM at d 84 . Although Brummer et al. (2013) did not analyze whole blood Se concentrations on d 112 as in the current study, the researchers reported significantly greater whole 77 78 blood Se concentrations at d 154 and 189 in horses consuming 0.3 mg organic and inorganic Se/kg DM as compared to horses consuming 0.12 mg Se/kg DM. Depletion Phase Erythrocyte Glutathione Peroxidase Activity Regression At initiation of the depletion phase (d 0), overall mean erythrocyte (RBC) GSHPx activity were 42.33 ± 7.07 mU/mg Hb. Overall RBC GSH-Px activity in horses consuming 23% of NRC Se recommendations depleted had large variation. Due to the large variation, a non-linear regression equation could not be developed. Individual RBC GSH-Px activities during the depletion phase can be observed in Figure B-1 in the Appendix. There are no published studies reporting the regression of a Se depletion phase. However, data for Se depletion in horses has been reported. Brummer et al. (2013) reported horses fed 60% of the NRC recommendation of Se for 196 d had significantly lower whole blood GSH-Px activity at d 140 and 196 as compared to horses fed 0.12 mg Se/kg DM. Furthermore, Brummer et al. (2013) reported significantly lower whole blood GSH-Px activity in horses fed 0.06 mg Se/kg DM at d 84, 140, 168, and 196 as compared to d 0. Whole blood GSH-Px activity concentrations in horses receiving 0.06 mg Se/kg DM was significantly lower at d 168 and 196 as compared to d 84 and 140. However, the authors reported no significant differences in whole blood GSH-Px activity between d 168 and 196. In the current study, overall mean RBC GSH-Px activity at the initiation of depletion were 42.33 mU/mg Hb, as compared to Brummer et al. (2013), who reported overall mean whole blood GSH-Px activity of 64.5 mU/mg Hb. In addition, at the end of the depletion phase (d 112) in the current study, overall mean whole blood GSH-Px 79 activity in horses consuming 23% of NRC recommendations were 27.38 mU/mg Hb. Brummer et al. (2013) reported whole blood GSH-Px activity in horses consuming 60% of NRC recommendations of 52.7 mU/mg Hb at d 140, 46.7 mU/mg Hb at d 168, and 43.1 mU/mg Hb at d 196. Repletion Phase Erythrocyte Glutathione Peroxidase Activity Regression At the initiation of repletion phase (d 112 of depletion), horses were stratified according to whole blood Se concentrations at d 84 of depletion, and assigned to 1 of 2 Se supplement treatments. Horses assigned to SE1 had overall mean RBC GSH-Px activity of 27.28 ± 6.45 mU/mg Hb. Horses assigned to SE3 had overall mean whole blood Se concentrations of 27.48 ± 4.12 mU/mg Hb. Non-linear regression equations could not be developed due to variation within sample. Individual RBC GSH-Px activities during repletion phase can be observed in Appendix Figure B-2. There are no studies reporting non-linear regression analysis of RBC GSH-Px activity during a repletion phase. However, Calamari et al. (2009) reported linear regressions of RBC GSH-Px activity {RBC GSH-Px activity, mU/L = 59.09 ± 9.70 x time (d) + 12059 ± 2585} in horses consuming 0.18 mg Se yeast/kg DM, {RBC GSH-Px activity, mU/L = 58.50 ± 7.87 x time (d) + 15149 ± 2555} in horses consuming 0.29 mg Se yeast/kg DM, {RBC GSH-Px activity, mU/L = 63.88 ± 8.01 x time (d) + 10407 ± 2827} in horses consuming 0.39 mg Se yeast/kg DM, and {RBC GSH-Px activity, mU/L = 90.74 ± 5.44 x time (d) + 3391 ± 1623} in horses consuming 0.29 mg Na selenite/kg DM. 80 Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 0 of Repletion There was no significant differences observed in overall mean RBC GSH-Px activity in horses assigned to SE1 (mean = 27.28 mU/mg Hb) and SE3 (mean = 27.48 mU/mg Hb) at d 0 of repletion (P = 0.477; Figure 13). Data for RBC GSH-Px activity at d 0 of repletion in the current study agree with that of Brummer et al. (2013), who reported no differences in whole blood GSH-Px activity in horses previously fed 0.06 mg Se/kg DM for 196 d at d 0 of repletion. The results also agree with Calamari et al. (2009), who reported no differences in whole blood GSH-Px activity in horses fed 0.085 mg Se/kg DM for 2 mo at the beginning of a repletion phase. Richardson et al. (2003) reported no significant differences at the beginning of a repletion phase in plasma and RBC GSH-Px activity in horses fed 0.15 mg Se/kg DM for 28 d. Richardson et al. (2006) reported no significant differences at the beginning of repletion in plasma, RBC, and muscle GSH-Px activity in horses fed 0.15 mg Se/kg DM for 28 d. Shellow et al. (1985) reported no significant differences in plasma GSH-Px activity in horses fed 0.06 ppm Se for a minimum of 4 wk. Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 14 of Repletion There was no significant effect of treatment observed on overall mean RBC GSHPx activity in horses consuming SE1 (mean = 30.38 mU/mg Hb) and SE3 (mean = 30.38 mU/mg Hb) at d 14 of repletion (P = 0.500; Figure 14). 81 82 83 Data for RBC GSH-Px activity at d 14 of repletion in the current study agree with that of Shellow et al. (1985), who reported no significant differences in plasma GSH-Px activity in horses consuming 0.06, 0.11, 0.16, and 0.26 ppm Se at d 14 of repletion. Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 28 of Repletion There was no significant effect of treatment observed on overall mean RBC GSHPx activity in horses consuming SE1 (mean = 28.81 mU/mg Hb) and SE3 (mean = 29.74 mU/mg Hb) at d 28 of repletion (P = 0.411; Figure 15). Data for RBC GSH-Px activity at d 28 of repletion in the current study agree with that of Brummer et al. (2013), who reported no differences in whole blood GSH-Px activity in horses consuming 0.06 and 0.3 mg Se/kg DM at d 28 of repletion. These results also agree with Calamari et al. (2009), who reported no differences in whole blood GSH-Px activity in horses consuming 0.085, 0.18, 0.29, and 0.39 mg Se/kg DM at d 28 of repletion. In addition, data from the current study agrees with that of Shellow et al. (1985), who reported no significant differences in plasma GSH-Px activity in horses fed 0.06, 0.11, 0.16, and 0.26 ppm Se at d 28 of repletion. The results of the current study partially agrees with Richardson et al. (2003), who reported horses consuming 0.6 mg organic Se/mg DM had significantly greater RBC GSH-Px activity at d 28 as compared to horses consuming 0.15 mg organic Se/kg DM, and 0.6 inorganic Se/kg DM. Further, Richardson et al. (2003) reported no significant differences at d 28 of repletion in plasma RBC GSH-Px activity in horses 0.15 mg organic Se/kg DM, 0.6 mg organic and inorganic Se/kg DM. Data from the current study also disagrees with that of Richardson 84 85 et al. (2006), who reported horses consuming 0.45 mg organic Se/kg DM tended to have greater RBC GSH-Px activity at d 28 of repletion as compared to horses consuming 0.15 mg organic Se/kg DM and 0.45 mg inorganic Se/kg DM. Richardson et al. (2006) reported no significant differences at d 28 of repletion in plasma GSH-Px activity in horses fed 0.15 mg Se/kg DM, and 0.45 mg organic or inorganic Se/kg DM. Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 56 of Repletion There was no significant effect of treatment observed on overall mean RBC GSHPx activity in horses consuming SE1 (mean = 25.99 mU/mg Hb) and SE3 (mean = 27.98 mU/mg Hb) at d 56 of repletion (P = 0.328; Figure 16). Data for RBC GSH-Px activity at d 56 of repletion in the current study agree with that of Brummer et al. (2013), who reported no differences in whole blood GSH-Px activity in horses consuming 0.06 and 0.3 mg Se/kg DM at d 56. The results from the current study agree with that of Shellow et al. (1985), who reported no significant differences in plasma GSH-Px activity in horses fed 0.06, 0.11, 0.16, and 0.26 ppm Se at d 56. The results of the current study disagree with that of Calamari et al. (2009), who reported significantly greater whole blood GSH-Px activity in horses consuming 0.18, 0.29, and 0.39 mg organic Se/kg DM and 0.39 mg inorganic Se/kg DM as compared horses consuming 0.085 mg Se/kg DM at d 56 of repletion. In addition, Calamari et al. (2009) reported horses consuming 0.39 mg inorganic Se/kg DM had significantly greater whole blood GSH-Px activity as compared to horses consuming 0.29 mg organic Se/kg DM. The results of the current study disagree with Richardson et al. (2003), who reported 86 87 horses consuming 0.6 inorganic Se/kg DM had significantly greater RBC GSH-Px activity at d 56 of repletion as compared to horses consuming 0.15 mg organic Se/kg DM, and tended (P = 0.057) to be greater as compared to horses consuming 0.6 mg organic Se/mg DM. Further, Richardson et al. (2003) reported no significant differences at d 56 of repletion in plasma RBC GSH-Px activity among horses consuming 0.15 mg organic Se/kg DM, 0.6 mg organic or inorganic Se/kg DM. Data from the current study also disagree with that of Richardson et al. (2006), who reported horses consuming 0.45 mg organic Se/kg DM tended to have greater RBC GSH-Px activity at d 56 of repletion as compared to horses consuming 0.15 mg organic Se/kg DM and 0.45 mg inorganic Se/kg DM. Richardson et al. (2006) reported no significant differences at d 56 of repletion in plasma and muscle GSH-Px activity in horses fed 0.15 mg Se/kg DM, and 0.45 mg organic or inorganic Se/kg DM. Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 84 of Repletion There was no significant effect of treatment observed on overall mean RBC GSHPx activity in horses consuming SE1 (mean = 31.88 mU/mg Hb) and SE3 (mean = 27.33 mU/mg Hb) at d 84 of repletion (P = 0.193; Figure 17). Data for RBC GSH-Px activity at d 84 of repletion in the current study agree with that of Shellow et al. (1985), who reported no significant differences in plasma GSH-Px activity in horses fed 0.06, 0.11, 0.16, and 0.26 ppm Se at d 84. The results of the current study disagree with that of Calamari et al. (2009), who reported significantly greater whole blood GSH-Px activity in horses consuming 0.18, 0.29, and 0.39 mg organic Se/kg 88 89 DM and 0.39 mg inorganic Se/kg DM as compared horses consuming 0.085 mg Se/kg DM at d 84 of repletion. In addition, Calamari et al. (2009) reported horses consuming 0.29, 0.39 mg organic Se/kg DM and 0.39 mg inorganic Se/kg DM had significantly greater whole blood GSH-Px activity as compared to horses consuming 0.18 mg organic Se/kg DM. Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 96 of Repletion There was no significant effect of treatment observed on overall mean RBC GSHPx activity in horses consuming SE1 (mean = 28.62 mU/mg Hb) and SE3 (mean = 27.37 mU/mg Hb) at d 96 of repletion (P = 0.371; Figure 18). There are no published studies reporting the effects of Se supplementations on RBC GSH-Px activity at d 96 of a repletion period. As previously stated, the unexpectedly low values of whole blood Se concentrations observed at d 84 led to an additional blood sampling at d 96. Overall mean RBC GSH-Px activity didn’t appear to be affected by the apparent decline in whole blood Se concentrations at d 84. Erythrocyte Glutathione Peroxidase Activity in Horses Consuming 0.1 and 0.3 ppm Selenium at d 112 of Repletion There was no significant effect of treatment observed on overall mean RBC GSHPx activity in horses consuming SE1 (mean = 36.15 mU/mg Hb) and SE3 (mean = 32.24 mU/mg Hb) at d 112 of repletion (P = 0.135; Figure 19). Data for RBC GSH-Px activity at d 112 of repletion in the current study disagree with that of Calamari et al. (2009), who reported significantly greater whole blood GSH- 90 91 92 Px activity in horses consuming 0.18, 0.29, and 0.39 mg organic Se/kg DM and 0.39 mg inorganic Se/kg DM as compared to horses consuming 0.085 mg Se/kg DM at d 112 of repletion. In addition, Calamari et al. (2009) reported horses consuming 0.29, 0.39 mg organic Se/kg DM and 0.39 mg inorganic Se/kg DM had significantly greater whole blood GSH-Px activity as compared to horses consuming 0.18 mg organic Se/kg DM. Although Brummer et al. (2013) did not analyze whole blood GSH-Px activity on d 112 as in the current study, the researchers reported significantly greater whole blood GSH-Px activity in horses consuming 0.12 mg inorganic Se/kg DM, or 0.3 mg organic and inorganic Se/kg DM as compared to horses consuming 0.085 organic Se/kg DM at d 154 of repletion. Additionally, the authors reported significantly greater whole blood GSH-Px activity in horses consuming 0.3 mg inorganic Se/kg DM as compared to horses consuming 0.12 mg inorganic Se/kg DM at d 154 of repletion. At d 189, Brummer et al. (2013) reported horses consuming 0.12 mg inorganic Se/kg DM, or 0.3 mg organic and inorganic Se/kg DM had significantly greater whole blood GSH-Px activity as compared to horses consuming 0.085 mg organic Se/kg DM. Furthermore, at d 189, the authors reported horses consuming 0.3 mg organic and inorganic Se/kg DM had significantly greater whole blood GSH-Px activity as compared to horses consuming 0.12 mg inorganic Se/kg DM. Possible Explanation for Differences in Erythrocyte Glutathione Peroxidase Activity between Studies One possible explanation for the different results observed in the current study and that of previous studies is the sample handling time and environmental temperature. 93 Koller et al. (1984) stated whole blood GSH-Px activity was less stable and reliable than was whole blood Se concentrations. Hussein and Jones (1981) measured whole blood GSH-Px activity in cattle, goats, and horses, and reported that both samples stored at room temperature (20 °C), or in a refrigerator (5 °C), had considerably reduced enzyme activity within 3 d, particularly in whole blood from horses. Jones (1985) reported that whole blood GSH-Px activity was reduced by approximately 20% unless samples were immediately frozen after blood draw. Abiaka et al. (2000) reported RBC GSH-Px activity was stable in samples stored at -80 °C for approximately 2 yr. Additionally, the authors reported that prior to freezing, plasma was separated and 0.9% NaCl solution was spun with RBC at 2500 x g for 5 min using a non-temperature controlled centrifuge. In the current study, the protocol for the assay kit (Bioxytech® GPx-340TM; OxisResearchTM , Portland, OR) recommended centrifuging samples at 4 °C. However, the centrifuge used in this study was not temperature controlled, therefore the possible change in temperature could have increased the oxidation of GSH-Px. The resultant differing oxidation rates could account for the differences in RBC GSH-Px activity observed in the current study with data observed in previously mentioned studies. 94 Chapter V CONCLUSIONS AND IMPLICATIONS Results from this experiment allowed for the development of a depletion curve for horses consuming 23% of the NRC Se recommendation for 112 d. The results also indicate that horses may benefit from organic Se supplementation at levels higher than those recommended by the NRC, during a 112-d repletion phase in previously depleted horses. Variation in RBC GSH-Px activity suggests the importance of proper handling and storage of GSH-Px samples to maintain the integrity of the blood samples. Whole blood Se concentration data indicate that horses fed 0.3 ppm organic Se supplementation will have higher whole blood Se concentrations over time as compared to horses receiving 0.1 ppm organic Se supplementation. The non-linear regression curve developed for horses consuming 23% of NRC Se recommendation for 112 d was: {predicted whole blood Se concentration = 184.95 * (1 * EXP (-0.005 * day))}. The non-linear regression curve for the change in whole blood Se concentrations in horses consuming SE1, previously depleted to 108 ng Se/mL whole blood was: {predicted change in whole blood Se concentration = 20.911 * (1 - EXP (- 0.062 * day))}. For horses consuming SE3, previously depleted to 109 ng Se/mL whole blood, the non-linear regression curve was: {predicted change in whole blood Se concentration = 38.249 * (1 - EXP (-0.070 * day))}. 95 Regression curves appeared to be greater in horses consuming SE3 as compared to SE1. Horses consuming SE3 had greater whole blood Se concentrations at d 14, 28, 56, 84, 96, and 112 as compared to horses consuming SE1. Due to variation in RBC GSH-Px activity, non-linear regression curves could not be developed for the depletion phase, and each treatment during the repletion phase. No significant differences were observed in RBC GSH-Px activity between treatments during repletion. During the depletion phase, whole blood Se concentrations in this study mostly agreed with that of Brummer et al. (2013). Whole blood Se concentrations during the repletion phase mostly agreed with previously reported repletion studies. Data for RBC GSH-Px activity from the current study both agreed and disagreed with previous studies. Sample handling and storage may have affected the results of the RBC GSH-Px activity assay. Data from the current study indicate that horses depleted to the extent of the current study never reach their original values, even after organic Se supplementation for 112-d. Further research may be necessary to determine the time and dietary concentration required to replenish Se stores in the body to adequate levels. In addition, further research needs to address the economic and possible environmental impact of Se supplementation in the horse industry. 96 LITERATURE CITED Abiaka, C., F. Al-Awadi, and S. Olusi. 2000. Effect of prolonged storage on the activities of superoxide dismutase, glutathione reductase, and glutathione peroxidase. Clinical Chem. 46: 566-567. Aleman, M. 2008. A review of equine muscle disorders. Neuromuscular Disorders. 18:277-287. Avellini, L., E. Chiaradia, and A. Gaiti. 1999. Effect of exercise training, selenium and vitamin E on some free radical scavengers in horses (Equus caballus) Compar. Biochem. Phys. 123:147-154. Blackmore, D.J., C. Campbell, C. Dant, J.E. Holden, and J.E. Kent. 1982. Selenium status of Thoroughbreds in the United Kingdom. Equine Vet. J. 14(2):139-143. Brummer, M., J.E. Ringler, A.G. Parks, S. Hayes, A.A. Adams, D.W. Horohov, and L.M. Lawrence. 2009. Selenium status and equine immune function. J. Equine Vet. Sci. 29:362-363. Brummer, M., S. Hayes, S.M. McCown, A.A. Adams, D.W. Horohov, and L.M. Lawrence. 2011a. Selenium depletion reduces vaccination response in horses. J. Equine Vet. Sci. 31:266. Brummer, M., S. Hayes, J.E. Earing, S.M. McCown, and L.M. Lawrence. 2011b. Effect of selenium depletion on oxidative stress in mature horses. J. Equine Vet. Sci. 31: 257. Brummer, M., S. Hayes, K.A. Dawson, and L.M. Lawrence. 2013. Measures of antioxidant status of the horse in response to selenium depletion and repletion. J. Anim. Sci. 91:2158-2168. Calamari, L., A. Ferrari, and G. Bertin. 2009. Effect of selenium source and dose on selenium status of mature horses. J. Anim. Sci. 87: 167-178. Calamari, L., F. Abeni, and G. Bertin. 2010. Metabolic and hematological profiles in mature horses supplemented with different selenium sources and doses. J. Anim. Sci. 88: 650-659. 97 Carmel, D.K., M.V. Crisman, W.B. Ley, M.H. Irby, and G.H. Edwards. 1989. A survey of whole blood selenium concentrations of horses in Maryland. Proc. 11th Eq. Nutri. Phys. Sym. Stillwater, OK. Carter, E.I., V.E. Valli, B.J. McSherry, F.J. Milne, G.A. Robinson, and J.H. Lumsden. 1974. The kinetics of hematopoiesis in the light horse I. the lifespan of peripheral blood cells in the normal horse. Can. J. Comp. Med. 38: 303-313. Charlton, S.J. and W.N. Ewing. 2007. The Minerals Directory. Context Products, Nottingham, UK. Chavez, E.R. 1979. Effect of dietary selenium depletion and repletion on plasma glutathione peroxidase activity and selenium concentration in blood and body tissue of growing pigs. Can. J. Anim. Sci. 59: 761-771. Chiaradia, E., L. Avellini, F. Rueca, A. Spaterna, F. Porciello, M.T. Antonioni, and A. Gaiti. 1998. Physical exercise, oxidative stress and muscle damage in racehorses. Compar. Biochem. Phys. 119:833-836. Daniels, L.A. 1996. Selenium metabolism and bioavailability. Bio. Trace. Element Res. 54: 185-199. Desta, B., G. Maldonado, H. Reid, B. Puschner, J. Maxwell, A. Agasan, L. Humphreys, and T. Holt. 2011. Acute selenium toxicosis in polo ponies. J. Vet. Diagn. Invest. 23(3): 623-628. Dunnett, C.E., and M. Dunnett. 2008. Organic selenium and the exercising horses. In: 4th European Workshop on Equine Nutri., Forssa, Finland. p. 33. (Nutrition of the exercising horse. Markku T. Saatamoinen. P. 255) Equine Nutrition. 2005. Selenium sources. J. Equine Vet. Sci. 25(1):35-36. EXRX. Selenium. 2013. Accessed April 19 2013. http://www.exrx.net.nutrition/ antioxidants/selenium.html Greiwe-Crandell, K.M., D.S. Kronfeld, G.A. Morrow, and W. Tiegs. 1993. Phosphorus and selenium depletion in thoroughbred mares and weanlings. Proc. 13th Equine Nutri. Phys. Sym. Gainesville, FL. Hussein, K.S., and B.E. Jones. 1981. Effects of different anticoagulants on determination of erythrocyte glutathione peroxidase. Acta Vet. Scand. 22: 472-479. 98 Janicki, K.M., L.M. Lawrence, T. Barnes, and C.J. Stine. 2001. The effect of dietary selenium source and level on selenium concentration, glutathione peroxidase activity, and influenza titers in broodmares and their foals. Proc. 17th Equine Nutri. Phys. Sym. Lexington, KY. Jones, D.G. 1985. Stability and storage characteristics of enzymes in cattle blood. Res. Vet. Sci. 38: 301-306. Karren, B.J., J.F. Thorson, C.A. Cavinder, C.J. Hammer, and J.A. Coverdale. 2010. Effect of selenium supplementation and plane of nutrition on mares and their foals: selenium concentrations and glutathione peroxidase. J. Anim. Sci. 88:991- 997. Knight, D.A. and W.J. Tyznik. 1990. The effect of dietary selenium on humoral immunocompetence of ponies. J. Anim. Sci. 68:1311-1317. Koller, L.D., P.J. South, J.H. Exon, G.A. Whitbeck, and J. Mass. 1984. Comparison of selenium levels and glutathione peroxidase activity in bovine whole blood. Can. J. Comp. Med. 48: 431-433. Koller, L.D., and J.H. Exon. 1986. The two faces of selenium- deficiency and toxicityare similar in animals and man. Can. J. Vet. Res. 50:297-306. Lescure, A., M. Rederstorff, A. Krol, P. Guicheney, and V. Allamand. 2009. Selenoprotein function and muscle disease. Biochimica et Biophysica Acta (BBA)-General Subjects, 1790(11): 1569-1574. Lewis, L.D. 1995. Feeding and care of the horse. 2nd ed. Williams and Wilkins. Baltimore, MD. Ludvikova, E., L. Pavlata, M. Vyskocil, and P. Jahn. 2005. Selenium status of horses in Czech Republic. ACTA Vet. BRNO. 74: 369-375. Mayer, G. 2009. Immunoglobulins- Structure and function. Microbiology and Immunology. http://pathmicro.med.sc.edu/mayer/igstruct2000.htm Montgomery, J.B., J.J. Wichtel, M.G. Wichtel, M.A. McNiven, J.T. McClure, F. Markham, and D.W. Horohov. 2012a. Effects of selenium source on measures of selenium status and immune function in horses. Canadian J. Vet. Res. 76: 281- 291. 99 Montgomery, J.B., J.J. Wichtel, M.G. Wichtel, M.A. McNiven, J.T. McClure, F. Markham, and D.W. Horohov. 2012b. The effects of selenium source on measure of selenium status of mares and selenium status and immune function of their foals. J. Equine Vet. Sci. 32:352-359. Nicholson, J.W., R.E. McQueen, and R.S. Bush. 1991a. Response of growing cattle to supplementation with organically bound or inorganic sources of selenium or yeast cultures. Can. J. Anim. Sci. 71: 803-811. Nicholson, J.W., J.G. Allen, and R.S. Bush. 1991b. Comparisons of responses in whole blood and plasma selenium levels during selenium depletion and repletion of growing cattle. Can. J. Anim. Sci. 71: 925-929. NRC. 1983. Selenium in Nutrition, Revised Ed. Natl. Acad. Press, Washington, D.C. NRC. 1989. Nutrient Requirements of Horses. 5th rev. ed. Natl. Acad. Press, Washington, DC. NRC. 2007. Nutrient requirements of horses. 6th rev. ed. Natl Acad. Press, Washington, DC. NIH. Selenium. 2013. Accessed August 14 2013. http://ods.od.nih.gov/factsheets/ Selenium-HealthProfessional/ Oldfield, J. E. 1998. Environmental implications of uses of selenium with animals. Environmetal Chemistry of Selenium, 129-142. Pagan, J.D., P. Karnezos, M.A.P. Kennedy, T. Currier, and K.E. Hoekstra. 2007. Effect of selenium source of selenium digestibility and retention in exercised thoroughbreds. Proc. 16th Equine Nutri. Phys. Soc., Raleigh, N.C. Podoll, K.L., J.B. Bernard, D.E. Ullrey, S.R. DeBar, P.K. Ku, and W.T. Magee. 1992. Dietary selenate versus selenite for cattle, sheep, and horses. J. Anim. Sci. 70: 1965-1970. Reddy, C.C. and E.J. Massaro. 1983. Biochemistry of selenium: a brief overview. Fund. App. Tox. 3: 431-436. Richardson, S.M., P.D. Siciliano, T.E. Engle, C.K. Larson, and T.L. Ward. 2006. Effect of selenium supplementation and source on the selenium status of horses. J. Anim. Sci. 84: 1742-1748. 100 Richardson, S.M., P.D. Siciliano, T.E. Engle, and T.L. Ward. 2003. Effect of selenium supplementation and source (organic vs. inorganic) on selenium status on horses. Proc. 18th Equine Nutri. Phys. Sym. East Lansing, MI. Shelle, J.E., W.D. VanHuss, J.S. Rook, and D.E. Ullrey. 1985. Relationship between selenium and vitamin E nutrition and exercise in horses. 9th Proc. Eq. Nutri. Phys. Sym. East Lansing, MI. Shellow, J.S., S.G. Jackson, J.P. Baker, and A.H. Cantor. 1985. The influence of dietary selenium levels on blood levels of selenium and glutathione peroxidase activity in the horse. J. Anim. Sci. 61: 590-594. Sors, T.G., D.R. Ellis, and D.E. Salt. 2005. Selenium uptake, translocation, assimilation and metabolic fate in plants. Photosynthesis Res. 86: 373-389. Stadtman, T.C. 1983. New biologic functions- Selenium-dependent nucleic acids and proteins. Fund. App. Tox. 3: 420-423. Stowe, H.D. 1967. Serum selenium and related parameters of naturally and experimentally fed horses. J. Nutri. 93: 60-64. Stowe, H.D. and T.H. Herdt. 1992. Clinical assessment of selenium status of livestock. J. Anim. Sci. 70:3928-3933. Stowe, H.D. 1998. Selenium supplementation for horse feed. Nottingham Univ. Press, Nottingham, UK. Ullrey, D.E. 1987. Biochemical and physiological indicators of selenium status in animals. J. Anim. Sci. 65: 1712-1726. Ullrey, D.E. 1992. Basis for regulation of selenium supplements in animal diets. J. Anim. Sci. 70:3922-3927. White, S. 2010. Dietary selenium and prolonged exercise alter gene expression and activity of antioxidant enzymes in equine skeletal muscle. MS Thesis. Univ. Florida, Gainesville. White, S., L.K. Warren, and J. Bobel. 2011. Effects of dietary selenium supplementation and exercise on antioxidant enzyme activity in equine blood and skeletal muscle. J. Equine Vet. Sci. 31:266-267. Wright, P.L. and M.C. Bell. 1966. Comparative metabolism of selenium and tellurium in sheep and swine. Am. J. Physiol. 211:6-10. 101 APPENDIX FIGURES A WHOLE BLOOD SELENIUM CONCENTRATIONS GRAPHS 102 103 104 105 APPENDIX FIGURES B ERYTHROCYTE GLUTATHIONE PEROXDIASE ACITIVITY GRAPHS 106 107 108 APPENDIX TABLES A 109 Table A-1. Individual Whole Blood Selenium Concentrations (ng/mL) Depletion phase (d) Horse 0 28 56 84 112 1 187 158 139 127 113 2 175 138 120 104 89 3 191 153 133 122 106 4 193 171 145 139 122 5 199 172 140 128 111 6 194 161 150 148 122 7 188 152 129 125 103 8 187 161 137 132 114 9 181 162 144 137 116 10 179 156 126 100 94 Repletion phase (d) Horse Supplement (ppm) 14 28 56 84 96 112 1 0.1 117 134 143 126 138 135 2 0.1 100 114 121 111 125 119 3 0.1 115 130 134 118 129 126 4 0.1 131 141 142 128 138 137 5 0.1 115 128 135 117 130 124 6 0.3 139 158 162 145 165 161 7 0.3 123 144 148 132 146 145 8 0.3 137 150 159 135 150 147 9 0.3 134 146 155 139 149 152 10 0.3 116 137 142 131 145 143 110 Table A-2. Individual Erythrocyte Glutathione Peroxidase Activity (mU/mg Hb) Depletion phase (d) Horse 0 28 56 84 112 1 34.82 42.35 34.65 48.32 24.28 2 48.72 34.47 50.75 34.76 22.65 3 39.86 38.87 42.87 40.84 29.14 4 47.80 40.67 56.14 46.93 37.77 5 34.59 33.31 47.39 32.44 22.54 6 40.56 33.37 43.34 46.41 31.63 7 40.32 43.92 46.29 40.32 23.29 8 57.41 45.60 41.02 26.88 24.97 9 37.77 45.65 44.32 37.83 32.21 10 41.48 46.29 35.69 26.19 25.32 Repletion phase (d) Horse Supplement (ppm) 14 28 56 84 96 112 1 0.1 32.04 28.10 20.28 22.25 22.48 33.54 2 0.1 21.03 30.13 23.06 40.55 30.24 36.62 3 0.1 24.68 24.04 25.09 26.01 25.20 44.96 4 0.1 39.34 35.28 38.82 38.12 37.89 35.28 5 0.1 34.82 26.48 22.71 32.44 27.29 30.36 6 0.3 26.65 28.16 32.91 31.69 21.55 26.24 7 0.3 33.89 34.94 31.23 30.13 37.02 32.85 8 0.3 33.89 33.66 32.79 21.49 25.78 38.18 9 0.3 34.30 35.22 24.10 36.33 26.30 28.21 10 0.3 23.17 16.74 18.89 17.03 26.19 35.69 111 Table A-3. Individual Body Weights (kg) Depletion Horse 0 14 28 56 84 112 1 570 550.7 540.7 538 552 547 2 494.2 478.7 475.7 466 481 491 3 509.8 487.6 486.8 482 492 493 4 484.2 441.4 433.7 433 443 445 5 601.2 522.3 507.4 507 517 519 6 506.5 491.3 495.3 488 486 490 7 565.4 539 540.7 540 549 540 8 604.4 549.6 546.6 540 539 536 9 569.5 559.1 540.1 528 544 544 10 537.8 529 519.4 514 538 534 Repletion Horse Supplement (ppm) 14 28 56 84 112 1 0.1 549 543 541 537 547.5 2 0.1 486 481 491 481 491 3 0.1 492 488 487 481 488.3 4 0.1 438 435 435 434 435.3 5 0.1 519 514 527 526 537 6 0.3 490 486 474 478 478.6 7 0.3 540 527 529 528 529.2 8 0.3 527 527 519 521 523.7 9 0.3 533 529 533 532 534 10 0.3 529 525 527 519 532.1
