218 norm. Metab. Res. 9.(1977) 218·222
© Georg Thieme Verlag Stuttgart
The Effect of Diet Fat on Rat Adipocyte Glucose Transport C. Ip*, Helen M. Tepperman, Janet De Witt and J. Tepperman Department of Pharmacology, State University of New York, Upstate Medical Center, Syracuse, New Y ork, USA
Rats were fed either a high fat diet (67% of calories as Iard) or high glucose diet (67% of calories as gluoose) for 7-8 days. Basal and insulin stimulated net uptake of D glucose (D-L) and 2 deoxy D glucose uptake by free fat cells of fat rats were depressed. Net transport of D glucose (D-L) by purified adipocyte plasma membranes of fat red rats was also diminished. Incubation of fat cells from glucose fed rats with insulin before homogenization for membrane preparation increased net D gluoose transport by subsequently purified membranes in two experiments to a greater extent than in similar preparations from rat fed rats. These exper· iments suggest that fat feeding modifies th plasma membranes of fat cells so that both gluoose transport and the stimulatory effect of insulin on the process are decreased. Key-Words: Fat Diet - Adipocytes - PlaS1TllJ Membrane Glucose Transport - Insulin
Introduction
Recent reports from this laboratory have shown that feeding rats a high fat diet produces adaptive changes in certain plasma membrane functions of their isolated adipocytes. Hormone stimulated adenylate cyclase activity of fat cell ghosts. (Gorman, Tepperman anti Tepperman 1972) and insulin binding capacity of fat cells (Ip, Tepperman. Holohan and Tepperman 1976) were decreased in rats fed high fat diets compared to controls fed chow or high glucose diets. The reduced fat cell insulin bin ding correlated with a decreased in vitro response to insulin as measured by its effect on glucose oxidation, a confirmation of previous reports of fat diet induced insulin resistance (Blazquez and Quijada 1968, Malaisse. Lemonnier, Malaisse-Lagae and Mandelbaum 1969. Smith Kralanti and Bjorntorp 1974). In those experiments, glucose oxidation by fat cells was measured in the presence of low glucose concentrations, a procedure which is believed to provide an indirect measure of glucose transport (Crottord and Renold 1965). In several experiments with rats fed isocaloric high fat or high glucose diets, fat cells isolated from the former oxidized less glucose than did those from the latter either in the presence or absence of insulin (Ip, Tepperman, Holohan and *Present address: Department of Breast Surgery, Cell and Virus Bldg., Roswell Park Memorial Institute Buffalo, New York Received: 29 Apr. 1976
Aeeepted: 2S Oet. 1976
Tepperman 1976). It seemed likely, therefore, that the plasma membrane glucose transport capacity , in addition to its response to insulin, was decreased in response to fat feeding. To test this hypothesis, two recently published procedures for measurement of adipocyte glucose transport were used to study the effect of diet fat on this additional plasma membrane function. The method of Livingston and Lockwood (1974) was used to measure fat cell basal and insulin stimulated uptake of metabolizable and non-metabolizable sugars during short incubations as one index of glucose transport. In addition, the method of Carter, Avruch anti Martin (1972) was used to measure glucose transport by a plasma membrane fraction purified by the procedure of McKeel and larett (1970). The results provide further evidence that fat feeding decreases fat cell glucose transport activity. Materials and Methods Animals and Diets. Young adult male rats (100-110 gm initial wt. from Charles River Laboratories, Wilmington, Mass., CD strain) were employed in all experiments. They were fed one of two synthetic diets ad libitum before each experiment for the lengths of time indicated in the results seetion. Both diets included 33% of calories as casin with 67% as lard (F) or gluoose (C). The diets were supplemented with 5% salt mixture and 4% vitamin mix from Nutritional Biochemicals.
Downloaded by: East Carolina University. Copyrighted material.
Summary
Preparation o[ isolated [ar cel1s and meallUrement o[ cell number. Rats were killed by decapitation and Cat ceDs were isolated from the epididymal fat pads by collage na se digestion acoording to the method of Rodbell (1964). Fat cell number was estimated by a modification of the method of DiGirolamo, Mendlinger and Fergit (1971). The diameters of 150 cells from each preparation were measured with a micrometer eyepice and the mean cell volume calculated. Triglyceride oontent oC an aliquot of the fat cell suspendion was measured by the method of Stern and Shapfro (1953) with triolein as standard. The number of cells per ml. of suspension was then calculated assuming a specific gravity of 0.915.
Fat cell glucose uptake. Fat cell gluoose uptake was meas.. ured by the method of Livingston and Lockwood (1974). Six experiments were done for each diet group. Cells for each experiment were prepared from the fat pads of five or six rats. The cells were preincubated for 45 min at 37 0 in Krebs.."enseleit bicarbonate buffer with 1% of bovine serum albumin (Armour) in an atomosphere of 95% 02, 5% C02 in the presence and absence of insulin (lOOIlV/ml). The assays were than carried out at 22 0 C by adding 0.1 ml ofcell suspension to 0.2 ml of 0-[2- 3" [-gluoose and L-[_14CJgluoose in bicarbonate albumin buffer. Final ooncentrations were 0.33 mM for each sugar. It was shown. as expectcd,
Diet Fan and Glucose Transport in Rat Adipocyte
AU results were reported as means ± S.E. Differences with a p value of .05 or less were considered statisticaUy significant.
Results
There was no significant difference in body weight between the rats fed the fat or the carbohydrate diet. The cell size difference were statistically significant in the 2 deoxyglucose experiment (mean volurne for F cells = 102.3 ± 2.5 and for C = 76.9 ± 1.3 picoliters) but not in the experiments in which D and L glucose uptake in fat cells were measured (mean volume for F cells = 77.3 ± .64 and for C d 72.3 ± 2.17). We have carried out many experiments on rats fed high fat diets. There is usually a tendency for the fat fed rats to gain slightly more weight and to have larger fat pads and fat cells than those fed chow or a high glucose diet. The differences someF'ractiofUltion o{ plo8mll membrane from adipocyte', Fat ceU times reach statistical significance, but on other ocplasma membranes were isolated from 16-20 rats for each casions may not. The explanation for this variability preparation as described by Jorett (1974). is not clear, but it does not appear to influence the In experiments designed to investigate the effect of insulin differences in fat cell membrane properties which on glucose transport by plasma membrane preparations, the foUowing procedure was employed. A single pool of isolated are regularly observed in rats fed lard diets. adipocytes flOm the same batch of animals was divided equally into two portions: one portion was incubated with I mUlml of bovine insulin (Sigma) at 37 0 for 15 minutes whlle the other portion served as the contro!. 80th ceU preparations were then washed and the plasma membranes were isolated. A'$Qy tor uptake o{ Iobeled sugars by plasma membrane preparation. Duplicate tubes were set up containing 300 "I of buffer (0.25 M sucrose, 10 mM Tris, pH 7.4). Reaction was initiated by adding 100 "I of a mixture of an equimolar concentration of D- and L"ßlucose containing D-[2- 3 HIglucose (Amersham/Searle) and L-[_14C) glucose (New England Nuclear) as tracers, so that the final concentration of each sugar in the reaction mixutre was 5 mM. Incubations were carried out at 24 0 in a metabolic shaker. (AU uptake experiments were done at room temperature which happened to be slightly higher for the experiments on isolated membranes than for those on whole ceUs.) At various intervals of time, 50 "I aliquots were removed from the reaction mixture and filtered over pre-chiUed MiUipore HAWP filters (pore size 0.45 ,,) by means of a Millipore sampling manifold. The filters were washed in a dropwise fashion with ice cold Krebs-Ringer bicarbonate buffer as described by Carter et 01. (1972). The filters were than transferred d irectly to counting vials, dissolved in 10 ml of scintillation fluid and the radioactivity determined in an Isocap Liquid Scintillation Counter by the double Iabeling technique. The scintülation fluid was composed of 3: I (vIv) of toluence and methylceUusolve with 0.3% of PPO and 0.01% of POPOP. Protein and enzyme measurements. The protein content of each subcellular fraction was deterrnined by the method of Lowry. ROlebrough. Fa" and Randall (1951) uSing bovine serum albumin as the standard. The marker enzymes for plasma fTIembrane, mitochondrial and microsomal fractions were 5 -nuc1eotidase, succinate dehydrogenase and NADH oxidase respectively. 80th 5'-nu~ leotidase and NADH oxidase were measured by the methods described by Avruch and Wallach (1971). The succinate dependent reduction of indophenol to leucoindophenol was followed spectrophotometrically for the assay of succinate dehydrogenase (Bachman. AI/mann and Green 1966).
Diet effect on glucose uptake by lat cells. A preliminary experiment showed that L-glucose uptake by C or F cells reached its maximum value in 15 seconds and was not influenced by insulin. D-glucose radioactivity, on the other hand, continued to increase for 120 seconds. The net glucose incorporated (D-L) was less in the presence and absence of insulin when cells from fat fed rats were compared with those from glucose fed rats. The diet had been fed for 6 to 13 days. The experiment was repeated with rats fed the diets for 7 or 8 days with the results shown in Fig. 1. Again there was less net glucose uptake into adipocytes from rat fed than those from glucose fed rats both in the presence and in the absence of insulin and insulin produced a smaller in creme nt of uptake. It seemed possible that even under these experimental conditions the intracellular metabolism of glucose could have influenced these results to some extent. The uptake of 2-deoxyglucose by cells from rats fed the two diets was therefore measured. In this case only glucose transport and phosphorylation could contribute to the net uptake. The results (Fig. 2) again show decreased uptake by F adipocytes compared to C and decreased insulin response. The uptake of 3-O-methyl glucose, which is not phosphorylated, would give a clear measure of glucose transport alone. Attempts to study this process by the technique reported by Livingston and Lockwood (1974) were unsuccessful because of the extremely rapid attainment of equilibrium. In order to obtain a measure of glucose transport uncomplicated by possible intracellular metabolism, purified adipocytes plasma membranes were prepared from rats
Downloaded by: East Carolina University. Copyrighted material.
that the presence of L glucose did not influence the amount of D glucose taken up by the cells. The amount of L-glucose taken up was considered to be a measure of glucose trapped in extraceUular fluid plus whatever glucose entered the ceUs by nonspecific diffusion. This value was substracted from the D"ßlucose uptake to give a measure of specific D"ßlucose transport. Uptake of 2-deoxyglucose was measured simllarly with D-(I-I4c1-2 deoxyglucose at a final concentration of 0.33 mM. In this case only total D-[1-14CJ-2 deoxyglucose uptake was measured. The uptake of L-[ t-t4c1glucose was the same when fat cells from rats on the two diets were compared. Therefore differences in total 2-deoxyglucose uptake reflect differences in specific uptake of the sugar. Five experiments, each with cells from five or six rats, were carried out for each diet group. The absolute amount of D glucose or D deoxyglucose taken up varied considerably between different groups of rats studies at different times. The reason for this is unknown, but similar variabllity between groups of rats purchased at different times has been found when other properties of fat cells were measured.
219
C. Ip, Helen M. Tepperman, Janet De Witt, J. Tepperman
220
2.0
"-I ~
~
a..
/!
:::l.!! 1.5
"-I-u
© Georg Thieme Verlag Stuttgart
The Effect of Diet Fat on Rat Adipocyte Glucose Transport C. Ip*, Helen M. Tepperman, Janet De Witt and J. Tepperman Department of Pharmacology, State University of New York, Upstate Medical Center, Syracuse, New Y ork, USA
Rats were fed either a high fat diet (67% of calories as Iard) or high glucose diet (67% of calories as gluoose) for 7-8 days. Basal and insulin stimulated net uptake of D glucose (D-L) and 2 deoxy D glucose uptake by free fat cells of fat rats were depressed. Net transport of D glucose (D-L) by purified adipocyte plasma membranes of fat red rats was also diminished. Incubation of fat cells from glucose fed rats with insulin before homogenization for membrane preparation increased net D gluoose transport by subsequently purified membranes in two experiments to a greater extent than in similar preparations from rat fed rats. These exper· iments suggest that fat feeding modifies th plasma membranes of fat cells so that both gluoose transport and the stimulatory effect of insulin on the process are decreased. Key-Words: Fat Diet - Adipocytes - PlaS1TllJ Membrane Glucose Transport - Insulin
Introduction
Recent reports from this laboratory have shown that feeding rats a high fat diet produces adaptive changes in certain plasma membrane functions of their isolated adipocytes. Hormone stimulated adenylate cyclase activity of fat cell ghosts. (Gorman, Tepperman anti Tepperman 1972) and insulin binding capacity of fat cells (Ip, Tepperman. Holohan and Tepperman 1976) were decreased in rats fed high fat diets compared to controls fed chow or high glucose diets. The reduced fat cell insulin bin ding correlated with a decreased in vitro response to insulin as measured by its effect on glucose oxidation, a confirmation of previous reports of fat diet induced insulin resistance (Blazquez and Quijada 1968, Malaisse. Lemonnier, Malaisse-Lagae and Mandelbaum 1969. Smith Kralanti and Bjorntorp 1974). In those experiments, glucose oxidation by fat cells was measured in the presence of low glucose concentrations, a procedure which is believed to provide an indirect measure of glucose transport (Crottord and Renold 1965). In several experiments with rats fed isocaloric high fat or high glucose diets, fat cells isolated from the former oxidized less glucose than did those from the latter either in the presence or absence of insulin (Ip, Tepperman, Holohan and *Present address: Department of Breast Surgery, Cell and Virus Bldg., Roswell Park Memorial Institute Buffalo, New York Received: 29 Apr. 1976
Aeeepted: 2S Oet. 1976
Tepperman 1976). It seemed likely, therefore, that the plasma membrane glucose transport capacity , in addition to its response to insulin, was decreased in response to fat feeding. To test this hypothesis, two recently published procedures for measurement of adipocyte glucose transport were used to study the effect of diet fat on this additional plasma membrane function. The method of Livingston and Lockwood (1974) was used to measure fat cell basal and insulin stimulated uptake of metabolizable and non-metabolizable sugars during short incubations as one index of glucose transport. In addition, the method of Carter, Avruch anti Martin (1972) was used to measure glucose transport by a plasma membrane fraction purified by the procedure of McKeel and larett (1970). The results provide further evidence that fat feeding decreases fat cell glucose transport activity. Materials and Methods Animals and Diets. Young adult male rats (100-110 gm initial wt. from Charles River Laboratories, Wilmington, Mass., CD strain) were employed in all experiments. They were fed one of two synthetic diets ad libitum before each experiment for the lengths of time indicated in the results seetion. Both diets included 33% of calories as casin with 67% as lard (F) or gluoose (C). The diets were supplemented with 5% salt mixture and 4% vitamin mix from Nutritional Biochemicals.
Downloaded by: East Carolina University. Copyrighted material.
Summary
Preparation o[ isolated [ar cel1s and meallUrement o[ cell number. Rats were killed by decapitation and Cat ceDs were isolated from the epididymal fat pads by collage na se digestion acoording to the method of Rodbell (1964). Fat cell number was estimated by a modification of the method of DiGirolamo, Mendlinger and Fergit (1971). The diameters of 150 cells from each preparation were measured with a micrometer eyepice and the mean cell volume calculated. Triglyceride oontent oC an aliquot of the fat cell suspendion was measured by the method of Stern and Shapfro (1953) with triolein as standard. The number of cells per ml. of suspension was then calculated assuming a specific gravity of 0.915.
Fat cell glucose uptake. Fat cell gluoose uptake was meas.. ured by the method of Livingston and Lockwood (1974). Six experiments were done for each diet group. Cells for each experiment were prepared from the fat pads of five or six rats. The cells were preincubated for 45 min at 37 0 in Krebs.."enseleit bicarbonate buffer with 1% of bovine serum albumin (Armour) in an atomosphere of 95% 02, 5% C02 in the presence and absence of insulin (lOOIlV/ml). The assays were than carried out at 22 0 C by adding 0.1 ml ofcell suspension to 0.2 ml of 0-[2- 3" [-gluoose and L-[_14CJgluoose in bicarbonate albumin buffer. Final ooncentrations were 0.33 mM for each sugar. It was shown. as expectcd,
Diet Fan and Glucose Transport in Rat Adipocyte
AU results were reported as means ± S.E. Differences with a p value of .05 or less were considered statisticaUy significant.
Results
There was no significant difference in body weight between the rats fed the fat or the carbohydrate diet. The cell size difference were statistically significant in the 2 deoxyglucose experiment (mean volurne for F cells = 102.3 ± 2.5 and for C = 76.9 ± 1.3 picoliters) but not in the experiments in which D and L glucose uptake in fat cells were measured (mean volume for F cells = 77.3 ± .64 and for C d 72.3 ± 2.17). We have carried out many experiments on rats fed high fat diets. There is usually a tendency for the fat fed rats to gain slightly more weight and to have larger fat pads and fat cells than those fed chow or a high glucose diet. The differences someF'ractiofUltion o{ plo8mll membrane from adipocyte', Fat ceU times reach statistical significance, but on other ocplasma membranes were isolated from 16-20 rats for each casions may not. The explanation for this variability preparation as described by Jorett (1974). is not clear, but it does not appear to influence the In experiments designed to investigate the effect of insulin differences in fat cell membrane properties which on glucose transport by plasma membrane preparations, the foUowing procedure was employed. A single pool of isolated are regularly observed in rats fed lard diets. adipocytes flOm the same batch of animals was divided equally into two portions: one portion was incubated with I mUlml of bovine insulin (Sigma) at 37 0 for 15 minutes whlle the other portion served as the contro!. 80th ceU preparations were then washed and the plasma membranes were isolated. A'$Qy tor uptake o{ Iobeled sugars by plasma membrane preparation. Duplicate tubes were set up containing 300 "I of buffer (0.25 M sucrose, 10 mM Tris, pH 7.4). Reaction was initiated by adding 100 "I of a mixture of an equimolar concentration of D- and L"ßlucose containing D-[2- 3 HIglucose (Amersham/Searle) and L-[_14C) glucose (New England Nuclear) as tracers, so that the final concentration of each sugar in the reaction mixutre was 5 mM. Incubations were carried out at 24 0 in a metabolic shaker. (AU uptake experiments were done at room temperature which happened to be slightly higher for the experiments on isolated membranes than for those on whole ceUs.) At various intervals of time, 50 "I aliquots were removed from the reaction mixture and filtered over pre-chiUed MiUipore HAWP filters (pore size 0.45 ,,) by means of a Millipore sampling manifold. The filters were washed in a dropwise fashion with ice cold Krebs-Ringer bicarbonate buffer as described by Carter et 01. (1972). The filters were than transferred d irectly to counting vials, dissolved in 10 ml of scintillation fluid and the radioactivity determined in an Isocap Liquid Scintillation Counter by the double Iabeling technique. The scintülation fluid was composed of 3: I (vIv) of toluence and methylceUusolve with 0.3% of PPO and 0.01% of POPOP. Protein and enzyme measurements. The protein content of each subcellular fraction was deterrnined by the method of Lowry. ROlebrough. Fa" and Randall (1951) uSing bovine serum albumin as the standard. The marker enzymes for plasma fTIembrane, mitochondrial and microsomal fractions were 5 -nuc1eotidase, succinate dehydrogenase and NADH oxidase respectively. 80th 5'-nu~ leotidase and NADH oxidase were measured by the methods described by Avruch and Wallach (1971). The succinate dependent reduction of indophenol to leucoindophenol was followed spectrophotometrically for the assay of succinate dehydrogenase (Bachman. AI/mann and Green 1966).
Diet effect on glucose uptake by lat cells. A preliminary experiment showed that L-glucose uptake by C or F cells reached its maximum value in 15 seconds and was not influenced by insulin. D-glucose radioactivity, on the other hand, continued to increase for 120 seconds. The net glucose incorporated (D-L) was less in the presence and absence of insulin when cells from fat fed rats were compared with those from glucose fed rats. The diet had been fed for 6 to 13 days. The experiment was repeated with rats fed the diets for 7 or 8 days with the results shown in Fig. 1. Again there was less net glucose uptake into adipocytes from rat fed than those from glucose fed rats both in the presence and in the absence of insulin and insulin produced a smaller in creme nt of uptake. It seemed possible that even under these experimental conditions the intracellular metabolism of glucose could have influenced these results to some extent. The uptake of 2-deoxyglucose by cells from rats fed the two diets was therefore measured. In this case only glucose transport and phosphorylation could contribute to the net uptake. The results (Fig. 2) again show decreased uptake by F adipocytes compared to C and decreased insulin response. The uptake of 3-O-methyl glucose, which is not phosphorylated, would give a clear measure of glucose transport alone. Attempts to study this process by the technique reported by Livingston and Lockwood (1974) were unsuccessful because of the extremely rapid attainment of equilibrium. In order to obtain a measure of glucose transport uncomplicated by possible intracellular metabolism, purified adipocytes plasma membranes were prepared from rats
Downloaded by: East Carolina University. Copyrighted material.
that the presence of L glucose did not influence the amount of D glucose taken up by the cells. The amount of L-glucose taken up was considered to be a measure of glucose trapped in extraceUular fluid plus whatever glucose entered the ceUs by nonspecific diffusion. This value was substracted from the D"ßlucose uptake to give a measure of specific D"ßlucose transport. Uptake of 2-deoxyglucose was measured simllarly with D-(I-I4c1-2 deoxyglucose at a final concentration of 0.33 mM. In this case only total D-[1-14CJ-2 deoxyglucose uptake was measured. The uptake of L-[ t-t4c1glucose was the same when fat cells from rats on the two diets were compared. Therefore differences in total 2-deoxyglucose uptake reflect differences in specific uptake of the sugar. Five experiments, each with cells from five or six rats, were carried out for each diet group. The absolute amount of D glucose or D deoxyglucose taken up varied considerably between different groups of rats studies at different times. The reason for this is unknown, but similar variabllity between groups of rats purchased at different times has been found when other properties of fat cells were measured.
219
C. Ip, Helen M. Tepperman, Janet De Witt, J. Tepperman
220
2.0
"-I ~
~
a..
/!
:::l.!! 1.5
"-I-u
