EFFECTS OF AGE AND DIET ON GLUCOSE AND INSULIN DYNAMICS IN THE HORSE By Julie Lynn Rapson A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Large Animal Clinical Sciences Master of Science 2015 ABSTRACT EFFECTS OF AGE AND DIET ON GLUCOSE AND INSULIN DYNAMICS IN THE HORSE By Julie Lynn Rapson Insulin resistance (IR) is a key metabolic disturbance in horses that develop obesity - associated laminitis. In addition to obesity, age and diet affect tissue sensitivity to insulin but these factors have received limited investigation in horses. This stud y tested the hypothesis that glucose and insulin responses to a sweet feed (SF, high CHO) meal would be greater in aged horses, as compared to adult horses, as well as in horses adapted to a forage - only diet. Three diets, grass hay (G), grass hay plus swee t feed (starch and sugar - rich, SS), and grass hay plus a fat and fiber feed (FF), were fed to 17 healthy mares, 8 adult (5 - 12 yr) and 9 aged (>19 yr), for a 6 - week adaptation period in a randomized design. Minimal model parameters during a frequently sampl ed intravenous glucose tolerance test (FSIGTT) and glucose and insulin responses to a standardized SF meal were determined after 30 and 42 days on each diet, respectively. Data were analyzed by repeated measures ANOVA. The acute insulin response to glucose (AIRg) was greater and tissue insulin sensitivity was lower in aged horses, regardless of diet. No differences in glucose responses to the SF meal were detected between age groups for any of the diets; however, both peak glucose concentration and AUC - G we re lower after adaptation to the SS diet. In contrast, peak insulin concentration and AUC - I were greater in aged horses than adult horses on all diets but no differences were found between diets within age groups. As hypothesized, the insulin response, bu t not the glycemic response, to a sweet feed meal was greater in aged horses, regardless of background diet. Further, the glycemic response was greatest after adaptation to a forage - only diet in aged horses only. iii ACKNOWLEDGMENTS I would like to thank my committee members Dr. Schott, Dr. Geor, and Dr. McCutcheon for their continued guidance throughout this process. I would also like to thank my family and friends for their support. Lastly, I would like to recognize all of the horses that were a part of th is project. iv TABLE OF CONTENTS LIST OF TABLES v LIST OF FIGURES vi INTRODUCTION 1 CHAPTER 1 LITERATURE REVIEW 2 Overview 2 Insulin 3 Insulin Resistance, Diabetes Mellitus, and Metabolic Syndrome 5 Measurement of Insulin Sensitivity 8 Non - specific Measurements of Insulin Sensitivity 9 Specific Measurements of Insulin Sensitivity 13 Factors Affecting Insulin Sensitivity 16 CHAPTER 2 EFFECTS OF AGE AND DIET ON GLUCOSE AND INSULIN DYNAMICS IN THE HORSE 28 Introduction 28 Materials and Methods 29 Horses and housing 29 Study design 30 Insulin - modified frequently sampled intravenous glucose tolerance test (FSIGTT) 31 Standardized meal challenge (SMC) 31 Sample analysis 32 Calculations and statistical analysis 32 Results 33 Minimal model parameters 34 Insulin and glucose dynamics during the SMC 34 Discussion 35 C o n c l u s i o n s 3 9 CHAPTER 3 CONCLUSIONS 41 APPENDIX 47 REFERENCES 55 v vi LIST OF TABLES Table 1 Nutrient composition of the three diets fed, expressed on a dry matter basis, determined by proximate analysis of well - mixed composite samples of several aliquots of each feedstuff collected multiple times during the study; digestible energy (DE) of each diet was calculated using measured DE for each food stuff and the amount fed: HAY = grass hay; FF = mix of hay and fat and fiber rich co ncentrate feed; and SS = mix of hay and cereal grain feed rich in nonstructural CHO (sugar and starch feed). 48 Table 2 Bodyweights (BW) and body condition scores (BCS) at barn entry (day 22) and at the end (day 41) of each feeding period in eight a dult horses and nine aged horses. Values reported as means ± SEM. 49 Table 3 Acute insulin response to glucose (AIRg) infusion, insulin sensitivity (SI), glucose effectiveness (Sg), and disposition index (DI) determined by minimal model analysis of glucose and insulin data obtained from an insulin - modified frequently sampled i.v. glucose tolerance test performed in eight adult horses and nine aged horses that had been adapted to three diets for 4 wk: HAY (grass hay only); SS (hay plus a cereal gra in feed rich in hydrolyzable CHO [sugar and starch feed]); and FF (hay plus a fat and fiber rich concentrate feed). Values reported as means ± SEM. 50 Table 4 Measures of glucose and insulin responses before, during, and after the standardized CH O - rich meal challenge (4.4 g/kg of the SS cereal grain feed) administered in eight adult horses and nine aged horses that were adapted to three diets for 6 wk: HAY (grass hay only); SS (hay plus a cereal grain feed rich in nonstructural CHO [sugar and star ch feed]); and FF (hay plus a fat and fiber ri ch concentrate feed). Values reported as mean ± SEM. 51 vii LIST OF FIGURES Figure 1 Mean ± SEM glucose concentrations in response to the standardized meal challenge (0.4 g/kg SS concentrate feed offered for 60 min) in adult (bottom panel) and aged (upper panel) mares. The filled bars above the x - axes indicate time points that were different ( P <0.05) from baseline; asterisks indicate time points at which SS < HAY ( P < 0.05); cross indicates a time point at which SS < FF ( P < 0.05). 52 Figure 2 Mean ± SEM insulin concentrations in response to the standardized meal challenge (0.4 g/kg SS concentrate feed offered for 60 min) in adult (bottom panel) and aged (upper panel) mares. The filled bars above the x - axes indicate time points that were different ( P <0.05) from baseline; asterisk indicates a time point at which SS > HAY ( P < 0.05). 53 Figure 3 - 1 3 , black - 1 3 , gray fill) in mares that consumed >90% (full) of the SS meal in the standardized meal challenge (n = 36 studies) as compared to mares that left more the 20% (partial) of the SS meal after 60 min (* denotes P < 0.05). 54 1 INTRODUCTION This thesis contains three chapters. The first chapter is a literature review that describes reduced tissue sensitivity to insu lin called insulin resistance. Insulin plays and important role in carbohydrate and lipid metabolism. A multitude of different factors that can affect insulin resistance are discussed in humans as well as horses. Insulin resistance is evaluated by both non - specific and specific methods with the two standard techniqu es used in a research setting being the euglycemic - hyperinsulinemic clamp and minimal model analysis of insulin and glucose dynamics during a frequently sampled intravenous glucose tolerance test . There are limitation s to both of these methods which have lead to the development of less invasive techniques for measuring insulin sensitivity such as the meal response. The second chapter describes the experimental proto col and r esults of the minimal model analysis using frequently sampled insulin glucose tol erance test and a meal response challenge in both adult and aged horses. In the third chapter, conclusions are made based on prev ious research. Additionally, possible underlying mechanisms for age and diet variations in tissue sensitivity to insulin in t he horse are discussed. Lastly, practical implications from introducing a high sugar and starch concentrate to horses with metabolic disease are considered. 2 CHAPTER 1 LITERATURE REVIEW Overview Insulin resistance is defined as a decreased response of peripheral tissues to a normal concentration of insulin . As a result, a greater amount of insulin is required to attain a biological effect. Insulin resistance is a component of many metabolic cond itions such as type II diabetes and is associated with other health problems such as obesity in both humans and horses. Specific to the horse, hyperinsulinemia and insulin resistance predispose horses to laminitis which is a painful, life threatening dise a se of the hooves [19 ] . The term Metabolic Syndrome has been used to describe a combination of different medical disorders that increase the risk of developing diseases such as diabetes and hypertension in humans. Recently, the term Equine Metabolic Synd rome (EMS) has been coined to describe a clinical syndrome of obesity, insulin resistance, and predisposition to l aminitis in horses [35 ] . There are many factors that can influence insulin sensitivity in both humans and horses. Two of these factors are age and diet. In both humans and horses, insulin sensitivity decreases with age leading to an increased prevalence of i nsulin resistance [80 ] . The appropriate diet for animals and humans that are insulin resistant is a subject of extensive research today . Both carbohydrate restriction and increased protein consumption increase insulin secretion in humans [1 13 ] . In equids, there is a great deal of debate as to what diet is most appropriate for an aged horse with decreased insulin sensitivity, as well as for an animal diagnosed with EMS. There are low starch diets and senior feeds marketed for the aged equid but limited research has been conducted to determine whether these are the most appropriate diet for the insulin resistant or 3 EMS horse. Therefore, the scope of this study was to look at the effects of age and diet on glucose and insulin responses of a cohort of healthy, non - obese mares. I nsulin Insulin is the crucial hormone involved in regulation of carbohydrate and fat metabolism. Storage of nutrients is the major function of insulin. Insulin is a secretory protein consisting of two peptide chains: an A chain consisting of 21 amino acids and a B chain with 30 amino acids. These two chains are joined by two disulfide bridges giving mature insulin a total of 51 amino acids. The half - life of endogenous insulin in circulation is 10 minutes [13 ] . Insulin is produced in the endocrine portion of the pancreas, the islets of Langerhans. cells, and F cells. Beta cells are located within the central core of the i slets and produce insulin, proinsulin, amylin, and C - peptide. Insulin is encoded by a gene located on the short arm of chromosome 11. Transcription of this gene produces mRNA that encodes preproinsulin. Shortly after it is synthesized, the signal sequen ce is cleaved from the N - terminus of preproinsulin resulting in proinsulin. Proinsulin consists of three domains: A, B, and C. Proinsulin is packaged into clathrin - coated secretory granules within the Golgi apparatus. While in the Golgi apparatus, protea ses cleave proinsulin at two sites resulting in a 31 - amino acid connecting or C - peptide and insulin, consisting of chains A and B linked by disulfide bonds. Maturation of the secretory granule is completed with the loss of the clathrin coating. A mature secretory vesicle released into the portal circulation contains insulin and C - peptide in equal quantities in addition to a smaller am ount of proinsulin [9 ] . Release of insulin is biphasic with the acute phase representing release of preformed insulin in s ecretory granules and the later phase associated with de novo synthes is of insulin [13 ] . 4 Once released into the portal blood, insulin travels to the liver where over half is bound, exerts its action, and is removed immediately from circulation. The rem aining insulin is available to act on other target tissues within the body. Once at a target tissue, insulin binds to specific receptors on the plasma membrane. The insulin receptor is a heterotetramer consisting require d for activation of downstream signaling pathways [9 ] . protein s as well as activate additional substrates, such as IRS - 1, IRS - 2, and Src homology C terminus (SHC), by phosphorylation. This causes activation of additional kinases, phosphatases, and other signaling molecules leading to a complex pathway divided into t wo major signalin g pathways [9 ] . The first signaling pathway is the metabolic pathway. Activation of phosphatidylinositol - 3 - kinase resulting from binding to a phosphorylated IRS causes movement of vesicles containing glucose transporter 4 (GLUT - 4) to the cell membrane. These events allow increased cellular glucose uptake down a concentration gradient and prevent a substantial postprandial rise in circulating glucose concentration. The second pathway is the mitogenic pathway that regulates growth effects of insulin. Binding of SHC proteins to either the insulin receptor or IRS protein leads to activation of SHC. This ultimately results in increas ed gene expression [9 ] . Insulin is a key regulator of carbohydrate metabolism. In the postprandial period , glucose is absorbed into portal circulation from the small intestine. The elevation in glucose in 5 portal blood triggers pancreatic release of insulin. Subsequently, insulin increases entry of glucose into muscle, liver, and adipose tissue via GLUT - 4 tr ansporters. In muscle, protein synthesis is also increased in the presence of insulin by increasing amino acid uptake and ribosomal protein synthesis. In the liver, insulin also activates glycogen synthesis and storage and inhibits breakdown of glycogen post - prandially by decreasing activity of glucose - 6 - phosphatase. Appropriate insulin action maintains the plasma concentration of glucose within a narrow range in order to prevent disease. Further, following a meal, insulin release also provides a s ign al for satiety [9,13 ] . Insulin also regulates lipid metabolism. Triglycerides are energy dense storage molecules and their production and storage in adipose tissue is increased in response to insulin. Additionally, insulin inhibits hormone - sensitive lipase that hydrolyzes triglycerides to fatty acids, therefore, limiting breakdown of fat. When insulin production is decreased, for example during a fasting state, hormone - sensitive lipase is strongly activated promoting hydrolysis of triglycerides and r elease of glycerol and non - esterified fatty acids (NEFAs) into circulation for gluconeogenesis and - oxidation to provide energy [13 ] . Insulin Resistance, Diabetes Mellitus, and Metabolic Syndrome Insulin resistance is a metabolic disturbance defined as a decrease in the uptake of glucose by the liver, skeletal muscle, and adipose tissue as a result of reduced tissue sensitivity to insulin. Insulin resistance involves a failure of in sulin signaling [62 ] . This failure can be due decreased numbers of ins ulin receptors on cell surfaces, decreased activation (downregulation) of insulin receptors following insulin binding, or altered triggering of downstream intracellular pathways that lead to translocation of GLUT - 4 transporters t o the plasma membrane [114 ] . Insulin resistance has been described as either compensated or uncompensated. Compensated 6 insulin resistance is characterized by greater postprandial insulin secretion by the pancreas to appropriately control blood glucose concentration. Maintenance o f normoglycemia is accompanied by intermittent or persistent hyperinsulinemia. If pancreatic beta cells eventually amounts of insulin, intermittent or persistent hypergly cemia may develop. This condition is termed uncompensated insulin resistance and can be accompanied by either high or low circulating insulin co ncentrations [74 ] . Insulin resistance is a component of many metabolic disorders and is a risk factor for a n umber of medical problems in humans and domestic animals. In humans, these include obesity, dyslipidemias, type 2 diabetes mellitus, microalbuminuria, and cardiovascular problems including hypertension (collectively termed Metabol ic Syndrome [MS]) [1] . I n the equine population, insulin resistance is also recognized with obesity or regional fat deposition, hyperlipidemia, hypertension (less well documented than in humans), and diabetes mellitus (much rarer in equids than humans). Due to similarities to the human condition, the term Equine Metabolic Syndrome (EMS) was coined to describe this cluster of medical pr oblems in equids [56 ] . However, the primary clinical problem affecting equids with EMS is development of insidious onset laminitis, often accompanying ingestion of excessive amounts of soluble CHOs, notably lush pasture [39 ] . In humans, diabetes mellitus (DM) is the most common metabolic disease characterized by persistent hyperglycemia resulting from reduced insulin secretion and/or insul in resistance. Type 1 DM most commonly develops in children as a consequence of immune - mediated pancreatic beta cell destruction resulting in decreased insuli n production [74 ] . In addition to a lack of insulin causing decreased glucose uptake by tissues, the liver produces additional glucose 7 and ketones. Both hyperglycemia and increased ketones produce an osmotic diuresis. Also, metabolic acidosis can develop as a consequence of increased production of ketones. Treatment for type 1 DM requires administr ation of exogenous insulin to control hyperglycemia. If a patient is not treated with insulin, they may die of diabetic ketoacidosis [9,13 ] . In the horse, this ty pe of DM is rare [56 ] . - standin g insulin resistance leads to the development of type 2 DM [62] . Individuals with type 2 DM can produce some insulin but pancreatic output to an increase in plasma glucose is inadequate to maintain normglycemia. Thus, both insulin secretion and glucose h omeostasis are abnormal in affected patients. The primary treatment for patients with this type of DM is diet and exercise in order to better control blood glucose levels. It is recommended that patients eat well - balanced meals on a regular schedule to a void extremes in fluctuation of blood glucose concentration. Increasing the fiber content in the diet can also attenuate fluctuations in blood glucose by slowing the rate of CHO absorption from the intestine. If diet and exercise alone do not adequately control blood glucose concentration, medications to lower blood glucose concentration and/or insulin therapy may be prescribed [13 ] . In the human population, type 2 DM is generally diagnosed in adulthood and is seen more frequently in obese individuals [9 ] . Although insulin resistance is common in obese equids, progression to type 2 DM is a rare occurrence. In fact, type 2 DM is seen most frequently in older equids with insulin resistance as a component of pituitary pars intermedia dysfunction [23 ] . Metab olic syndrome is a clustering of medical disorders that increase the risk of type 2 DM and cardiovascular disease. Other names for MS include insulin resistance syndrome and syndrome X [56 ] . There a multiple criteria required to establish a diagnosing of metabolic 8 include presence of impaired glucose tolerance/insulin resistance or overt type 2 DM, along with two of the following abnormalities: hypertension, dyslipidemia, central obesit y, or microalbuminuri a [1] . Development of this complex disease syndrome is not completely understood but risk factors include diet, obesity, genetic predisposition, aging, stress, and a s edentary lifestyle [35 ] . As previously mentioned, EMS is the term commonly used to describe a similar cluster of metabolic dis orders in equids [56 ] . Although pathophysiology may not be identical in both species, overfeeding of CHO, genetic predisposition, and a sedentary lifestyle also appear to be important risk factors for development of EMS in equids. However, development of laminitis is a clear clinical difference between patients with MS and EMS and the pathophysiologic processes that culminate in this painful and debilitating lameness remain incompletely understood. Measurement of Insulin Sensitivity Because insulin resistance is a key metabolic disorder in both MS and EMS, measuring tissue sensitivity to insulin has been a strong focus of both research into the syndromes and managing affected patients. Tests to me asure insulin sensitivity can be divided into non - specific (indirect) and specific (direct) assessment of insulin resistance. Measurement of basal circulating glucose and insulin concentration is the simplest non - specific test. Next, various mathematical manipulations of these two measurements, termed proxy measurements, have been performed in an attempt to improve diagnostic sensitivity and specificity of these measurements. Measurement of changes in glucose and insulin concentrations in response to enteral or intravenous glucose (or glucose and insulin) administration is additional tests that provide dynamic assessment of pancreatic insulin production and tissue glucose uptake but they do not directly assess tissue insulin sensitivity. There are three specific or direct, quantitative tests that 9 have been used to measure tissue insulin sensitivity: the euglycemic - hyperinsulinemic clamp, the frequently sampled intravenous glucose tolerance test (FSIVGT) with mi nimal model analysis, and the insulin s uppression test [62 ] . Non - specific measurement of insulin sensitivity Measurement of basal (fasting) glucose concentration is a common screening test fo r DM in human medicine [62 ] . Fasting hyperglycemia indicates th at there is either a decrease in tissue glucose uptake due to decreased insulin sensitivity or that pancreatic beta cells are secreting inadequate insulin to maintai n normoglycemia [31 ] . Thus, this measurement does not determine a definitive cause for hyp erglycemia. Further, glucose concentrations within the blood can vary substantially over a short period of time due to environmental factors such as stress, feeding, a nd diurnal variation [113 ] . Despite limited use to assess tissue insulin sensitivity, me asuring glucose concentration remains a simple and cost - effective test in human medicine where DM is more commonly recognized than in equids. Measurement of basal (fasting) insulin concentration is another simple test and values are inversely correlated with ins ulin sensitivity [106 ] . A fasting insulin concentration of > 20 mU/L in equids is defined by the American College of Veterinary Internal Medicine Consensus Statement as supportive of insulin resistance [35 ] . Hyperinsulinemia develops when more ins ulin is secreted due to decreased tissue sensitivity to insulin (compensated insulin resistance). Hyperinsulinemia may be persistent or may only be detected for a few hours post - prandially ; therefore, false - negative (normal) results may be found in fastin g samples in equids with insulin resistance. In theory, persistent hyperinsulinemia would seem to support more severe insulin resistance than intermittent postprandial hyperinsulinemia but this has not been proven in equids 10 and, similar to glucose, insulin concentration can change with stress and other external factors [62,113 ] . Various proxies, or mathematical manipulations of glucose and insulin concentrations measured in the same fasting blood sample, have been developed in an attempt to better assess pancreatic function and insulin sensitivity. In human medicine, proxies have been validated in large patient populations and can more useful in identifying insulin resistance that glucose or insulin conc entrations alone [112 ] . One example is the glucose - to - insulin ratio that correlates positively with insulin sensitivity. Additionally, the insulin - to - glucose ratio is positively correlated with in sulin secretion [ 62 ] . The insulin - to - glucose ratio has been found to decrease within 15 minutes after intravenous injection of glucose in ponies and horses with pituitary pars intermedia dysfunction ( PPID) when compared to normal h orses [37 ] . Also, a modified insulin - to - glucose ratio (MIRG) and the reciprocal of the square root of insulin (RISKI) values have been reported to be useful to identify ponies with pre - laminitic met abolic syndrome [11 3 ] . Although simple to calculate, validation of these proxies in equine medicine remains limited as they have only been evaluated in small groups of equids. The g lucose tolerance test is another non - specific measure of insulin resistance. This test involves serial measurement of glucose concentration following enteral or intravenous administration of glucose. The time required for plasma glucose concentration to return to ) glucose [31 ] . Glucose intolerance, indicated by either a greater increase or longer duration of hyperglycemia as compared to normal subjects, can be a con sequence of inadequate pancreatic production or secretion of insulin or impaired ability of tissue glucose uptake. Thus, these tests only provide indirect support of tissue insu lin sensitivity [62 ] . 11 The oral glucose tolerance test was introduced in equin e medicine in 1973 to diagnose small intestinal disease and to assess p ancreatic endocrine function [95 ] . Following an overnight fast, horses are administered glucose (typically 1 g/kg) via a nasogastric tube or oral dosing syringe. A baseline blood samp le is collected prior to glucose administration followed by blood collection every 30 - 60 minutes for 6 hours following glucose administration. Blood samples are then analyzed for glucose and insulin concentration. Glucose concentration typically peaks fr om 90 - 120 minutes and should return to the baseline valu e after 4 hours [31 ] . When compared to an intravenous glucose tolerance test, variation in gastric emptying and intestinal absorption makes interpretation of oral glucose tolerance test results diffi cult for assessment of tissue insulin sensitivity, although an excessive rise in insulin concentration provides indirect support [91 ] . Also, stress accompanying nasogastric tube placement can cause transient increases in blood glucose and insulin concentr ations [31 ] . Although it can provide support for insulin resistance , the oral glucose tolerance test has more commonly been used as a diagnostic aid to document small intestinal malabsorption syndromes in horses [68 ] . Recently, however, there has been r and useful screening tool to document insulin resistance in equids. In this test Karo Syrup Lite TM is administered orally (15 mL/100 kg = 5 g sucrose) and blood samples are collected prior to dosing and 75 min after administration. An exaggerated insulin response ( > 60 mU/L at 75 minutes) is supportive of tissue in sulin resistance [99 ] . The intravenous glucose tolerance test eliminates gastrointestinal factors that must be considered with the oral glucose tolerance test [62 ] . The intravenous glucose tolerance test requires an intravenous catheter to be placed in the jugular vein a few hours prior to th e test. The animal should be fasted for at least 12 hours prior to receiving a rapid bolus of glucose (0.1 - 0.3 12 g/kg) intravenously. Blood glucose and insulin concentrations are determined from blood samples collected from time zero to 4 hours post - g lucos e infusion [31 ] . Peak glucose concentration is typically measured in the initial post - administration sample (15 minutes) with a decline to the baseline value within 1 hour. Insulin concentration peaks at 30 minutes and return to baseline following the pa ttern of the glucose response curve. If the blood glucose peak is either higher at 15 min, when compared to normal horses, or does not return to baseline values within 1 hour, test results support decreased tissue s ensitivity to insulin [37,91 ] . However, to better document insulin resistance, insulin concentrations should also be measured to determine whether the pancreas in not producing enough insulin (low insulin concentrations) or the peripheral tissues are not responding to insulin (high insulin conc entrations) [31 ] . The insulin tolerance test is another non - specific assessment that has been used to document insulin resistance. This test measures the blood glucose response to an intravenous dose of insulin. In a normal animal, blood glucose concentr ation falls to approximately 50% of the baseline value within 30 minutes of insulin injection. Blood glucose concentration should return to normal within 1.5 - 2 hours. If the animal is insulin resistant, the decrease in blood glucose concentration followi ng insulin administration will be blunted [31 ] . The combined glucose - insulin tolerance test (CGIT) was subsequently developed to assess the effects of simultaneous administration of glucose (0.15 g/kg) and insulin (0.1 U/kg), after an overnight fast. The C GIT essentially couples the dynamic responses observed in both the glucose tolerance and insulin suppression tests. Blood glucose is measured serially over the next 3 hours and results typically show an initial period of hyperglycemia followed by a period of hypoglycemia. Variation in the magnitude of these responses (excessive hyperglycemia or blunting of hypoglycemia) is supportive of reduced tissue sens itivity to insulin [27 ] . 13 Specific measurement of insulin sensitivity Three specific, quantitative methods have been used to determine tissue sensitivity to insulin. The first method is the insulin suppression test that measures the effect that insulin has on plasma glucose concentration when insulin and glucose are infused at a fixed rate. First, an inhibitory hormone, typically somatostatin is administered intravenously at a constant rate throughout the test to inhibit release of endogenous insulin from the pancreas and to suppress hepatic glucose release. Next, insulin and glucose are also infused intravenously at a constant rate. After an equilibration period, the infusion rate is manipulated during the last 60 minutes of the 150 minute test to produce steady - state glucose and insulin concentration. Mean glucose infusion rate during this time per iod is a measure of insulin - mediated glucose disposal rate that directly correlates with tissue insu lin sensitivity [62 ] . The purpose of this test is to limit the effects of endogenous insulin and hepatic glucose production in order to specifically measur e the effect of exogenous insulin on ti ssue glucose uptake [100 ] . Difficulties with the insulin suppression test, including adverse reactions to somatostatin, challenges in maintaining steady - state glucose and insulin concentrations, and development of hy perglycemia and glucosuria, have limited both clinical and research u se of this test [62 ] . The second quantitative method is the euglycemic - hyperinsulinemic clamp. During this test hyperinsulinemia is induced by an initial bolus followed by a constant ra te intravenous infusion of insulin and euglycemia is maintained by adjusting the intravenous infusion rate of glucose through a catheter in the contralater al jugular vein [30 ] . Insulin infusion suppresses endogenous glucose production and the goal of the test is to maintain a physiologic glucose concentration, typically 5 mmol/L, in the face of hyperinsulinemia. In order to maintain this glucose concentration, frequent blood samples are collected and immediately analyzed in order 14 to adjust the glucose inf usion rate appropriately. After an initial equilibration period of 60 - 120 minutes, mean glucose infusion rate during the last 60 minutes of the test is a measure of glucos e disposal rate . A low glucose disposal rate indicates reduced tissue sensitivity t o insulin or ins ulin resistance [62 ] . tissue insulin sensitivity, the euglycemic - hyperinsulinemic clamp technique has several limitations. First, it is an intensive techn ique that requires skilled assistants. Next, multiple intravenous catheters must be placed, for sample collection and for insulin and glucose infusions. Further, it has been argued that the test is non - physiologic in nature because the insulin concentrat ion achieved is often above the physiologic range [29 ] . Finally, the relationship between glucose disposal rate and concentrations of glucose and insulin is complex and could lead to substantial variation under different experimental conditions [44,62 ] Mi nimal model analysis of insulin and glucose dynamics during a FSIVGT is the third quantitative test used to directly measure insulin resistance. The goal of this test is to assess insulin and glucose dynamics under physiologic insulin concentrations. Min imal model analysis also allows estimation of additional parameters including the acute pancreatic insulin release in response to an intravenous bolus of glucose, glucose effectiveness, and the d isposition index [49 ] . The procedure involves placing bilate ral jugular intravenous catheters, one for glucose and insulin administration and one for sample collection, usually the night before the test to limit any stress of catheter placement on test results. Subjects are also fasted overnight. On the day of th e test, a bolus of glucose (0.1 - 0.3 g/kg) is administered followed by a bolus of insulin (10 - 20 mU/kg) 20 minutes later. Approximately 25 blood samples are collected over the subsequent 4 - 6 hours and glucose and insulin concentrations are subjected to min imal model analysis [62 ] . 15 Minimal model analysis uses two differential equations to describe glucose - time curves, dividing glucose disposal into two phases: glucos e and insulin - mediated disposal. The first equation calculates glucose - mediated glucose dis posal using a single - rate constant. The result provides an estimate of glucose utilization (uptake) following the intravenous bolus of glucose. The second equation calculates insulin - mediated glucose disposal by using a rate constant that incorporates in sulin sensitivity. The result provides an estimate of the ability of insulin to remove glucose from the blood by decreasing endogenous glucose production and increasing the use of glucose by tissue, in response to the bolus of ex ogenous insulin [62 ] . Lim itations of minimal model analysis include an argument that a single - rate constant in the first equation may be inadequate to model the complex interaction of glucose and insulin. Next, another concern has been that hyperglycemia in response to the glucos e bolus could exceed the renal threshold and lead to loss of glucose in the urine during the first 20 minutes of the test. The minimal model does not account for this and, as a consequence, the test has been modified by lower the dose of glucose admin ister ed to 0.1 g/kg [109 ] . Glucose effectiveness (Sg), insulin sensitivity (Si), acute insulin response to glucose (AIRg), and the disposition index (Di) are parameters calculated using minimal model equations. Glucose effectiveness is an estimate of the abi lity of glucose to drive its own disposal (tissue uptake) without the influence of insulin. Insulin sensitivity is a measure of the ability of insulin to promote glucose disposal. Acute insulin response to glucose measures the endogenous insulin output b y the pancreas in response to glucose administration. Lastly, disposition index is the product of acute insulin response to glucose and insulin sensitivity and provides an assessment of beta cell responsiv eness to glucose [49 ] . 16 Along with comparing peak values and general shape of the curves, the statistical analysis of the various glucose and insulin tolerance tests described above includes calculating area under the curves for plasma glucose and serum insulin. These values are calculated with commercia lly available computer software using trapezoida l integration [11 5 ] . The value of the area under the curve can be used to compare different experimental conditions as it is a representation of overall height and width of the glucose or insulin response. In human medicine, it has been shown that there is an increase in the area under the curve for glucose when a high - star ch diet is cons umed [17 ] . In equine research, area under the curve was used to determine the appropriate dosage of dextrose to be admini stered during the frequently sampled intravenous glucose tolerance test to minimize uri ne glucose spill ing [109 ] . Factors Affecting Insulin Sensitivity There are certain factors which affect insulin sensitivity. Age has been shown to affect in sulin sensitivity throughout certain stages in life. In a newborn foal, for the first 24 to 48 hours, the beta cells within the pancreas are still maturing which leads to transient insulin resistance [52 ] . This maturation of beta cells within the pancrea s continues for approximately three months post - parturition at which time the pancreatic cells are functioning at the level of a mature h orse [103 ] . In human children at the onset of puberty, there is a stage of insulin resistance which corresponds with a time of rapid deve lopment and growth [88 ] . Human literature has reported that glucose tolerance deteriorates with age. Using an oral glucose tolerance test in healthy men, peripheral glucose uptake was 3 times greater in young and middle - aged men as co mpared with elderly men. Glucose and insulin responses were similar in all three age groups which suggest that the impairment in the elderly is not at the level of the insulin but a result of impairment in downstream mechanisms of gl ucose metabolism [53 ] . 17 When using the euglycemic - hyperinsulinemic clamp at different levels of glycemia in elderly versus nonelderly subjects, the glucose disposal rate was decreased by 30 - 35 % in the elderly at each level of glycemia with similar levels of hyperinsulinemia. This study reported similar affinity of glucose utilization indicating that there was no impairment in the affinity for the glucose receptor but rather a reduction in the num ber of functional recep tors [30 ] . In conclusion, aging can affect insulin resist ance by both reduced glucose uptake with fewer receptors and impaired intracellular glu cose metabolism [43 ] . Limited research has been conducted investigating the effect of age on insulin sensitivity in the horse. In 2002, following a 0.25 g/kg of body weight oral glucose tolerance challenge, old Standardbred mares (27.0 ± 0.4 yrs) had a significantly greater insulin response as compared to middle - aged (15.2 ± 0.4 yrs) and young (6.8 ± 0.4 yrs) Stand ar dbred mares [69 ] . Furthermore in 2010, sixteen Arabian horses, eight 2 - year olds and eight mature horses (14 ± 0.5 year), underwent weekly glycemic response tests using various feedstuffs. Despite minimal differences in glucose responses between 2 - year olds and mature horses, mature horses had a great er insulin response suggesting that mature horses have reduced ins ulin sensitivity [83 ] . When comparing old (22 years ± 0.7 yrs) and young (7 years ± 0.6 yrs) Standardbred mares using the frequently sampled intravenous glucose tolerance test, old horses h ad a greater AIRg suggesting a greater pancreatic output of insulin in response to the glucose challenge [64 ] . In both humans and horses, it has been shown that insulin sensitivity decreases with age, leading to an increased pr ev alence of insulin resistan ce [80 ] . The mechanisms of digestion post - prandially are also influenced by age. In humans, there seems to be a higher prevalence of gastrointestinal disorders of function and mo tility with a ging [96 ] . It has been reported that peristalsis and gastric contractile force are reduced in aged 18 individuals following a meal. Lipids also have been shown to delay gastric emptying time in e lderly subjects [82 ] . This delay in gastric emptying may also be a result of a reduction in the number of myenteric neurons available to signal neurotran smitter res ponses [96 ] . Aging increases the proliferation of gastric mucosa, however, this mucosa is reported to be more susc eptible to injury [78 ] . The number of gastric and colonic mucosal cells undergoing apoptosis increa sed with a restrictio n in caloric intake [126 ] . Also, the responsiveness of functional receptors for peptides such as gastrin is reduce d with age [96 ] . This suggests that aging influences nutritional and hormonal impacts on gastric mucosa. Disease can h ave major effects on intestinal absorption including atrophic gastritis. Helicobacter pylori infections are associated with a decrease in gastr ic acid secreti on [78 ] . Consequences of atrophic gastritis include bacterial overgrowth in the proximal intesti nal tract and malabsorption [96 ] . While the effect of age on digestion has been researched extensively in the human literature, very little has been reported in the horse. As the horse ages, dentition becomes an important management factor. The process of digestion begins in the mouth with thorough mastication of a primarily fibrous diet. A sufficient surface area is needed for proper mastication of feed. For grinding forage, the horse has cheek teeth that form solid arcades of enamel ridges. The hors e has hypsodont, or low - crowned, teeth which erupt throughout life until approximately twenty years of age. Around this age, the reserve crown is shortened to the extent that t hat tooth is shed [42 ] . This reduces the circumference of the tooth which allo ws for diastemas, spaces between teeth, to form. Additionally, the surface area for gr inding is reduced [65 ] . The type of diet affects the dentition of the horse as well. Horses fed a primarily forage diet have a greater lateral excursion of the arcades and a larger amount of occlusal surface to contact the opposite tooth than when fed a 19 high - concentrate diet. Reduced occlusal wear from a high - concentrate diet also restricts the lateral chewing ac tion of the horse [42 ] . Aging and diet effect the dental health of the horse Another trait influencing insulin sensitivity is breed. It has been documented that ponies are less sensitive to insulin than Standarbred horse s [55 ] . In support, a study reported that donkeys are less insulin sensitive th an ponies or horses [58] . In humans, certain ethnicities are more susceptible to developing insulin resistance. For example, the prevalence of metabolic syndrome in Mexican Americans is significantly higher (32%) than non - Hispanic whites (24%) and non - Hispanic blacks ( 22%) [93] . More specifically, black South African women are more insulin resistant with a lower insulin sensitivity and higher acute insulin response to glucose than white South African women following minimal model analysis [41] . Lastly, it was shown that African - American women have lower insulin sensitivity than Europe an - Amer ican women [44 ] . Specific horse breeds have been identified as having a greater risk of developi ng insulin resistance. These breeds include Morgan Horses, European Warmbloods, American Saddlebreds, Spanish mustangs, an d all pony breeds . One recent stu dy comparing 8 Standardbred , 8 mixed - breed ponies, and 7 Andalusian - cross horses showed that ponie s and Andalusian horses have a lower insulin sensitivity and AIRg as compared to Standardbreds [5] . In summary, there appears to be a genetic predisposition impacting insulin sensitivity. Pregnancy can also have an effect on insulin sensitivity. In human s, insulin resistance during pregnancy, known as gestational diabetes, is an important risk factor le adin g to abortion [70 ] . Pregnant mares have a significantly higher basal insulin and greater insulin response to an intravenous glucose challenge as compa red to non - pregnant controls at less than 270 days of gestation [34] . Additionally, there was a significantly greater insulin secretion at 28 weeks of 20 gestation in pregnant mares versus non - pregnant mares following minim al model analysis [40 ] . Fetal grow th rate has a negative consequence on insulin sensitivity as well. Pony fetuses transferred to Thoroughbred mares during gestation have higher basal insulin levels and greater beta cell response to glucose when allowed to grow larger in u tero than normal [33 ] . This suggests that factors related to pregnancy do have an impact on insulin sensitivity of both mother and offspring. Obesity is very important risk factor influencing insulin sensitivity. Obesity generally results from an imbalance between the e nergy consumed and lost. Through evolution, horses conditions. These genes are detrimental to the majority of horses that live a sedentary life with minimal controlle d exercise. The daily energy requirement of such horses is typically less than what the horses are actually fed whether the horse is stalled or on pasture. This results in excessive caloric intake and deposition of adipose tis sue [39 ] . In the horse, obe sity can be determined by body condition scoring using the Henneke scale (1 - 9). Horses with a score of 8 or 9 are generally considered obese [47 ] . It has been shown in a variety of species including mice, dogs, humans, and rats, overfeeding and accumulat ion of adipose tissue results in a decreased in sulin sensitivity [11 ] . Adipose tissue regulates metabolism by releasing inflammatory cytokines, hormones such as leptin and adiponectin, non - esterified fatty acids (NEFAs), and glycerol. Obesity induces a ch ronic inflammatory state by increasing the production and circulation of these products which have been associated wi t h insulin resistance [127 ] . The exact mechanism is unclear, but it is thought that interference with insulin signaling pathways leads to decreased insulin sensitivity. In previous research comparing lean and obese horses, the results showed 21 that the obese horse has lower insulin sensitivity as well as higher plasma concentrations of insulin, glucose, leptin, and NEFAs. In this study, a 20 % increa se in body weight resulted in com p ensated insulin resistance [11 ] . Other research in the horse has shown that increased body condition and percent fat are associated with increased expression of TNF - - 1 which may inh ibit insulin action [117 ] . In human literature, it has been shown that within hours of an acute increase in NEFA levels in plasma, the development of insulin resistance occurred. Obesity increases the production of adipokines which are cytokines released from adipose tissue th at cause insulin resistance [66] . Also, adipocytes contain an enzyme, 11 - - - HSD1), which converts cortisone to cortisol. Cortisol opposes the effects of insulin on carbohydrate metabolism causing a decrease in glucose up take by peripheral tissues and increased production of gluco se by the liver [71 ] . By examining transgenic mice that overexpress 11 - - HSD1, there was an increased development of visceral obesity, insulin resistance, type 2 diabetes, an d hyperlipidemia [56 ] . Lastly, as a result of adipose tissue reaching maximum storage capacity with excessive glucose being converted to fat, lipotoxicity in other tissues can occur causing alteration in normal cellular function including insuli n resistance [35 ] . Obesity is a causative factor leading to a variety of health disturbances in humans and domestic animals. In the horse, insulin resistance in conjunction with obesity has been associated with an increased preva lence of laminit is [39 ] . However, it has not been dete rmined whether this is a direct correlation or whether obesity increases the prevalence of factors such as insulin resistance and inflammation leading to laminitis. Insulin is a vasoregulatory hormone using nitric oxide to invoke vasodilation. When there is a decrease in the tissue sensitivity to insulin in the lamina of the hoof, there is a reduction in vasodilatory effects of insulin resulting in 22 decreased blood flow to the hoof. Deprivation of nutrients such as glucose to the lamina can lead to separa tion [35 ] . Also, keratinocytes of the lamina have glucose transporters which are reliant on insulin for regulation of hoof growth [111 ] . In the equine literature, it has been shown that prolonged hyperinsulinemia and euglycemia triggered laminitis within 72 hours in healthy, moderate body condition ponies with no his tory of laminitis [4] . To further support this finding, experimental induction of hyperinsulinemia triggered laminitis in young (3 - 4yr), non - obese Standardbred horses within 48 hours [19 ] . T herefore, a reduction in the insulin sensitivity of the hoof and hyperinsulinemia are risk factors for the development of laminitis. Physical activity has been shown to improve insulin sensitivity in many species of animals. In both humans and rats, short periods of acute exercise improved insulin action [48,59] . In the horse, exercise training decreased the acute insulin response to glucose (AIRg) in yo ung Standardbreds [64 ] . Additionally, in both obese and lean mares, insulin sensitivity improved, 60% and 48% respectively, following acute exercise training with no change in body weight. Nine days post exercise training, there was no difference in insulin sensitivity from baseline suggesting physical activity caused an acute improvement only [89] . Di et can influence the prevalence of insulin resistance. Carbohydrate restriction has been shown to improve all markers of metabolic syndro me in humans [118 ] . A low - carbohydrate diet for fourteen days was reported to improve insulin sensitivity by 75% [8] . Following a 56 week diet adaptation to a ketogenic (high fat, low carbohydrate) diet, there was a significant reduction in blood glucose in 31 obese, diabetic patients [15] . In order to reduce the glucose content of the diet, the total carbohydrate cont ent as well as the type of carbohydrate must be considered. Starches are 100% glucose whereas sucrose is approximately 50 % glucose. Substituting sucrose or lactose for starch in a single meal was reported to reduce the glucose area response by 23 approxima tely 40 - 50 % [60] . Dietary protein does not increase blood glucose concentration but does increase insulin secretion. Therefore, consumption of protein with ingestion of glucose will reduce the blood glucose response to the glucose ingestion by increas ing insulin secretion [36] . When adapted to a high monounsaturated fat/low - carbohydrate diet, Type 2 diabetic patients showed significant decrease in postprandial plasma glucose and insulin. Additionally, using euglycemic - hyperinsulinemic clamp, a significa ntly higher insulin - mediated glucose disposal was documented suggesting an increased uptake of glucose by peripheral tissue or improvement in ins ulin sensitivity [86] . In summary, dietary adjustments including limiting carbohydrate intake and increasing p rotein intake can improve tissue sensitivity to insulin in humans. As a non - ruminant herbivore, the horse is well adapted to a diet high in carbohydrates. There are two main sources of carbohydrates in the equine diet: structural carbohydrates (SC) and n onstructural carbohydrates (NSC). Simple sugars, starches, and fructans are examples of NSC while cellulose and hem icelluloses are SC [35 ] . In general, the fibrous part of the diet constitutes the SC content while the high - energy yielding concentrate por tion comprises the NSC content. It should be noted that the forage species in some pastures can contain a high NSC content. Additionally, growing conditions, maturity when harvested, and drying conditions all effect the amount of NSC in hay. Forage ana lysis is a method used to determine the exact amo unt of NSC in hay [46 ] . The type of carbohydrate is also described by a measurement of glycemic index. The glycemic index of a feed is the plasma glucose response to consumption of a measured amount of fee d when compared to a standard challenge. In humans, this standard challenge is a meal consisting of white bread. In the equid, the standards vary from an equivalent weight of oats to an oral dose of glucose or a standardized amount of NSCs in the feed. Hay is generally 24 classified as having a low glycemic index whereas grains have a highe r glyce mic index [92 ] . In general, foods that produce a higher peak and greater overall glucose response are considered to have a high er glycemic index [55 ] and can rang e from 40 - 60% NSC on a dr y matter basis [106 ] . When comparing types of grain supplementation, a starch and sugar concentrate (sweet feed) has a higher glycemic index (> 30 % NSC) than a high fat and fiber concent rate (<20 % NSC) [92 ] . The effect of di et on glucose metabolism was first recognized by Argenzio and Hintz in 1972. Greater glucose availability and improved glucose utilization was seen when ponies were adapted to an oat diet compared with a high fiber diet of alfalfa and beet pulp [ 3 ] . Jacobs and Bolton further explored the effect of diet on glucose metabolism in 1982 by using the OGTT. Horses adapted to a diet of oats and alfalfa hay had a significantly lower glycemic response when compared with horses on a pastur e diet [54] . The OGTT was also used to document a significantly lower glycemic response in ponies adapted to a high fiber pellet diet versu s a hay only di et [80 ] . A lower area under the curve and reduced insulin sensitivity has been documented in horses adapted to a starch an d sugar diet as compared to a fa t and fiber diet [90 ] . In healthy Thoroughbred geldings adapted to a high starch and sugar feed versus a fat and fiber rich feed, there was a decrease in insulin sensitivity when adapted to the starch and sugar diet. Also reported in this study was a lower acute insulin response to glucose and disposition index. This was not an expected finding as acute insulin response to glucose typically compensates for lower insulin sensitivity by increasing in value. This finding sug gests that there was a reduced beta - cell response when adapted to the starch and sugar diet [50 ] . When comparing the insulin and glucose concentrations post - prandially, horses fed a starch and sugar grain had significantly higher plasma concentrations the n horses f ed a fat and fiber diet [50 , 112 ] . In support, Thoroughbred weanlings adapted to a high glycemic meal had a 37 % decrease in insulin 25 sensitivity when compared to a fat and fiber feed. Furthermore, the decreased insulin sensitivity was compensat ed with an increase in the acute insulin response to gl ucose [112 ] . Lean horses fed three diets differing in NSC content showed no difference in glucose and insulin response within treatment suggesting no alteration in peripheral glucose disposal followin g a 90 day diet adaptation. Interestingly, AUCg was significantly lower, peak insulin concentration was significantly greater, and AUCi did not change on day 90 when com pared with day 0 [106 ] . An explanation for these results is unknown as a decreased AU Cg and increased peak insulin do not allow for distinguishing between increased tissue sensitivity to insulin versus reduced insulin sensitivity. Furthermore, a reduction in insulin - induced adipose tissue lipolysis was reported on high NSC diet indication diet associated insulin resistance. In a different animal model, rats fed a high glycemic index starch developed more body fat when compared to rats fed a low glycemic index starch. This suggests that the high glycemic index meal which induces hyperinsu linemia increases the amount of nutrients deposited as f at [87 ] . Meals with a high glycemic index can pose a variety of problems in the horse. Chronic adaption to meals with a high glycemic index causes fluctuations in glycemia and insulinemia. In prev ious research, it was shown that the main energy source of the equid changes in the spring from stored fat to soluble carbohydrates. This switch is associated with an increased prevalence of insulin resistance and increased ri sk of laminiti s [113 ] . It ha s been shown in horses that the greater NSC content of a meal, the higher the insulin and glucose response which may potentiate the development of ins ulin resistan ce [116 ] . Horses given a 6 - hour insulin infusion as a model for a horse adapted to a high NS C diet had a decreased abundance of GLUT - 4, GLUT - 1, and insulin receptor in adipose tissue while GLUT - 1 was increased and insulin receptor was 26 decreased in skeletal muscle. G lucose and lipid transport was a ffected by hyperinulinemia in this study promotin g in sulin resistanc e [105 ] . Large meals consisting of NSC increase the rate of gastric emptying and gut transit time while decreasing digestion of the starch and sugars in the small intestines before reaching the hindgut. Disturbances in the micro flora and fermentation of the hindgut with a diet overload of carbohydrates results in an increase in intestinal permeability [77 ] . This can induce acidosis of the hindgut of the equid as a consequence of undigested starch being rapidly fermented by gram positi ve bacteria. This allows for an influx of endotoxins, resulting in inflammatory responses including onset of laminitis, increased risk of colic and higher risk of effecting mineral content of bone [52, 92 ] . Replacing high NSC concentrates such as sweet feed with a fat and fiber feed has been a popular interest in research due to the possible benefits in the horse industry such as reducing rhabdomyolysis, enhancing both anaerobic and aerobic exercise and reduci ng excitabi lity [125 ] . When looking at gluco se and insulin dynamics, feeding a high fat and fiber diet results in an increase in insulin sensitivity and reduced hyperinsulinemia and more closely mimics the natural state of a grazing ho rse versus a high starch diet [50, 92 ] . Compared to sweet feed a nd a pelleted concentrate, a feed high in fat and fiber had the same energy density but with an insulin response of a typical alfa lfa hay [125 ] . Following a six week diet adaptation period, young, healthy Standardbred horses exhibited no change in insulin sensitivity or glucose tolerance on the fat and fiber diet versus when adapted to a high NSC concentrate diet where insulin sensitivity and glucose tolera nce were decrea sed [90 ] . In humans, both high fat and high carbohydrate diets are known to cause in s ulin resistance [55 ] . 27 Equine metabolic syndrome is being researched heavily today as the direct cause and effect relationships of this syndrome are not completely understood but are very extensive and intertwined. In particular, limited research has been devoted to the effect that age has on the development of equine metabolic disease. Additionally, there is confusion as to what feed is appropriate for the aged or senior horse. There are a variety of different concentrates marketed to both the senior and obese horse including low - starch and senior feeds. Senior feeds are high in starch which has led to veterinarians to recommend low - starch feeds to the aged horse due to the perceived increase in insulin resistance with age. However, low - starch feeds are not a complete feed which is problematic for the aged horse with poor dentition. This has led to some confusion as to what the appropriate diet is for the aged horse. Therefore, the purpose of the following study was to investigate the effect of diet on age - related changes in glucose and insulin dynamics in the horse. 28 CHAPTER 2 EFFECTS OF AGE AND DIET ON GLUCOSE AND INSULIN DYNAMICS IN THE HORSE Introduction Factors affecting insulin sensitivity in equids include breed, body condition, pregnancy and lactation, physical activity, age , and diet [5, 6, 31, 35, 72] . With respect to the latter two, ageing is associated with development of glucose intolerance and insulin resistance (IR) in people, manifested by exaggerated glucose and insulin responses to car bohydrate (CHO) challenge [20, 30, 53] . G reater insulin response s to enteral glucose challenge [69] , as well as to cereal grain meals rich in hydrolysable CHO [83] , have also been documented in mature horses, as compared to young horses. These findings support an age - related decrease in insulin sensitivity in this species as well. Importantly, IR, especially when coupled with genetic predisposition and obesity, is considered a risk factor for development of medical problems including cardiovascular disea ses, dyslipidemias, and type 2 diabetes in people [85] and laminitis i n equids [35, 45, 72, 75] . Diets high in nonstructural CHOs (NSC) have also been implicated in development of IR and type 2 diabetes in people, even after controlling for other risk fact ors [73] . Equids are usually fed forage diets that are relatively low in NSC. Similar to people, when horses ingest larger amounts of NSCs, either in lush pasture grass or when forage is supplemented with cereal grains, greater postprandial glucose and ins ulin responses, along with decreased insulin sensitivity, have been document ed [11, 49, 83, 90] . Ingestion of diets high in NSC may further affect glucose and insulin dynamics by altering the enteroinsular axis, consisting of enteric neuronal signals and gut - derived hormones (incretins) that influence pancreatic insulin release [6,18] . As a 29 conseque nce, when equids require greater caloric intake to meet demands of exercise or lactation, use of oil - fortified feeds has been recommended to decrease the risk of inducing or exacerbati ng IR [122 ] . To date, there has been limited investigation of the inter action of age and diet on glucose and insulin dynamics in healthy, non - obese horses. The objectives of this study were: 1) to measure minimal model parameters during an insulin - modified frequently sampled intravenous glucose tolerance test (FSIGTT) after a daptation to a forage only diet or a forage diet supplemented with either a cereal grain based, starch and sugar (hydrolysable CHO) rich or an oil and fiber rich complementary concentrate feed; and 2) to assess glucose and insulin responses to a standardi zed hydrolysable CHO rich meal challenge (SMC) after diet adaptation in groups of healthy adult and aged horses. It was hypothesized that aged horses, as compared to adult horses, would have a greater acute insulin response to glucose administration (AIRg) and lower insulin sensitivity (SI) during the FSIGTT, regardless of diet. Similarly, glucose and insulin responses to the SMC were hypothesized to be greater in aged than adult horses, again regardless of diet. Finally, in both age groups adaptation to a cereal grain supplemented diet was hypothesized to produce a greater AIRg and a lower SI during the FSIGTT and greater glucose and insulin responses to the SMC, as compared to the other diets. Materials and methods Horses and housing Seventeen mares, inc luding eight adult (8.1 ± 1.6, range 5 - 10 yr) and nine aged (21.9 ± 1.6, range 19 - 24 yr) horses, were studied. Breeds included 12 stock - type horses, four Thoroughbreds, and one Standardbred. Aged mares weighed less than adult horses (455 ± 12 vs. 500 ± 13 kg, P<0.02) throughout the study but body condition score (BCS scale 1 - 9 [47] ) was 30 not different between age groups (median [range] 5 [4 - 7] for adults and 5 [3 - 6] for aged, P=0.20). Prior to the study start, all horses underwent a complete oral examination to ensure presence of all premolars plus molars and minor dental abnormalities were corrected. All horses also received a dose of ivermectin paste. All aged mares had normal overnight dexamethasone suppression test results (ODST, cortisol suppression [<1. 0 ug/dL] 19 h after dexamethasone administration, 4 0 µg/kg, IM [24 ]) and lacked clinical signs of pituitary pars intermedia dysfunction (PPID) [98] . Horses were grouped as pairs, one adult and one aged mare, and maintained in snow covered paddocks or dry l ots to minimize pasture access (January - June). Study design Mares were studied in a Latin square design and fed three diets for 41 d: forage only (HAY); forage supplemented with a high starch and sugar (SS, rich in hydrolysable CHO) cereal grain based fee d (Pleasure Sweet, Buckeye® Nutrition); or forage supplemented with a low starch and sugar, oil and fiber rich feed (FF, Equilibrium Growth, Winergy®). After 21 d on each diet, pairs were moved indoors and housed in adjacent stalls (ambient temperature, 15 ± 7°C) and fed individually for an additional 20 d. While housed in stalls, horses were turned out for exercise in a dry lot for 2 h, 3 d each week. The start of each diet period was staggered by either 2 or 4 wk due to the limited number of stalls availa ble; consequently, testing was performed with groups of 5 - 6 mares at one time. Mares were initially fed 1.6% of body weight (BW), as fed, divided into two equal feedings (0700 and 1700 h): HAY = 1.6% BW hay; SS or FF = 1.0% BW hay and 0.6% BW SS or FF. On d 22 and d 41 of each period, horses were weighed and BCS was assessed by three trained individuals. By d 41 of the first period for the initial cohort of mares studied (February), BW had decreased (~16 and ~27 kg for aged and adult horses, respectively). Consequently, for the remainder of the study, the amount fed was increased 31 to 1.84% BW, as fed: HAY = 1.84% BW hay; SS or FF = 1.15% BW hay and 0.69% BW SS or FF. Insulin - modified frequently sampled intravenous glucose tolerance test (FSIGTT) An i nsulin - m odifi ed FSIGTT [49,109 ] was performed on d 31 or 32 of each feeding period (two or three horses each day). During the afternoon prior to testing, 13.3 cm, 14 G polyurethane catheters were aseptically inserted into both jugular veins and patency was maintained by injection of 1 m L of Na heparin (1000 IU/mL). Mares were fasted overnight (~12 h) prior to testing. Starting at 0900 h, two baseline blood samples were collected 15 min apart prior to glucose administration (0.1 g/kg, IV bolus, 50% dextrose solution, VEDCO, Inc.), followe d 20 min later with insulin administration (20 mU/kg, IV bolus, Novolin R, Novo Nordisk, Inc.). Blood samples were collected from the opposite IV catheter at - 15 , - 1, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16, 19, 22, 23, 24, 25, 27, 30, 35, 40, 50, 60, 70, 8 0, 90, 100, 120, 150, and 180 min after glucose administration. At each sampling time, a 15 mL blood sample was collected with 4 mL transferred to a plastic tube containing lithium heparin, placed on ice and min of collection and 10 mL transferred to a tube without anticoagulant that was allowed to clot at room temperature for 2 h prior to centrifugation. Plasma and serum were harvested and stored at - Standardized meal challenge (SMC) On d 42 of each feeding period, horses were f ed a standardized amount of the cereal based meal (4 g/kg BW of SS) without hay for their morning feeding. A 13.3 cm, 14 G polyurethane catheter was aseptically inserted into the jugular vein during the afternoon prior to the SMC and patency of the catheter was maintained by injection of 1 mL of Na heparin (1000 IU/mL). Again, mares were fasted overnight (~12 h) prior to testing and three baseline blood 32 samples were collected during the 30 min prior to the SMC (of fered at 0900 h) to determine baseline serum glucose and insulin concentrations. Feed remaining 1 h after offering the SMC was removed and weighed. Blood samples were collected 15, 30, 45, 60, 75, 90, 105, 120, 150, 180, 210, 240, and 360 min after the mea l was offered. At each time point, a 15 mL blood sample was collected and processed as described for the FSIGTT. Due to the staggered start of diet period one, six FSIGTT and SMC tests were performed when mares were on the lower plane of nutrition (1.6% BW , two mares on each diet) while all other tests were performed after at least 1 week on the higher plane of nutrition (1.84% BW). The study protocol and all procedures performed were approved by the Michigan State University Institutional Animal Care and U se Committee (approval #11/09 - 174 - 00). Sample analysis Hay and concentrate feed samples were collected several times during the course of the study and proximate analysis of pooled, well - mixed samples was performed [28] . Additional aliquots of pooled feed samples were submitted for analysis of starch, water soluble CHO, and ethanol soluble CHO (Dairy One DHIA Forage Testing Laboratory). Plasma glucose concentrations were measured in duplicate using a membrane based glucose oxidase system to catalyse oxidati on of glucose to gluconic acid and hydrogen peroxide (YSI 2300 STAT Plus Glucose and Lactate Analyzer, Yellow Springs Instruments). Serum insulin concentrations were determined in duplicate using a radioimmunoassay (Coat - a - Count Insulin, Siemens Medical So lutions Diagnostics) previously validated for equine sampl es [108 ] . Inter - assay CV was 9.3% and intra - assay CVs were 5.2% and 8.6% for high and low control samples, respectively. Calculations and statistical analysis 33 SI, AIRg, glucose effectiveness (Sg) and disposition index (DI = SI × AIRg) were calculated by minimal model analysis of glucose and insulin data from the FSIGTT (MinMod Millenium, V ersion 6.02; [10] ). For the SMC, peak glucose and insulin concentrations and time to peak concentrations were d etermined, and area under the 360 min curves for glucose (AUCg) and insulin (AUCi) were calculated by the trapezoidal method, using commercially available computer software (GraphPad Prism, version 6.0, GraphPad Software Inc., San Diego, CA). Outlier data One outlier (adult) for the FF diet was removed from analysis of insulin response during the SMC. Differences between ages and diets for minimal model variables and SMC we re determined via mixed ANOVA for repeated measures and Tukey - Kramer post hoc adjustment (SAS, version 9.4, SAS Institute Inc., Cary, NC). Data in the text are presented as means ± SD Results Nutrient composition of the three feedstuffs is presented in Table 1. The SS concentrate contained 43% NSC, ~3 and 4 times more NSC than FF and HAY, respectively, while the FF feed contained 11% fat, ~1.5 and 2.3 times more fat than SS and HAY, respectivel y. When either supplemental feed was combined with hay, digestible energy provided was ~12% (FF) to ~19% (SS) greater when compared to HAY alone: 40.6 ± 0.9, 45.6 ± 0.7, and 48.3 ± 0.9 kcal/kg/d for HAY, FF and SS, respectively, (P<0.01, with SS and FF sig nificantly greater than HAY). Overall, horses consumed the diets well over the course of the study. As mentioned, BW decreased (P<0.05) in both age groups from barn entry on d 22 to barn exit on d 41 during period 1, regardless of diet (Table 2). BW loss w as attributed to the combined effects of season (winter) and a transient period of decreased feed intake when horses were initially placed into stalls 34 during period 1, despite having been previously housed in the same facility. Once the amount of feed prov ided was increased, BW and BCS tended to increase a small amount over the remainder of the study in both adult and aged mares (Table 2). Minimal model parameters AIRg was significantly affected by age (P<0.02), with mean AIRg greater in aged when compared to adult mares across all diets. In aged mares AIRg was lower (P<0.05) after adaptation to the HAY diet when compared to SS; AIRg did not differ among diets in adult mares. SI was significantly greater (P<0.01) in adult when compared to aged mares, regard less of diet (Table 3). Further, a significant age × diet interaction (P<0.01) was also detected. For adult mares there was no difference in SI among the three diets but in aged mares SI was lower with HAY and FF when compared to SS (Table 3). Although the re was no difference (P=0.08) between age groups, DI was greater (P<0.01) after adaptation to the SS diet, as compared to the HAY diet. In adult mares, mean DI was also greater (P<0.01) with FF than HAY. There were no significant differences in Sg with eit her age or diet (Table 3). Insulin and glucose dynamics during the SMC Glucose (Fig. 1) and insulin (Fig. 2) concentrations reached peak values within 90 - 120 min after the meal was offered. There was no effect of age (P=0.32) or diet (P=0.10) on peak gluc ose concentration after eating a meal rich in hydrolysable CHO. Similarly, time to peak glucose concentration was not affected by age (P=0.13) or diet (P=0.10). In contrast, peak insulin concentration, but not time to peak concentration, was greater (P<0.0 3) in aged when compared to adult mares. There was no effect of diet on peak (P=0.80) or time to peak (P=0.18) insulin concentration (Table 4). 35 AUCg was not different (P=0.66) between age groups, regardless of diet (Table 4). However, AUCg was lower (P < 0. 05) after adaptation to SS, in comparison to HAY, in both age groups. AUCi was lower (P<0.03) in adult mares, as compared to aged mares, regardless of diet (Table 4). Although AUCi in adult horses was 40% lower after adaptation to SS, as compared to the ot her diets fed to this age group, this was not a significant finding (P=0.29). Not surprisingly, AUCg and AUCi were positively correlated (r=0.44, P<0.01). A limitation of the SMC was incomplete ingestion, defined as leaving more than 0.4 kg (~20%) of the m eal after 60 min, by four mares (two adult and two aged mares) in eight of the trials (one, four, and two of these instances occurring after adaptation to HAY, FF, and SS, respectively). When data for these eight studies were compared to data for the remai ning 36 studies during which >90% of the meal was ingested, AUCg was lower (P<0.03) for the eight studies with partial meal consumption but there was no difference in AUCi (P=0.78) (Fig. 3). Discussion In the study reported here, glucose and insulin dynami cs were affected by age and diet in healthy horses. Consistent with findings of Liburt et al. [64] , AIRg was greater and SI was lower in aged mares when compared to adult mares. Notably, diet did not significantly affect AIRg or SI in adult mares; whereas, both SI and AIRg were greater in aged mares after adaptation to SS when compared to HAY (and FF for SI). These age - and diet - associated differences in glucose and insulin dynamics in response to IV glucose challenge were, for the most part, corroborated b y findings during the SMC. Specifically, peak insulin concentration and AUCi were greater in aged than in adult mares with all diets. In contrast to our hypothesis, however, AUCg in aged mares was lower after adaptation to SS when compared to the other die ts, although this finding was consistent with the greater SI observed in aged mares adapted to this diet. 36 A greater increase in circulating insulin concentration, in response to both IV glucose (AIRg) and ingestion of the SMC, in aged mares could be a con sequence of increased pancreatic insulin release, decreased insulin clearance, or a combination of the two. A growing body of evidence in other species suggests that insulin action on target tissues decreases with age, due to decreased responsiveness of in sulin signaling pathways that lead to translocation of GLUT - 4 vesicles to the plasma m embrane [12, 43, 79, 123 ] . Multiple factors including increased visceral adiposity, decreased lean muscle mass, mitochondrial dysfunction, inflammation and oxidative stress, and reduced physical activity all contribute to development of IR with ageing in people [20] - cell insulin release is required to re - establish euglycemia following a meal. Similar to people, aged horses tend to have decreased muscle mass (supported by a lower BW despite similar size in aged mares in this study) and less structured physical activity and these factors likely contribute to aging - associated IR in this species as well. In support, exercise trainin g of aged mares has been shown to decrease AIRg and improve SI, although the former was not a significa nt finding [64] . Cellular mechanisms of IR with aging in horses have been litt le studied. Waller et al. [121 ] demonstrated decreased basal and insulin - s timulated skeletal muscle cell surface GLUT - 4 expression in IR mares, as compared to insulin sensitive mares, despite similar skeletal muscle cell GLUT - 4 content. This finding suggests that insulin signaling pathways are dysfunctional in IR horses, unrelat ed to age. Increased amounts of inflammatory cytokines (tumor necrosis factor - - 1, and interleukin - 6) have also been measured in blood and insulin - sensitive tissues in IR horses, although studies have yielded conflicting results and age may ha ve confounded the resul ts [105, 117, 120 ] . A possible role for inflammatory cytokines and oxidative stress, implicated as contributing factors to mitochondrial dysfunction w ith age in people [20] , in 37 development of IR with aging in horses remai ns unclear [ 64] . Finally, although all aged mares had normal ODST results, this test may be insensitive for detection of earlier sta ges of PPID ([76] ; D. McFarlane, personal communication 2015). Thus, differences in function of the hypothalamic - pituitary - adrenal axis between adult and aged mares cannot be excluded as an additional factor contributing to development of IR with aging in horses. - cell release) from portal blood during first pass through the liver and the rema inder is cleared by uptake and degradation in target tissues and renal elim ination [21] . A potential role for altered hepatic insulin clearance in development of IR with aging in horses remains uncertain. An early study found a longer half - life of exogenou s insulin in horses (33 min) as compared to people (3 - 8 min) [67] , suggesting that peripheral insulin clearance can also differ between species. Finally, a decline in renal function in aged animals could prolong insulin half - life; however, decreased hepati c or renal elimination would be an unlikely explanation for either the immediate insulin response during the FSIGTT (AIRg) or postprandially following the SMC. These acute insulin responses suggest that - cell secretion plays a greater role, as compared to alterations in clearance, in the exaggerated insulin response observed in aged mares. With regard to diet, our hypothesis that AIRg would be greater after adaptation to SS, when compared to HAY or FF, was only supported in aged horses (AIRg was numerically highest in adult horses after adaptation to SS, but this was not a significant finding). In contrast, our hypothesis that SI would be lower after adaptation to SS, as compared to HAY or FF, was refuted. In fact, SI was greatest in ag ed horses after adaptation to SS (again, although not a significant finding, SI was also numerically highest after adaptation to SS in adult horses). A similar finding of a greater SI (but not a greater AIRg) after 20 wk of feeding a glucose 38 supplemented ( 1.5 g/kg BW of glucose) meal once daily to increase adiposity in horses and ponies (aged 5 - 19 years) was recently re ported by Bamford et al. [7] . This finding also contradicted their hypothesis that inducing obesity by increasing NSC intake would decrease SI, as had been previously reported in Arabian horses that were fed increasing amounts of a sweet feed to indu ce obesity [11] . They suggested that obesity per se may not affect insulin sensitivity, at least over a short period of time, while chronic obesit y may pose a different risk. Similar to our findings, when obesity was induced by feeding an oil supplemented diet (25% of DE as oil), SI was u nchanged [7] . Disposition index, the product of AIRg and SI, was nearly 3 - fold greater in both age groups after a - cell responsiveness to IV glucose adm inistration [16] after adaptation to a diet high in hydrolysable CHO. In contrast, the ability of glucose to drive its own disposal (tissue uptake) without the influence of insulin (Sg) was not different between diets or age groups. This finding suggests that neither diet nor age had a significant effect on peripheral tissue glucose transporters that are not regulated by insulin (i.e., GLUT - 1 and GLU T - 1 2) [2, 119 ] . The finding of a lower AUCg response to the SMC after adaptation to SS in both age groups was also contrary to our hypothesis. However, adaptation to SS likely increased small intestinal capacity for glucose uptake by upregulation of sodium/glucose cotrans porters (SGLT1) on enterocytes [14, 102 ] . SGLT1 expression is regulated by the enteric nervous system. Increased amounts of luminal sugars lead to greater secretion of gut hormones (incretins), via stimulation of the sweet receptor expressed on enteroendocrine cells that are dispersed within the intestinal mucos a [102 ] . These hormones include glucagon - like peptides 1 and 2 (GLP - 1 and GLP - 2) and glucose - dependent insulinotrophic peptide (GIP). Intravenous administration of GLP - 2 increases SGLT1 ex pression in absorptive enterocytes indirectly, through stimulation of 39 GLP - 2 receptors expressed on ent eric neurons [17] . In people, release of incretins GLP - 1 and GIP during a meal a ccounts for 50 60% of pancreatic insulin secretion by stimulating specific r - cells [81] . Both upregulation of SGLT1 transporters and increased activity of the enteroinsular axis after adaptation to SS should increase the rate of glucose absorption and uptake by peripheral tissues. Thus, the lower AUCg res ponse to the SMC after adaptation to SS in both age groups could potentially be explained by increased intestinal glucose uptake in combination with enhanced hepatic and peripheral tissue glucose disposition, blunting the rise in circulating glucose concen tration. There were several limitations of this study including transient weight loss in the first cohort of horses studied in period 1; however, this problem occurred with both age groups and across diets. It should be emphasized that the SMC was designe d to be an integrated assessment of intestinal CHO absorption, the enteroinsular axis, and tissue glucose disposition, rather than an attempt to develop a diagnostic test for IR. Incomplete meal ingestion would represent a substantial limitation as a diagn ostic test, as compare d to oral sugar administration [99 ] . Fortunately, in this study incomplete ingestion of the SMC was similar across age groups and diets, minimizing any confounding effect. Conclusions Insulin responses to IV or enteral CHO challenge i ncrease with age in healthy horses, - cell senescence appears to be an - cell failure (and development of uncompensated IR and type 2 d iabetes mellitus) [20] . Our group of aged horses studied was not overweight and did not have - be established for horses, as many horses between 20 - 30 years of age remain healthy and in 40 regular exer cise. Curiously, despite a greater SI and a modest postprandial increase in insulin response, in response to a high hydrolysable CHO meal challenge, after adaptation to the SS diet in aged mares, glucose clearance (AUCg) improved in both age groups after a daptation to the SS. This finding could suggest that addition of some hydrolysable CHO to a forage diet may actually sensitize the enteroinsular axis and enhance postprandial glucose clearance, regardless of age. However, further studies are needed before such a dietary recommendation could be made, especially for ponies and other breeds with lower insulin sensitivity as well as for overweight equids. Finally, it remains unclear whether or not intermittent postprandial hyperinsulinemia, and possibly subclin ical PPID, may place older horses at greater risk for development of laminitis. 41 CHAPTER 3 CONCLUSIONS This study was the first to look at the effects of age and diet on tissue sensitivity to insulin in adult and aged horses. Con sistent with previous reports [64, 69 ] , older horses have reduced tissue sensitivity to insulin. L imited research has been conducted looking at the underlying mechanism. The greater circulating insulin concentrations in aged horses could be a result of increased pancreatic ou tput, decreased insulin clearance or a combination of both. The insulin pathway could have fewer insulin receptors available on the cell surface, decreased receptor affinity to insulin, or some downstream pathway defect. There is evidence that tissue sen sitivity to insulin decreases with age due to a decrease in the translocation of GLUT - 4 vesicles to the plasma membrane [ 12, 43, 79, 123 ]. In human literature, it is widely accepted that increased adipose tissue, decreased lean muscle mass, mitochondrial d ysfunction, inflammation and oxidative stress, and reduced physical activity all contribute to the development of in sulin resistance [20 ] . Aged horses have been reported to be in a pro - inflammatory state which could lead to insulin resistance. Two pro - in flammatory cytokines, TNF - alpha and IL - 1beta, are thought to impair insulin sensitivity whereas IL - 6 appears to improve insulin sensitivity by down regulating TNF - alpha and IL - 1 beta. Increased amounts of inflammatory cytokines (TNF - alpha, IL - 1, and IL - 6) have been measured in blood and tissues of insulin resistance horses [105, 117, 119 ] . The role of these inflammatory cytokines and oxidative stress on mitochondrial dysfunction in the development of insulin resistance remains unclear however. In addition to age, diet is an important consideration when examining glucose and insulin dynamics. When comparing the three diets (HAY, SS, and FF), horses adapted to a HAY diet 42 had a significantly greater glycemic response. This finding is supported by Murphy and colleagues who reported a greater area under the curve for glucose when adapted to hay only diet as compared to a high fiber p elleted meal [80 ] . Interestingly, we also found greater insulin sensitivity when mares were adapted to the SS diet wh ich did not support our hypothesis or previo us research [90 ] which reported a greater area under the curve for insulin following a high sugar and starch diet. Pratt et al. also reported an increase in body weight and body condition score which may have in fluenced their results. Additional research reported lower insulin sensitivity when adapted to the high starch and sugar diet as compared to a fat and fiber diet [49 ] . One limitation of this paper is only 4 horses were studied so sample size could be a c onfounding variable for the results. Like the effect of age on insulin sensitivity, limited research has been done to identify the mechanism behind diet adaptation and the effect on insulin sensitivity. There are a variety of possible mechanisms leading t o improved glucose uptake following diet adaptation to the high sugar and starch meal. Diet adaptation may increase the number of insulin receptors on the cell surface, increase the affinity of the insulin receptor, or up regulate downstream signaling wi thin the insulin pathway leading to an increased number of GLUT - 4 receptors for glucose uptake. These are just a couple of the many possible causes for why mares adapted to a starch and sugar feed showed improved insulin sensitivity. To further explore m echanisms, a review of how glucose is absorbed by the gastrointestinal tract is warranted. Many horses today have been adapted to a diet of forage and concentrate. The concentrate is typically added to provide enough energy for the demands of work and pe rformance. These concentrates are high in hydrolysable carbohydrates which are broken down in the small intestines by pancreatic alpha - amylase and the brush border disaccharidases to 43 monosaccharides such as glucose. Glucose and galactose are transported across the brush border membrane of the enterocytes by SGLT1/Na+/glucose transporter while fructose is transported by a sodium independent transporter, GLUT5. These monosaccarides accumulate in the enterocytes until moving down a concentration gradient on the basolateral side by GLUT2 [9 ] . Horses adapted to a pasture diet have the greatest glucose absorption in the proximal intestine (duodenum>jejunum) with very low expres sion of SGLT1 in the ileum [25 ] . Although limited research has been conducted lookin g at diet adaptation in the small intestines of the horse, Dyer and colleagues studied the expression of SGLT1 within the duodenum and ileum of 6 adult Standardbred geldings. Following 3 months of adaptation to a timothy hay diet, biopsies of the duodenum and ileum were harvested via laparoscopic technique. Horses were then introduced to 60 % hay, 40 % grain (3.3 g starch/kg bwt/day) diet. Biopsies were taken after 1 week and 1 month. Lastly, horses were adapted to 40 % hay, 60 % grain (6.0 g starch/kg bwt/day) and biopsies were harvested after 1 month. When adapted to the hay only diet, the greatest abundance of SGLT1 receptors was in the duodenum as compared to the ileum. After 1 week of eating the concentrate, there was no change in the duodenum exp ression but a 2 fold increase in SGLT1 expression in the ileum. After 1 month adaptation to the diet, the ileum expression did not change from 1 week of adaptation but the duodenum expression doubled. Lastly, after increasing the amount of concentrate fo r 1 additional month, no further expression was appreciated in the duodenum but the ileum had a 3 fold higher expression than the hay only diet. A similar pattern was seen with the GLUT2 transporter suggesting that adaptation to the high sugar and starch diet increased glucose transport into the blood stream by the enterocytes [26] . More recen t work supports the upregulation of the SGLT1 on enterocytes following adaptation to a high sugar and starch meal [ 17 , 102 ] . 44 The amount of insulin released to accele rate glucose uptake is partly regulated by the enteroinsular axis which consists of both neural and hormonal factors. When the small intestine lumenal glucose is above threshold, the sweet receptor is activated. This receptor causes secretion of gastroin testinal hormones called incretins. Incretins are synthesized by the endocrine cells of the gastrointestinal tract and are released with food absorption. Incretins promote release of insulin under hyperglycemic conditions. Insulin secretion following an oral glucose tolerance test is greater than an intravenous bolus of glucose because incretins are ac tivated [22] . In humans, there are two gut hormones shown to act on incretins, glucose - dependent insulinotropic polypeptide (GIP) and glucagon - like peptid e (GLP - 1). Another hormone, GLP - 2 does not increase insulin secretion but does modulate intestinal growth, blood flow, and SGLT1 in several species. The receptors for GLP - 2 are in the enteric neurons suggesting the enteric nervous system is involved in u pregulation of SGLT1 receptors as well. It has been shown that GIP, GLP - 1, GLP - 2 and sweet receptors are in the horse intestines [6,18 ] . GLP - 1 concentration was positively correlated with insulin concentration postprandially documenting an association between incretins and insulin concentration in equids [6] . Duhlmeier and colleagues investigated the enteroinsular axis in Shetland ponies and Standardbred horses following an oral glucose tolerance test and intravenous glucose t olerance test. Results showed an increased GIP following the oral glucose challenge only. One pony had an exaggerated response to both of the challenges indicating insulin resistance. This pony had a 3 times higher GIP level than the other ponies and ho rses in the study supporting a greater insulin secretion requirement to uptake glucose for an insulin resistant animal [22] . A similar study investigated the effect of different diets on the enteroinsular axis of 8 Shetland ponies. Ponies fed a high fat diet for 5 weeks had a greater plasma GIP concentration that those adapted 45 to a sugar and starch diet. This increase in GIP may be a stimulus for insulin hypersecretion and insulin r esistance [97] . Glucose uptake from the blood stream by the skeletal musc le and adipose tissue is stimulated by increased insulin concentration via the insulin receptor on the cell surface. The main receptor that facilitates glucose uptake is GLUT4 and partially by GLUT 12 [104 ] . When there is an increase in insulin concentrat ion, translocation of intracellular GLUT4 to the cell surface occurs resulting in glucose uptake. This is the rate limiting step in glucose uptake in insulin sensitive tissues. Waller and colleagues assessed the GLUT4 and GLUT12 expression in both muscle and adipose tissue of 10 light breed mares, 5 of which were considered insulin resistant based on FSIGT results. They found that insulin resistance selectively decreased active basal cell surface GLUT4 in skeletal muscle without altering GLUT12 or total GLUT4 or GLUT 12 content. Furthermore, there was no enhancement with in vitro insulin stimulation. These insulin resistant mares showed diminished expression of GLUT4 in omental fat which had the greatest GLUT4 content in insulin sensitive mar es [120, 12 1 ] . These results suggest that the mares in our study could have had an upregulation of GLUT4 transporters to the cell surface following diet adaptation. This would allow for increased glucose uptake following a glucose bolus when adapted to the high sug ar and starch diet. Our research further supports an age - related increase in insulin resistance in healthy, non - obese horses. Diet is an important consideration when studying glucose and insulin dynamics. We reported increased tissue sensitivity to insul in following adaptation to a high sugar and starch diet. This finding raises a variety of management concerns. For example, would a horse with the phenotype of a Metabolic Syndrome horse benefit from introducing small amounts of a high sugar and starch d iet prior to being placed on lush, spring pasture. The amount of concentrate 46 and the duration needed to see the improved insulin sensitivity was beyond the scope of this study but would be one area where further investigation is warranted. The exact mech anisms behind how age and diet affect tissue sensitivity to insulin remain unknown but are interesting areas for future research. 47 APPENDIX 48 Table 1 Nutrient composition of the three diets fed, expressed on a dry matter basis, determined by proximate analysis of well - mixed composite samples of several aliquots of each feedstuff collected multiple times during the study; digestible energy (DE) of each diet was calculated using measured DE for each food stuff and the amount fed: HAY = grass hay; FF = mix of hay and fat and fiber rich concentrate feed; and SS = mix of hay and cereal grain feed rich in nonstructural CHO (sugar and starch feed). HAY SS FF NDF (%) 61.4 25.0 42.3 CP (%) 7.9 13.2 14.9 Fat (%) 3.56 5.28 8.30 Calcium (g/kg) 7.9 11.4 15.9 Phosphorus (g/kg) 1.6 7.5 5.4 Lignin (%) 6.9 2.8 4.1 WSC (%) 10.6 7.6 8.6 ESC (%) 5.7 7.0 6.9 Starch (%) 0.5 35.2 5.4 49 Table 2 Bodyweights (BW) and body condition scores (BCS) at barn entry (day 22) and at the end (day 41) of each feeding period in eight adult horses and nine aged horses. Values reported as means ± SEM. Period 1 Period 2 Period 3 Group day 22 day 41 day 22 day 41 day 22 day 41 BW (kg) adult* 509 ± 10 a 482 ± 12 x 517 ± 12 a 522 ± 12 y 518 ± 11 a 520 ± 7 y aged 466 ± 13 ab 449 ± 13 x 456 ± 15 a 453 ± 17 x 464 ± 15 b 464 ± 16 y BCS (1 - 9) adult 5.1± 0.1 a 4.9 ± 0.2 x 5.2 ± 0.1 ab 5.3± 0.1 xy 5.4 ± 0.2 b 5.6 ± 0.2 y aged 4.8 ± 0.3 ab 4.6± 0.2 x 4.5 ± 0.3 a 4.7 ± 0.3 x 5.0± 0.3 b 5.1± 0.3 x * indicates significant (p<0.05) difference between adult and aged horses for all times Within age group, means with different superscript letters differ at P < 0.05 (a,b for day 22 [barn entry] and x,y for day 41 [period end]) 50 Table 3 Acute insulin response to glucose (AIRg) infusion, insulin sensitivity (SI), glucose effectiveness (Sg), and disposition index (DI) determined by minimal model analysis of glucose and insulin data obtained from an insulin - modified frequently sampled i.v. glucose tolerance test performed in eight adult horses and nine aged horses that had been adapted to three diets for 4 wk: HAY (grass hay only); SS (hay plus a cereal grain feed rich in hydrolyzable CHO [sugar and starch feed]); and FF (hay plus a fat and fiber rich concentrate feed). Values reported as means ± SEM. Group HAY SS FF AIRg - 1 adult* 113.1 ± 20.4 a 157.9 ± 24.7 a 140.4 ± 19.2 a aged 209.7 ± 29.5 a 286.0 ± 35.5 b 238.1 ± 28.5 ab SI - 1 - 1 - 4 adult* 2.52 ± 0.53 a 3.64 ± 0.55 a 3.39 ± 0.52 a aged 0.77 ± 0.27 a 2.19 ± 0.35 b 0.99 ± 0.24 a Sg (min - 1 - 2 ) adult 1.87 ± 0.25 2.73 ± 0.32 2.34 ± 0.27 aged 1.81 ± 0.23 1.90 ± 0.17 1.80 ± 0.21 DI 2 ) adult 2.41 ± 0.76 a 6.22 ± 1.1 b 5.11 ± 0.78 b aged 1.56 ± 0.65 a 5.13 ± 0.89 b 2.45 ± 0.67 a * indicates significant (p<0.05) difference between adult and aged horses for all dietary treatments Within age group, means with different superscript letters differ at P < 0.05 51 Table 4 Measures of glucose and insulin responses before, during, and after the standardized CHO - rich meal challenge (4.4 g/kg of the SS cereal grain feed) administered in eight adult horses and nine aged horses that were adapted to three diets for 6 wk: HAY (grass hay only); SS (hay plus a cereal grain feed rich in nonstructural CHO [sugar and starch feed]); and FF (hay plus a fat and fiber ric h concentrate feed). Values reported as mean ± SEM. Group HAY SS FF - 1 ) adult 90.9 ± 2.7 90.6 ± 1.7 90.3 ± 2.4 aged 87.3 ± 3.0 89.5 ± 1.9 88.4 ± 2.7 - 1 ) adult 139.2 ± 6.5 124.2 ± 4.8 144.8 ± 9.3 aged 142.9 ± 8.1 123.5 ± 4.4 134.3 ± 5.9 time to peak glucose (min) adult 122.5 ± 12.0 105.0 ± 4.6 108.8 ± 6.8 aged 136.9 ± 9.8 106.9 ± 7.3 118.3 ± 7.1 - 1 3 adult 88.0 ± 15.9 a 49.1 ± 12.0 b 81.2 ± 16.0 a aged 112.8 ± 15.3 a 52.8 ± 8.5 b 79.2 ± 10.1 ab - 1 ) adult 2.7 ± 0.2 2.8 ± 0.6 3.3 ± 0.3 aged 4.6 ± 1.0 4.9 ± 0.8 4.0 ± 0.3 - 1 ) adult* 64.5 ± 11.0 53.4 ± 7.9 65.2 ± 8.5 aged 117.8 ± 16.4 151.6 ± 25.6 144.2 ± 19.9 time to peak insulin (min) adult 120.0 ± 14.8 97.5 ± 3.7 113.6 ± 12.5 aged 185.6 ± 36.1 125.6 ± 17.7 143.3 ± 18.4 AUCi - 1 3 adult* 12.9 ± 2.7 8.1 ± 1.3 12.2 ± 2.2 aged 24.1 ± 3.0 22.6 ± 5.6 26.4 ± 3.9 * indicates significant (p<0.05) difference between adult and aged horses for all dietary treatments Within age group, means with different superscript letters differ at P < 0.05 52 Figure 1 Mean ± SEM glucose concentrations in response to the standardized meal challenge (0.4 g/kg SS concentrate feed offered for 60 min) in adult (bottom panel) an d aged (upper panel) mares. The filled bars above the x - axes indicate time points that were different ( P <0.05) from baseline; asterisks indicate time points at which SS < HAY ( P < 0.05); cross indicates a time point at which SS < FF ( P < 0.05). 53 Figure 2 Mean ± SEM insulin concentrations in response to the standardized meal challenge (0.4 g/kg SS concentrate feed offered for 60 min) in adult (bottom panel) and aged (upper panel) mares. The filled bars above the x - axes indicate time points that were different ( P <0.05) from baseline; asterisk indicates a time point at which SS > HAY ( P < 0.05). 54 Figure 3 - 1 3 , black fill) - 1 3 , gray fill) in mares that consumed >90% (full) of the SS meal in the standardized meal challenge (n = 36 studies) as compared to mares that left more the 20% (partial) of the SS meal after 60 min (* denot es P < 0.05). 55 REFERENCES 56 REFERENCES 1. Albert i, K. G. and P. Z. Zimmet, Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisiona l report of a WHO consultation. Diabet Med,1998. 15 (7): p. 539 - 553 . 2. Annandale, E. J., S. J. Valberg, et al., Insulin sensitivity and skeletal muscle glucose transport in horses with equine polysaccharide storage myopathy. Neuromuscul Disord, 2004. 14 (10): p.666 - 674. 3. Argenzio, R.A. and H.F. Hintz, Effect of diet on glucose entry and oxidation rates in ponies, J Nutr, 1972. 102 (7): p. 879 - 892. 4. Asplin, K. E., M. N. Sillence, et al., Induction of laminitis by prolonged hyperinsulinae mia in clinically normal ponies, Vet J , 2007. 174 (3): p. 530 - 535. 5. Bamford, N. J., S. J. Potter, et al., Breed differences in insu lin sensitivity and insulinemic responses to oral glucose in horses and ponies of moderate body condition score, Domest Anim Endocrinol , 2014. 47 : p. 101 - 107. 6. Bamford, N.J.,C.L . Baskerville, et al., Postprandial glucose, insulin, and glucagon - like peptide - 1 responses of different equine breed adapted to meals containing micronized maize, J Anim Sci, 2015a. in press. Doi:10.2527/jas.2014 - 8736. 7. Bamford, N.J., S.J. Potter, et al., Effect of increased adiposity on insulin sensitivity and adipokine concentrations in h orses and poines fed a high fat diet, with or without a once daily high glycaemic meal, Equine Vet J, 2015b. in press. Doi:10.1111/evj.12434. 8. Boden, G., K. Sargrad, et al., Effect of a low - carbohydrate diet on appetite, blood glucose levels, and insulin resistance in obese patients with type 2 diabetes, Ann Intern Med, 2005. 143(6): p. 403 - 411. 9. Boron, W.F. and E.L. Boulpaep, Medical Physiology , 2009. Philadelphia, PA: S aunders. 10. Boston, R. C., D. Stefanovski, et al., MINMOD Millennium: a computer program to calculate glucose effectiveness and insulin sensitivity from the frequently sampled intr avenous glucose tolerance tes t , Diabetes Technol Ther , 2003. 5 (6): p. 1003 - 1015. 11. Carter, R. A., L. J. McCutcheon, et al., Effects of diet - induced weight gain on insulin sensitivity and plasma hormone and lipid concent rations in horses, Am J Vet Res , 2009. 70 (10): p.1250 - 1258. 57 12. Couet, C., J. Delarue, et al., Age - related insu lin resistance: a review, Horm Res ,1992. 38 (1 - 2): p. 46 - 50. 13. Cunningham, J.G. and B.G. Klein , Veterinary Physiology. 2007. St. Louis, MO: Saunders 14. Daly, M. E., C. Vale, et al., Acute effects on insulin sensitivity and diurnal metabolic profiles of a high - sucrose com pared with a high - starch diet, Am J Clin Nutr ,1998. 67 (6): p. 1186 - 1196. 15. Dashti, H.M., T.C. Matt h ew, et al., Beneficial effects of ketogenic diet in obese diabetic subjects, Mol Cell Biochem, 2007. 302 (1 - 2): p. 249 - 256. 16. Denti, P., G. M. Toffolo, et al., The disposition index: from individual to population approach, Am J Physiol Endocrinol Metab , 2012. 303 (5): p. E576 - 586. 17. Daly, K., M. Al - Rammahi, et al., Expression of sweet receptor components in equine small intestine: relevance to inte stinal glucose transport, Am J Physiol Regul Integr Comp Physiol, 2012. 303 (2): p. R199 - 208. 18. de Graaf - Roelfsema, E., Glucose homeostasis and the enteroinsula r axis in the horse: A possible rol e in equine metabolic syndrome,Vet J, 2014. 199 : p. 11 - 18. 19. de Laat, M. A., C. M. McGowan, et al., Hyperinsulinemic laminitis, Vet Clin North Am Equine Pract , 2012. 26 (2): p. 257 - 264. 20. de Tata , V., Age - related impairment of pancreatic beta - cell function: pathophysiol ogical and cellular mechanism, Front Endocrinol , 2014. 5 : article 138. 21. Duckworth, W. C., R. G. Bennett, et al., Insulin degradation: progress and potential, Endocr Rev, 1998. 19 (5): p. 608 - 624. 22. Duhlmeier, R., E. Deegen, et al., Glucose - dependent insulinotropic polypeptide (GIP) and the enteroinsular axis in equine (Equus caballus), Comp Biochem Physiol A Mol Integr Physiol, 2001. 129(2 - 3): p.563 - 575. 23. Durham, A. E., K. J. Hughes, et al., Type 2 d iabetes mellitus with pancreatic beta cell dysfunction in 3 horses confirme d with minimal model analysis, Equine Vet J , 2009. 41 (9): p. 924 - 929. 24. Dybdal, N. O., K. M. Hargreaves, et al., Diagnostic testing for pituitary pars intermedia dysfunction in horse s, J Am Vet Med Assoc, 1994. 204 (4): p. 627 - 632. 25. Dyer, J., E. Fernandez - C astano Merediz, et al., Molecular characteriz ation of carbohydrate digestion and absorpt ion in equine small intestine, Equine Vet J , 2002. 34 (4): p. 349 - 358. 58 26. Dyer, J., M. Al - Rammahi, et al., Adaptive response of equine intestinal Na+/glucose co - transporter (SGLT1) to an increase in dietary soluble carbohydrate, Pflugers Arch, 2009. 458 (2): p. 419 - 430. 27. Eiler , H., N. Frank, et al., Physiologic assessment of bloo d glucose homeostasis via combined intravenous glucose and insulin testing in horses, Am J Vet Res , 2005. 66 (9): p. 1598 - 1604. 28. Elzinga, S., Nielsen, B.D., et al. , Comparison of nutrient digestibility bet ween adult and aged h orses, J Equine Vet Sci , 2014. 34 (10): p. 1164 - 1169. 29. Ferrannini, E. and A. Mari, How t o measure insulin sensitivity, J Hypertens , 1998. 16 (7): p. 895 - 906. 30. Fink, R. I., P. Wallace, et al., Effects of aging on glucose - mediated glucose disposal and glucose transport, J Clin Invest, 1986. 77 (6): p. 2034 - 2041. 31. Firshman, A. M. and S. J. Valberg, Factors affecting clinical assessment of insulin sensitivity in hors es, Equine Vet J, 2007. 39 (6): p. 567 - 575. 32. Firshman , A. M., S. J. Valberg, et al., Serum creatine kinase response to exercise during dexamethasone - induced insulin resistance in Quarter Horses with po lysaccharide storage myopathy, Am J Vet Res , 2005. 66 (10): p. 1718 - 1723. 33. Forhead, A. J ., J. C. Ousey, et al., Postnatal insulin secretion and sensitivity after manipulation of fetal growth by embryo transfer in the horse, J Endocrinol , 2004. 181 (3): p. 459 - 467. 34. Fowden, A.L., R.S. Comline, et al., Insulin s ecretion and carbohydrate metabolism during pregnancy in the mare, Equine Vet J, 1984. 16 (4): p. 239 - 246. 35. Frank, N., R. J. Geor, et al., Equine metabolic syndrome, J Vet Intern Med, 2010. 24 (3): p. 467 - 475. 36. Gannon, M.C. , F.Q. Nuttall, et al., An increase in dietary protein improves to blood glucose response in persons with type 2 diabetes, Am J Clin Nutr, 2003. 78 (4): p.734 - 741. 37. Garc ia, M. C. and J. Beech, Equine intravenous glucose tolerance test: glucose and insulin responses of healthy horses fed grain or hay and of horses with pituitary adenoma, Am J Vet Res , 1986. 47 (3): p. 570 - 572. 38. Geor, R. and N. Frank, Metabolic syndrome - From human organ disease to laminar failure in equids, Vet Immunol Immunopathol , 2009. 129 ( 3 - 4): p. 151 - 154. 59 39. Geo r, R. J. and P. Harris, Dietary management of obesity and insulin resistance: countering risk for laminitis, Vet Clin North Am Equine Pract , 2009. 25 (1): p. 51 - 65. 40. George, L. A., W. B. Staniar, et al., Evaluation of the effects of pr egnancy on insulin sensitivity, insulin secretion, and glucose dynamics in Thoroughbred m ares, Am J Vet Res , 2011. 72 (5): p. 666 - 674. 41. Goedecke, J.H., N.S. Levitt, et al., Differential effects of abdominal adipose tissue distribution on insulin sensitivity in black and white South African women, Obesity (Silver Spring), 2009. 17 (8): p. 1506 - 1512. 42. Graham, B. P., D ental care in the older h orse, Vet Clin North Am Equine Pract , 2002. 18 (3): p. 509 - 522. 43. Gumbiner, B., K. S. Polonsky, et al., Effects of aging on insulin secretion, Diabetes, 1989. 38 (12): p. 1549 - 1556. 44. Gungo r, N., R. Saad, et al., Validation of surrogate estimates of insulin sensitivity and insulin secretio n in children and adolescents, J Pediatr , 2004. 144 (1): p. 47 - 55. 45. Harris, P., S. R. Bailey, et al., Countermeasures for pasture - associated laminitis in ponies and horses, J Nutr, 2006. 136 (7 Suppl): p. 2114S - 2121S. 46. Harris, P. and R. J. Geor, Primer on dietary carbohydrates and utility of the glyce mic index in equine nutrition, Vet Clin North Am Equine Pract , 2009. 25 (1): p. 23 - 27. 47. Henneke, D. R., G. D. Potter, et al., Relationship between condition score, physical measurements and body fat percentage in mares, Equine Vet J, 1983. 15 (4): p. 371 - 372. 48. Henriksen, E.J., Invited review:Effects of acute exercise and exercise training on insulin resistance, J Appl Physiol, 2002. 93 (2): p. 788 - 796. 49. Hoffman, R. M., R. C. Bo ston, et al., Obesity and diet affect glucose dynamics and insulin sensitivity in Thoroughbred geldings, J Anim Sci, 2003. 81 (9): p. 2333 - 2342. 50. Hoffman , R. M., D. S. Kronfeld, et al., Glucose clearance in grazing mares is affected by diet, pregnancy, and lactation, J Anim Sci , 2003. 81 (7): p. 1764 - 1771. 51. Hoffman , R. M., L. A. Lawrence, et al., Dietary carbohydrates and fat influence radiographic bone min eral content of growing foals, J Anim Sci , 1999. 77 (12): p. 3330 - 3338. 52. Holdsto ck, N. B., V. L. Allen, et al., Development of insulin and proinsulin secretion in newborn pony foals, J Endocrinol , 2004. 181 (3): p. 469 - 476 . 60 53. Jackson, R.A., Mechanisms of a ge - related glucose intolerance, Diabet Care , 1990. 13 (Suppl ement 2): p. 9 19. 54. Jacobs, K.A. and J.R. Bolton, Effect of diet on the oral glucose tolerance test in the horse, J Am Vet Med Assoc, 1982. 180 (8): p. 884 - 886. 55. Jeffcott, L. B ., J. R. Field, et al., Glucose tolerance and insulin sensitivity in ponies and Standardbred horses, Equine Vet J , 1986. 18 (2): p. 97 - 101. 56. Johnson, P. J., The equine metabolic syndrome peripheral Cushing's syndro me, Vet Clin North Am Equine Pract , 2002. 18 (2): p. 271 - 293. 57. Johnson, P. J., C. E. Wiedmeyer, et al., Laminitis and the equine metabolic syndrome , Vet Clin North Am E quine Pract , 2010. 26 (2): p. 239 - 255. 58. June, V., V. Soderholm, et al., Glucose tolerance in the horse, pony and donkey, Equine Vet Sci, 1992. 12 (2): p.103 - 105. 59. Kim, J.Y., L.A. Nolte, et al., High - fat diet - induced muscle insulin resistance: relationship to visceral fat mass, Am J Physiol Regul Integr Comp Physiol, 2000. 279 (6): p. R2057 - 2065. 60. Krezowski, P. A., F.Q Nuttall, et al., Insulin and glucose responses to various startch - containing foods in type II diabetic subjects, Diabetes Care, 1987, 10 (2): p. 205 - 212. 61. Kronfeld, D. S., Dietary fat affects heat production and other variables of equine performance, u nder hot and humid conditions, Equine Vet J Suppl 1996 . (22): p. 24 - 34. 62. Kronfel d, D. S., K. H. Treiber, et al., Comparison of nonspecific indications and quantitative methods for the assessment of insulin re sistance in horses and ponies, J Am Vet Med Assoc , 2005. 226 (5): p. 712 - 719. 63. Kronfeld , D. S., K. H. Treiber, et al., Metabolic syndrome in healthy ponies facilitates nutritional countermeasu res against pasture laminitis, J Nutr , 2006. 136 (7 Suppl): p. 2090S - 2093S. 64. Liburt, N.R., Fugaro, M.N., et al., The effect of age and exercise t raining on insulin sensitivity, fat and muscle tissue cytokine profiles and body composition of old and young Standardbred mares, Comp Exer Phys, 2012. 8 (3 - 4): p. 173 - 187. 65. L owder, M. Q. and P. O. Mueller, Denta l disease in geriatric horses, Vet Clin North Am Equine Pract , 1998. 14 (2): p. 365 - 380. 66. Lyon, C. J., R. E. L aw, et al., Mini review: adiposity, inflammation, and atherogenesis." Endocrinology , 2003. 144 (6): p. 2195 - 2200. 61 67. Madigan, J. E. and J. W. Evans, Insulin turnover and irreversible loss rate in ho rses, J Anim Sci, 1973. 36 (4): p. 730 - 733. 68. Mair, T. S., M. H. Hillyer, et al., Small intestinal malabsorption in the horse: an assessment of the specificity of t he oral glucose tolerance test, Equine Vet J ,1991. 23 (5): p. 344 - 346. 69. Malinowski, K., C. L. Betros, et al., Effect of training on age - related changes in plasma insulin and glucose, Equine Vet J Suppl, 2002. (34): p. 147 - 153. 70. Maresh, M., Diabetes in pregnancy, Curr Opin Obstet Gynecol , 2001. 13 (2): p. 103 - 107. 71. Ma suzaki, H., J. Paterson, et al., A transgenic model of visceral obesity and the metabolic syndrome, Science , 2001. 294 (5549): p. 2166 - 2170. 72. McCue, M.E., R. J. Geor, et al., Equine metabolic syndrome: a complex disease influenced b y genetics and the environment, Journal of Equine Veterinary Science , 2015, 35 (5): p. 367 - 375. 73. McEvoy, C.T., C.R. Cardwell, et al., A posteriori dietary pattern s are related to risk of type 2 diabetes: findings from a systematic review and meta - analysis, J of Academy of Nutr and Dietetics, 2014. 114 (11) : p. 1759 - 1775. 74. Menzies - Gow, N., Diabetes in the horse: a condition of increasing clinical awareness for differential diagnosi s and interpretation of tests, Equine Vet J , 2009. 41 (9): p. 841 - 843. 75. Menzies - Gow, N. J., L. M. Katz, et al., Epidemiological study of pasture - associated laminitis and concurrent risk factors in the South of England, Vet Rec, 2010. 167 (18): p. 690 - 694. 76. Miller, M. A., I. D. Pardo, et al., Correlation of pituitary histomorphometry with adrenocorticotrophic hormone response to domp eridone administration in the diagnosis of equine pituitary pars intermedia dysfunction, Vet Pathol, 2008. 45 (1): p. 26 - 38. 77. Milinovich, G. J ., D. J. Trott, et al., Changes in equine hindgut bacterial populations during ol igofructose - induced laminitis, Env iron Microbiol , 2006. 8 (5): p. 885 - 898. 78. Morley, J. E., The aging gut: physiology, Clin Geriatr Med , 2007. 23 (4): p. 757 - 767, v - vi. 79. Morino, K., K. F. Petersen, et al., Molecular mechanisms of insulin resistance in humans and their potential links with mitochondrial dysfunction, Diabetes , 2006. 55 Suppl 2 : p. S9 - S15. 80. Murphy, D . Reid, S.W.J., et al., The effect of age and diet on the oral glucose tolerance t est in ponies, Equine Vet J, 1997. 29 (6): p. 467 - 470. 62 81. Nauck, M. , F. Stockmann, et al., Reduced incretin effect in type 2 (n on - insulin - dependent) diabetes, Diabetologia , 1986 . 29 (1): 46 - 52. 82. Nakae, Y., H. Onouchi, et al., Effects of aging and gastric lipolysis on gastric emptying of lipid in liquid meal, J Gastroenterol , 1999. 34 (4): p. 445 - 449. 83. Nielsen, - Robison, C.I., et al., Glycemic and ins ulinemic responses are affected by age of hors e and method of feed processing, Journal of Equine Veterinary Science , 2010. 30 (10): p. 249 - 258. 84. Pagan, J. D., R. J. Geor, et al. , Effects of fat adaptation on glucose kinetics and substrate oxidation during low - intensity exercise, Equine Vet J Suppl , 2002. (34): p. 33 - 38. 85. Paneni, F., S. Costantino, et al., Insulin resistance, diabetes, and cardiovascular risk, Curr Atheroscler Rep, 2014. 16 (7): 419. 86. Parillo, M., A.A. Rivellese, et al., A high - monounsaturated - fat/low - carbohydrate diet improves peripheral insulin sensitivity in non - insulin - dependent diabetic patients, Metabolism, 1992. 41 (12): p. 1373 - 1378. 87. Pawlak , D. B., J. A. Kushner, et al., E ffects of dietary glycaemic index on adiposity, glucose homoeostasis, and plasma lipids in animals, Lancet , 2004. 364 (9436): p. 778 - 785. 88. Potau, N., L. Ibanez, et al., Pubertal changes in insulin secretion and peripheral insulin sensitivity, Horm Res , 1997. 48 (5): p. 219 - 226. 89. Powell, D.M., S.E. Reedy, et al., Effect of short - term exercise training on insulin sensitivity in obese and lean mares, Equine Vet J Suppl, 2002. (34): p. 81 - 84. 90. Pratt, S. E., R. J. Geor, et al., Effects of dietary energy sour ce and physical conditioning on insulin sensitivity and glucose tolerance in standardbred horses, Equine Vet J Suppl, 2006.(36): p. 579 - 584. 91. Ralston, S. L., Insulin and glu cose regulation, Vet Clin North Am Equine Pract , 2002. 18 (2): p. 295 - 304, vii. 92. Ral ston, S. L., Ev idence - based equine nutrition, Vet Clin North Am Equine Pract , 2007. 23 (2): p. 365 - 384. 93. Razzouk, L. and P. Muntner, Ethnic, gender, and age - related differences in patients with the metabolic syndrome, Curr Hypertens Rep, 2009. 11 (2): p. 127 - 132. 94. Roberts, M.C. and F.W. Hill, The oral glucose tolerance test in the horse, Equine Vet J, 1973. 5 (4): p. 171 - 173. 63 95. Roth, J., X. Qiang, et al., The obesity pandemic: where have we been and where are we going?, Obes Res , 2004. 12 Suppl 2 : p. 88S - 101 S. 96. Salles, N., Basic mechanisms of the aging gastrointestinal tract, Dig Dis , 2007. 25 (2): p. 112 - 117. 97. Schmidt, O., E. Deegen, et al., Effects of fat feeding and energy level on plasma metabolites and hormones in Shetland ponies, J Vet med A Physiol Pathol Clin Med, 2001. 48 (1): p. 39 - 49. 98. s disease, Vet Clin North Am Equine Pract, 2002. 18 (2): p. 237 - 270. 99. Schuver, A., Frank, N., et al., Assessment of insulin and glucose dynamics by using an oral sugar test in horses , Journal of Equine Veterinary Science , 2014. 34 (4): p. 465 - 470. 100. She n, S. W., G. M. Reaven, et al., Comparison of impedance to insulin - mediated glucose uptake in normal subjects and in subjects with latent diabetes, J Clin Invest , 1970. 49 (12): p. 2151 - 2160. 101. Shepherd, P. R. and B. B. Kahn, Glucose transporters and insulin action implications for insulin re sistance and diabetes mellitus, N Engl J Med , 1999. 341 (4): p. 248 - 257. 102. Shirazi - Beechey, S.P., Moran, A.W., et al., Influences of food cons tituents on gut health; glucose sensing and signaling ; regulation of intestinal glucose transport, Proceedings of the Nutrition Society , 2011. 70 : p. 185 193. 103. Smy th, G. B., D. W. Young, et al., Maturation of insulin and glucose responses to normal feeding in foals , Aust Vet J , 1997. 70 (4): p. 129 - 132. 104. Stuart, C. A., M.E. Howell, et a l., Insulin - stimulated translocation of glucose transporter (GLUT) 12 parallels that of GLUT4 in normal muscle, J Clin Endocrinol Metab, 2009. 94 (9): p. 3535 - 3542. 105. Suagee, J.K., B.A. Corl, et al., Effects of hyperinsulinemia on glucose and lipid transporter expression in insulin - sensitive horses, Domest Anim Endocrinol, 2011. 40( 3): p. 173 - 181. 106. Suagee, J.K., B.A. Corl, et al., A 90 - day adaptation to a high glycaemic diet alters postprandial lipid metabolism in non - obese horses without affecting peripheral insulin sensitivity, J Anim Physiol Anim Nutr (Berl), 2013. 97 (2): p. 245 - 254. 107. Suagee, J. K., B. A. Corl, et al., Relationships between body condition score and pl asma inflammatory cytokines, insulin, and lipids in a mixed population of light - breed horses, J Intern Med 2013. 27 (1): p. 157 - 163. 64 108. Tinworth, K.D., P.C. Wynn, et al., Evaluation of commercially available assays for the measurement of equine insulin, Domest Anim Endocrinol, 2011. 41 (2): p. 81 - 90. 109. Toth, F., N. Frank, et al., Optimization of the frequently sampled intravenous glucose tol erance test to reduce urinary glucose spilling in horses, Equine Vet J, 2009. 41 (9): p. 844 - 851. 110. Treiber, K. H., R. C. Boston, et al ., Insulin resistance and compensation in Thoroughbred weanlings adapted to high - glycemic meals , J Anim Sci , 2005b. 83 (10): p. 2357 - 2364. 111. T reiber, K. H., T. M. Hess, et al., Glucose dynamics during exercise: dietary energy sources affect minimal model parameters in trained Arabian geldi ngs during endurance exercise, Equine Vet J Suppl , 2006. (36): p. 631 - 636. 112. Treiber, K. H., D. S. Kronfeld, et al., Use of proxies and reference quintiles obtained from minimal model analysis for determination of insulin sensitivity and pancreatic beta - cell responsiveness in horses, Am J Vet Res , 2005. 66 (12): p. 2114 - 2121. 113. Treiber, K. H., D. S . Kronfeld, et al., Evaluation of genetic and metabolic predispositions and nutritional risk factors for pasture - associated laminitis in ponies , J Am Vet Med Assoc , 2006a. 228 (10): p. 1538 - 1545. 114. Treiber, K. H., D. S. Kronfeld, et al., Insulin resistance i n equids: possible role in l aminitis , J Nutr , 2006b. 136 (7 Suppl): p. 2094S - 2098S. 115. Tsch ritter, O., A. Fritsche, et al., Assessing the shape of the glucose curve during an oral glucose tolerance test, Diabetes Care , 2003. 26 (4): p. 1026 - 1033. 116. Vervuert, I, S. Klein, et al., Effects of feeding state on glycaemic and insulinaemic responses to a starchy meal in horses: a methodological approach, Animal, 2009, 3 (9): p. 1246 - 1253. 117. Vick, M. M., A. A. Adams, et al., Relationships among inflammatory cytokines, obesity, and insulin sensitivity in the horse, J Anim Sci, 2007. 85 (5): 1144 - 1155. 118. Volek, J.S. and R.D. Feinman, Carbohydrate restriction improves the features of Metabolic Syn drome. Metabolic Syndrome may be defined by the response to carbohydrate restriction, Nutr Metab (Lond), 2005 . 2 : p. 31. 119. Waller, A. P ., T. A. Burns, et al., Insulin resistance selectiv ely alters cell - surface glucose transporters but not their total protein expression in equine skeletal muscle, J Vet Intern Med, 2011. 25 (2): 315 - 321. 6 5 120. Waller, A.P., L. Huettner, et al., Novel link between inflammation and impaired glucose transport during equine insulin resistance, Vet I mmunol Immunopathol, 2012. 149 (3 - 4): P.208 - 215. 121. Waller, A.P., M. George, et al., GLUT 12 functions as a basal and insulin - independent glucose transporter in the heart, Biochem Biophys Acta, 2013. 1832 (1): p. 121 - 127. 122. Warren , L.K., Vineyard, K.R., Fat and fatty acids. In: Equine Applied and Clinical Nutrition. Elsevier, St. Louis, MO, USA, 2013: p. 136 - 155. 123. Watson , R.T., Pessin, J.E., Intracellular organization of insulin signaling and GLUT4 translocation. Recent Progress in Hormone Research , 2011. ( 56): p. 175 - 193. 124. Watts, K., Pasture management to minimize the risk of equine laminitis, Vet Clin North Am Equine Pract , 2010. 26 (2): p. 361 - 369. 125. Williams, C.A., D.S. Kronfeld, et al., Plasma glucose and insulin responses of Thoroughbred mares fed a meal high in starch and sugar or fat and fiber, J Anim Sci, 2001. 79 (8): p. 2196 - 2201. 126. Xi ao, Z. Q., L. Moragoda, et al., Aging is associated with incre ased proliferation and decreased a poptosis in the colonic mucosa, Mech Ageing Dev , 2001. 122 (15): p. 1849 - 1864. 127. Xu, H., G. T. Barnes, et al., Chronic inflammation in fat plays a crucial role in the development of obe sity - related insulin resistance, J Clin Invest , 2003. 112 (12): p. 1821 - 1830.