| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ANIMAL NUTRITION |
* Department of Large Animal Clinical Sciences, University of Tennessee, Knoxville 37996;
and
McCauley Bros. Inc., Versailles, KY 40383; and
and
Department of Clinical Studies, University of Pennsylvania, Kennett Square 19348
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Key Words: Corn Oil • Horses • Insulin Sensitivity • Lipoprotein • Rice Bran Oil
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
; Cicero and Gaddi, 2001; Berger et al., 2004). A hypolipidemic response characterized by decreases in serum concentrations of TG, TC, very low-density lipoprotein cholesterol (VLDL-C), and LDL-C has been associated with feeding RBO to rats (Rukmini and Raghuram, 1991; Sunitha et al., 1997). Rice bran also attenuates hyperglycemia in humans suffering from diabetes mellitus (Qureshi et al., 2002). Soluble components of rice bran lowered fasting glucose concentrations by 29 and 33%, respectively, in subjects with Type I and Type II diabetes mellitus (Qureshi et al., 2002).
Increasing dietary fat intake lowers plasma TG concentrations and increases plasma TC concentrations in horses (Geelen et al., 1999, 2001b; Frank et al., 2004). This blood lipid response to dietary fat has been detected after addition of soybean oil (Geelen et al., 1999, 2001b; Frank et al., 2004) or corn oil (Duren et al., 1987) to the diet, whereas plasma TG concentrations increase when medium-chain triglycerides from coconut and palm kernel oils are fed (Hallebeek and Beynen, 2001). To our knowledge, blood lipid responses to RBO have not been evaluated previously in horses.
The current study was undertaken to determine the effect of RBO on blood lipid concentrations, lipoprotein composition, and glucose dynamics in mares. We hypothesized that RBO would exert a hypolipidemic effect in horses, and that this effect would modify the blood lipid response to higher dietary fat intake. It was also hypothesized that RBO would alter glucose dynamics in mares.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
) ranged between 3.5 and 6, with a mean of 5.1 (SD = 0.4) at the beginning of the study. A 5-yr-old mare weighing 385 kg (BCS = 4.5) was withdrawn from the study after the first treatment period when acute neurological deficits were observed at the beginning of the first washout interval. This mare was replaced by a 5-yr-old mare weighing 379 kg (BCS = 3.5) that had been kept on the same pasture.
Beginning 10 d before each feeding period, mares were placed in stalls measuring 3.7 m x 3.7 m. Each mare was turned out into a round pen for 2 h on an alternating day schedule. Composition of feeds is outlined in Table 1 based on information provided by independent analysis of feeds by the Dairy One DHIA Forage Testing Laboratory (Ithaca, NY). Daily quantities of feed provided to mares were calculated to provide 1.5 times the DE requirement for maintenance according to NRC (1989) guidelines. Sweet feed (COOP 13% Supreme Performance; Tennessee Farmers Cooperative, Levergne, TN) was fed at 1% of BW (as-fed basis), and the balance to meet the DE target was met by feeding grass (primarily fescue) hay. Quantities of feed were calculated based on BW at the beginning of each feeding period after feeds were analyzed for DE content (Table 1). This basal diet was supplemented with 240 mL of water, corn oil (CO; Ventura Foods, Brea, CA), refined rice bran oil (RR; RITO, Inc., Stuttgart, Arkansas), or crude rice bran oil (CR; McCauley Bros., Inc., Versailles, KY). Inclusion of a dietary oil in the diet provided 240 g of additional fat and 2.16 Mcal (9 Mcal/kg) of DE (NRC, 1989). Oils were stored at room temperature and protected from light until fed. Water or oil was mixed with the daily allotment of grain and fed at 0700, and hay was divided equally between 0700 and 1600 feedings. Mares were monitored to ensure that they consumed all of the feed provided. Grain was introduced to the diet over the first 6 d of each feeding period by providing 25% of the daily amount of grain for 2 d, 50% for 2 d, and then 75% for 2 d.
|
It should be noted that this study was designed to incorporate two experiments. Mares were concurrently examined to determine whether dietary oils affect the development of gastric ulcers, and results of that study are published elsewhere (Frank et al., 2005). To induce gastric ulcers, mares were subjected to a 7-d intermittent feed deprivation period, during which they were held off feed for 24 h or fed as normal on alternating days. Oil or water was added to the feed or administered orally via dose syringe throughout this week. Therefore, mares received 240 mL of oil or water per day for 6 wk, and washout intervals each lasted 5 wk. Omeprazole (Gastrogard; Merial, Duluth, GA) was administered to mares once daily for 14 d before each feeding period, and endoscopic examinations of the stomach were performed at 0, 5, and 6 wk of each experimental period after collection of blood samples (0 and 5 wk).
Blood samples (50 mL) were collected via jugular venipuncture into EDTA-coated tubes at 0 and 5 wk of each period after depriving mares of feed overnight. Blood was collected between 0830 and 0930 and stored immediately in ice for transport to the laboratory. Combined intravenous glucose-insulin tolerance tests (IVGITT) were performed at 0 and 4 wk.
Isolation and Quantification of Lipoproteins
Plasma lipoprotein fractions were isolated using a sequential ultracentrifugation method first described by Havel et al. (1955). Briefly, low-speed centrifugation (1,000 x g) at 4°C for 20 min was used to separate plasma from chilled blood samples. Six-milliliter plasma samples were placed in a fixed-angle rotor (Beckman Instruments Inc, Fullerton, CA) for ultracentrifugation at 112,000 x g for 18 h at 10°C. A 1-mL fraction of density <1.006 g/mL plasma was isolated from each tube. This fraction is referred to as VLDL. Triglyceride, phospholipid (PL), free cholesterol (FC), and TC components of VLDL were measured using enzymatic colorimetric reagents (Wako Chemicals U.S.A., Richmond, VA) in an automated discrete analyzer (Cobas Mira, Roche Diagnostic Systems Inc., Somerville, NJ). Lipoprotein lipase, phospholipase D, and cholesterol oxidase, respectively, were the principal reagents of the TG, PL, and TC assays. Protein content of VLDL was analyzed using BSA standards and a spectrophotometer (UV-160; Shimadzu, Kyoto, Japan) in accordance with a modified (Markwell et al., 1978) Lowry et al. (1951) procedure. Plasma VLDL concentrations were calculated by summing concentrations of lipid (TG, TC, and PL) and protein components.
Low-density lipoprotein was isolated from the remaining plasma by ultracentrifugation under the same conditions as for VLDL after addition of potassium bromide to raise the plasma density to 1.063 g/mL. A 1.063 g/mL solution of KBr prepared in EDTA saline was added to each sample to raise the volume to 6 mL. After ultracentrifugation, a 1-mL fraction of density <1.063 g/mL plasma was isolated from each tube and dialyzed overnight in 0.5 M ammonium bicarbonate solution to remove dissolved salts. Postdialysis volumes were recorded for each sample, and duplicate samples were processed. This fraction is referred to as LDL. Compositional analysis of dialyzed LDL samples was performed as already described. Plasma LDL concentrations were adjusted to account for addition of KBr solution and volume expansion during dialysis. High-density lipoprotein (HDL) was isolated and quantified using the same methods as described for LDL, with the exception that the plasma density was raised to 1.21 g/mL. Equine VLDL, LDL, and HDL have previously established density limits of <1.006 g/mL, 1.019 to 1.063 g/mL, and 1.063 to 1.21 g/mL, respectively (Watson et al., 1991).
Analysis of Other Plasma Lipids
Concentrations of plasma TG and TC were measured using the enzymatic colorimetric reagents and instrumentation already described. Plasma NEFA concentrations were measured using an in vitro enzymatic, colorimetric test kit (Wako Chemicals U.S.A.) using acyl CoA synthetase, acyl CoA oxidase, and ascorbate oxidase reactions.
Measurement of Glucose Dynamics
On d 28, a 14-gauge polypropylene catheter was inserted into the left jugular vein between 0730 and 0830. At approximately 0930, a baseline blood sample was collected, and then a combined IVGITT was performed by infusing 0.15 g/kg of BW of a 50% dextrose solution (Dextrose 50% Injection; Abbott Laboratories, North Chicago, IL), immediately followed by 0.10 USP units/kg of BW of regular insulin (Humulin R; Eli Lilly and Co., Indianapolis, IN). Blood samples were collected from the catheter at 1, 5, 15, 25, 35, 45, 60, 75, 90, 105, 120, 135, and 150 min after infusion. At each time, 3 mL of blood was withdrawn from the infusion line and discarded, and then a 6-mL blood sample was collected. From the blood sample, blood was immediately transferred to a tube containing sodium fluoride and potassium oxalate, immediately cooled on ice and refrigerated. Additional blood was immediately transferred to a glass tube that was left at room temperature for 1 h to clot. Serum was subsequently harvested after low-speed centrifugation. Plasma and serum samples were stored at –4°C until further analysis. In an attempt to decrease agitation associated with the procedure, mares had ad libitum access to grass hay before and during the test.
Glucose concentrations were measured using a colorimetric assay (Glucose; Roche Diagnostic Systems, Inc.) on an automated discrete analyzer (Cobas Mira). Insulin concentrations were determined using a RIA (Coat-A-Count Insulin, Diagnostic Products Corp., Los Angeles, CA) previously validated for equine insulin (Freestone et al., 1991). Duplicate assays were performed on each sample and intraassay CV <5% were required for acceptance of results from both assays.
Insulin sensitivity (SI) and glucose effectiveness (SG) values were calculated for each combined IVGITT using MinMod Millenium (Version 5.10, R. Bergman, Los Angeles, CA) and Stata 8 (Stata Corp., College Station, TX) software according to the minimal model using previously described methods (Hoffman et al., 2003). Net insulin response (NIR) is presented here instead of the acute insulin response to glucose (AIRg) because exogenous insulin was injected at time = 0, instead of after a delay of 20 min in the insulin modified frequent-sampling i.v. glucose tolerance test (FSIGT) used by Hoffman et al. (2003). Net insulin response was calculated by dividing AIRg values obtained from the model by the plasma volume of each mare, which was assumed to equal 48 mL/kg of BW based on a published value (Carlson et al., 1979). Area under the curve (AUC) values were calculated using the trapezoidal method for plasma glucose and serum insulin concentrations.
Statistical Analyses
Week, period, and period x week effects were examined by repeated measures ANOVA using the MIXED procedure of SAS (SAS Inst., Inc., Cary, NC). Where significance was established, differences of least squares means among the groups were compared using Bonferroni tests. Differences in blood lipid concentrations, lipoprotein composition, and measures of glucose dynamics were then calculated by subtracting baseline values from wk 4 (glucose dynamics) or wk 5 (lipids and lipoproteins) values. Treatment, period, and treatment x period effects were examined by both a single degree of freedom comparison of added-oil treatments to the water treatment and a comparison of all four treatment groups by ANOVA, using the same procedure and software. Differences of least squares means were compared using Bonferroni tests. Correlations between wk 0 values for BW and other variables were examined by calculating Pearson correlation coefficients using SAS procedures. Correlations between changes in BW over 5 wk and alterations in other variables over the same time were also examined. An 
level of P < 0.05 was used for determination of significance.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
Plasma concentrations of NEFA, TG, and VLDL decreased (P < 0.001) by 55, 30, and 39%, respectively, and plasma HDL and TC concentrations increased (P < 0.001) by 15 and 12%, respectively, over 5 wk (Table 2). None of the variables examined was affected by the increase in wk = 0 BW as the study progressed, but changes in plasma NEFA (r = 0.58; P < 0.001) and TC (r = 0.44; P = 0.013) concentrations were positively correlated with weight gain over 5 wk.
|
). Total and free cholesterol content increased (P < 0.001) and percentages of PL and TG decreased (P < 0.001) within LDL over 5 wk (Table 3). Percentages of TG and TC decreased (P < 0.001) and percentage of protein increased (P < 0.001) in HDL over 5 wk (Table 3). Percentage of PL within VLDL was negatively correlated (r = –0.37; P = 0.037), and LDL phospholipid content was positively correlated (r = 0.51; P = 0.003) with weight gain over 5 wk.
|
) than when water was added.
|
). In contrast, mean percentage of FC within LDL increased(P = 0.008) by a greater magnitude when oil was supplemented instead of water (Table 5). A smaller (P = 0.088) decrease in percentage of PL within LDL over 5 wk was detected, and the TG content of HDL decreased (P = 0.053) further over 5 wk in mares that received oil instead of water (Table 5).
|
). Mean percentage of FC within VLDL decreased by 1.6% over 5 wk in mares that received CR, which was different (P < 0.05) to the zero change in percentage of FC detected when water was added(Table 5). Plasma glucose concentrations were less (P < 0.001) during combined IVGITT performed after 4 wk compared with 0 wk values, and mean glucose concentrations differed (P = 0.023 and 0.013) between wk 0 and 4 at 1 and 5 min after injection, respectively. Serum insulin concentrations did not differ between 0 and 4 wk, but SI decreased (P = 0.018) by 38% over 4 wk (Table 6). Glucose dynamics values were not affected by addition of oil to the basal diet.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
). It should be noted that experimental diets were not isocaloric in this study because additional fat and calories were consumed when mares received oil. Mean plasma NEFA concentrations decreased over 5 wk, which was attributed to slower synthesis of fatty acids by the liver and enhanced clearance of NEFA from the blood. Both of these mechanisms were affected in ponies that were fed a diet containing 12% of DM as fat (Geelen et al., 2001a). Hepatic acetyl-CoA carboxylase and fatty acid synthase activities both decreased by approximately 50% and skeletal muscle carnitine palmitoyltransferase-I (CPT-I) activity increased by 46% over 45 d on the high-fat diet (Geelen et al., 2001a). Beta oxidative pathways that depend on CPT-I are upregulated in response to an increase in dietary fatty acids (Orme et al., 1997). This finding is supported by the further decrease in plasma NEFA concentrations in response to supplementation with oil.
Baseline plasma VLDL and TG concentrations detected in this study compare favorably with those previously reported for horses (Watson et al., 1991; Frank et al., 2004). Alterations in plasma VLDL and TG concentrations mimic each other because greater than 50% of circulating TG is carried within VLDL in equine blood (Watson et al., 1991). In this study, plasma VLDL and TG concentrations decreased by 39 and 30%, respectively, over 5 wk. These decreases may be explained by slower VLDL production by the liver, increased lipase activity, or accelerated receptor-mediated clearance of lipoproteins from circulation. Results of previous studies suggest that TG carried within VLDL is hydrolyzed more rapidly as tissues adapt to using more fatty acids for energy production (Geelen et al., 2001a). Greater plasma lipoprotein lipase (LPL) and hepatic lipase activities have been detected in mares fed diets containing 3.2 or 7.3% fat (Frank et al., 2004), and LPL activity has been positively correlated with dietary fat intake in horses and ponies (Geelen et al., 2001b). In the current study, oil supplementation lowered plasma TG concentrations even further, which is consistent with the previously reported (Geelen et al., 2001b) inverse relationship between fat consumption and plasma TG concentrations in horses. Plasma TG concentrations were least in the CR group, but they did not differ from the CO or RR group, which was surprising because CR contains gamma oryzanol and other plant sterols that are not found within the other oils. Lower plasma TG concentrations have been associated with RBO consumption in humans (Rajnarayana et al., 2001) and rats (Sunitha et al., 1997), but explanations for this finding have not been provided.
Plasma TC concentrations increased (P < 0.001) in mares across the experimental period and addition of oil to the basal diet further raised (P = 0.072) the concentrations of this lipid. Higher plasma HDL concentrations also were detected at the end of the 5-wk experimental period, which likely contributed to the increase in plasma TC concentration. Watson et al. (1993b) reported that 61% of circulating cholesterol is carried by HDL in equine blood, whereas only 29% is found in LDL. Plasma HDL concentrations increased over 5 wk in this study, but LDL concentrations remained unchanged. Greater TC and HDL-C concentrations have previously been detected in horses fed high-fat diets and were attributed to enhanced transfer of cholesterol from VLDL to HDL during lipolysis (Geelen et al., 2001b).
Very low-density lipoprotein lipid content decreased when mares were placed on experimental diets. Smaller, denser VLDL may have been secreted by the liver in response to the diets fed, but it is more likely that additional lipid was removed from circulating lipoproteins by lipases. Higher lipase activities have been detected in horses (Geelen et al., 1999, 2001b; Frank et al., 2004) and ponies (Schmidt et al., 2001) fed high-fat diets. In humans, LPL hydrolyzes TG carried by VLDL, which stimulates the transfer of FC, PL, and apolipoprotein A-I to HDL (Watson et al., 1993a). An inverse relationship exists between plasma VLDL and HDL concentrations in humans because lipases catalyze the hydrolysis of TG and transfer of FC and PL from VLDL to HDL (Watson et al., 1993a). This inverse relationship can be recognized in the results presented here, and has been previously detected in horses that are fed high-fat diets (Geelen et al., 1999, 2001b; Frank et al., 2004). Alterations in VLDL protein content were attributed to a relative decrease in lipid content, but additional apolipoproteins may have been acquired by VLDL.
The composition of LDL and HDL from horses fed high-fat diets has not been evaluated previously. Low-density lipoproteins are generated from VLDL in circulation as TG and are hydrolyzed by lipases located on endothelial surfaces. In this study, LDL contained less TG and PL, and more cholesterol after 5 wk. Low-density lipoproteins may have contained less TG because they originated from lipid-depleted VLDL, or because more TG was exchanged for cholesterol as LDL and HDL particles interacted in circulation (Watson et al., 1993a). Although cholesterol ester transfer protein is not active in equine blood (Guyard-Dangremont et al., 1998), cholesterol is transferred between HDL and LDL via an alternative mechanism, and this mechanism is stimulated by high-fat diets (Geelen et al., 2001c). This association is supported by the inverse relationship between LDL cholesterol content and plasma HDL concentrations detected in the current study.
High-density lipoprotein cholesterol content decreased, and the percentage of protein within HDL increased over 5 wk, despite an overall increase in mean plasma HDL concentration. Because the percentage of FC within HDL remained the same, the decrease in percentage of TC was attributed to a decrease in cholesterol ester content. Interactions with LDL particles and the resulting exchange of cholesterol for TG may have contributed to this effect. Although this theory does not at first seem to be supported by the decrease in percentage of TG observed, plasma HDL concentrations increased over 5 wk, so the total amount of TG carried within HDL increased overall. Percentage of protein within HDL also increased over 5 wk in this study. This finding may be attributed to the relative decrease in lipid content, or it may be due to HDL particles acquiring additional apolipoproteins. Apolipoprotein A-I is the most abundant protein within HDL, but apolipoprotein A-II, apolipoprotein C, and apolipoprotein E are also present in smaller quantities (Watson et al., 1993a).
Percentage of FC and TC within VLDL decreased over time when the basal diet was supplemented with oil. This finding is consistent with the stimulatory effect of dietary fat on plasma lipase activities (Geelen et al., 2001b). Free cholesterol is transferred to HDL after lipases hydrolyze TG carried by VLDL (Geelen et al., 2001b). Our results suggest that lipase activities were greater in mares that received oil because plasma TG concentrations were less when oil was supplemented instead of water.
Potential confounding factors including weight gain and variation in feed composition may have affected the results of this study. Although period effects were detected for some variables, consistent patterns were not identified with respect to weight gain or alterations in basal diet composition. Mares gained weight during the 5-wk experimental periods and overall across the 39-wk study. This alteration was attributed to feeding diets that contained 1.5 times the recommended DE requirement for maintenance (NRC, 1989), with additional calories provided by oil during three of the four periods. Mares were fed more calories than required for maintenance in an attempt to induce gastric ulcers by feeding a large quantity of sweet feed each day as a single meal (Frank et al., 2005). Week 0 plasma lipid concentrations were not affected by mares gaining weight as the study progressed, but the amount of weight gained over 5 wk affected plasma NEFA and TC concentrations. Plasma NEFA concentrations rose in response to weight gain, but decreased when oil was added to the diet. In contrast, both weight gain and oil supplementation raised plasma TC concentrations over time.
Selection of a sweet feed as a component of the basal diet introduced greater variability than anticipated because both the fat and DE content of this feed varied considerably between periods. More importantly, subsequent questioning of the manufacturer revealed that this feed contained added dietary oil. An equal mixture of soybean oil and corn oil had been added in varying quantities to maintain the targeted 5% minimum crude fat content listed on the label. Unfortunately, it was not possible to ascertain the exact amounts of oils added. It must therefore be assumed that the amount and type of fat in the basal diet varied across the different periods. Because CO was already present in the basal diet, supplementation with CO may have had additive effects on blood lipid concentrations, particularly when the fat content of the basal diet was greater during Period 3 (Table 1). Results of statistical analyses do not support this theory because responses to CO did not vary according to the fat content of the basal diet, but it is conceivable that the effects of RBO on blood lipid variables were affected by the presence of other oils in the basal diet. A feed with a more consistent nutrient content such as oats should be selected in the future and feed from a single batch should be fed across the entire study period.
Two experimental studies were combined here with the aim of concurrently evaluating the effects of dietary oils on gastric ulcer development. Sweet feed was selected as a component of the basal diet because these feeds are thought to raise gastric VFA concentrations and decrease the gastric pH, which promotes the formation of gastric ulcers (Nadeau et al., 2000, 2003). Endoscopic examinations of the stomach were performed at 0, 5, and 6 wk of each experimental period and mares underwent 7 d of intermittent feed deprivation between wk 5 and 6. It is unlikely, however, that these events affected blood lipid or glucose dynamics variables because wk 0 and 5 blood samples were collected before endoscopic examinations were performed, and the week of intermittent feed deprivation was followed by a 5-wk washout period.
Supplementation with oil lowered plasma TC concentrations over 5 wk, but CR did not elicit any specific cholesterol-lowering properties. This result was unexpected because rice bran and its oil markedly lower plasma TC and LDL-C concentrations in rats (Rukmini and Raghuram, 1991; Sunitha et al., 1997) and humans (Wilson et al., 2000; Cicero and Gaddi, 2001; Berger et al., 2004). Components of RBO include fatty acids, triterpene alcohols, phytosterols, tocotrienols, and -tocopherol (Cicero and Gaddi, 2001). Of these components, phytosterols including gamma oryzanol are thought to be responsible for changes in blood cholesterol concentrations (Vissers et al., 2000). Plasma TC and LDL-C concentrations decreased by 7.35 and 7.73 mg/dL, respectively, after healthy normolipemic human subjects consumed 2.1 g of plant sterols daily from RBO for 3 wk (Vissers et al., 2000). Both crude and refined forms of RBO were evaluated in this study to assess the importance of plant sterols because it was assumed that plant sterol concentrations would be higher in CR. Surprisingly, cholesterol responses to RR and CR did not differ from those induced by CO. This result may be explained by the absence of cholesterol in the equine diet. It has been proposed that plant sterols decrease blood cholesterol concentrations by inhibiting the absorption of dietary cholesterol and increasing fecal excretion of bile acids (Rukmini and Raghuram, 1991). Fecal excretion of neutral sterols and bile acids increased in rats fed RBO at a 10% level for 8 wk (Sharma and Rukmini, 1986). Alternatively, plant sterol concentrations may have been too low in the oils evaluated in the current study, or the volume of oil administered may have been insufficient to affect blood lipid concentrations. It also is possible that plant sterols degraded in stored oils as a result of oxidative damage; however, this would have been unlikely because the CR used in this study contained -tocopherol (440 IU/kg of oil) and was packaged in sealed 3.8-L bottles. Unfortunately, plant sterol concentrations were not measured in this study, but oil should be analyzed in future studies to determine the effects of storage on RBO composition.
A combined IVGITT was used in this study, but the simplicity of this test may have limited application of the minimal model. After this study had started, Hoffman et al. (2003) reported the use of an FSIGT to evaluate glucose dynamics in horses. Compared with the FSIGT, the combined IVGITT has two major limitations. First, only 14 blood samples are collected during the combined IVGITT compared with 30 samples in the FSIGT. Data generated from this larger number of samples are likely to better represent glucose metabolism. Second, glucose and insulin were infused at the same time in the combined IVGITT, which limits evaluation of the pancreatic response to exogenous glucose. In contrast, the insulin-modified FSIGT includes a 20-min interval between glucose and insulin infusions so that the pancreatic response may be quantified (Hoffman et al., 2003). It is for this reason that values for NIR are presented here instead of the AIRg because serum insulin concentrations reflect contributions from both endogenous and exogenous sources.
Values for SI and SG detected in this study were greater than those reported by Hoffman et al. (2003), but testing methods differed between studies. In our study, SI, SG, and NIR values varied considerably between and within mares, which hindered comparisons between groups. For instance, when glucose dynamics values from the beginning of each feeding period were compared, marked variation was detected within the same mare, even though grass hay was fed exclusively on each occasion. This variability was unlikely to have resulted from differences in hay composition alone (Table 1). Variability may have resulted from limitations of the combined IVGITT with respect to the minimal model or occurred because of other factors including stress. Blood concentrations of catecholamines and glucocorticoids were not measured in our study, but it has been previously established that these stress hormones decrease insulin sensitivity and increase blood glucose concentrations in horses (Sticker et al., 1995; Cartmill et al., 2003).
Lower plasma glucose concentrations were detected during combined IVGITT at 4 wk, and these concentrations differed (P = 0.023 and 0.013) at 1 and 5 min after injection, respectively; however, mean AUC values for glucose and insulin did not differ between testing times or diets, and only SI changed over time. Insulin sensitivity was less after 4 wk on experimental diets, but inclusion of oil in the diet did not alter the response. Insulin sensitivity may therefore have decreased in response to the greater fat content of experimental diets, or because more sugar and starch were provided in the sweet feed. In previous studies, ponies that were fed a diet containing 20% of DE as fat exhibited decreased glucose tolerance (Schmidt et al., 2001), and Thoroughbred geldings fed a diet rich in sugar and starch instead of fat and fiber had lower SI values (Hoffman et al., 2003). Glycemic responses to RBO did not differ from those induced by water or CO, so our hypothesis was rejected. Rice bran solubles that alter glucose dynamics in diabetic humans (Qureshi et al., 2002) may contain components that are not present in rice bran oils.
We conclude that 1) concentrations and composition of blood lipoproteins were altered by switching mares from a grass hay-only diet to experimental diets composed primarily of sweet feed and grass hay; 2) addition of oil to the basal diet instead of water decreased (P = 0.002 and 0.020, respectively) plasma NEFA and TG concentrations further and increased (P = 0.072) plasma TC concentrations by a greater magnitude over 5 wk; 3) only CR elicited responses that differed (P < 0.05) from those of water when all four treatment groups were included in the analysis; and 4) supplementation with oil did not affect glucose dynamics as determined by study procedures.
|
Implications |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
|
Footnotes |
|---|
2 Correspondence—phone: 865-974-5701; fax: 865-974-5773; e-mail: nfrank{at}utk.edu.
Received for publication October 22, 2004. Accepted for publication July 20, 2005.
|
Literature Cited |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionImplicationsLiterature Cited |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
