Hoppe-Seyler's Z. Physiol. Chem. Bd. 356, S. 1055 - 1066, Juni 1975
The Influence of Insulin on Glucose and Fatty Acid Metabolism in the Isolated Perfused Rat Hind Quarter Felix Reimer, Georg Löffler, Gerd Hennig and Otto H. Wieland
(Received 6 February 1975) Dedicated to Prof. Dr. Günther Weitzel on the occasion of his 60th birthday. Summary: Glucose and fatty acid metabolism of resting skeletal muscle were studied by perfusion of the isolated rat hind leg with a hemoglobin-free medium. Tissue integrity was demonstrated by normal ATP, ADP and creatine phosphate levels, by a sufficient oxygen supply, and by a normal appearance of perfused muscle specimens under the electron microscope. The rates of glucose and fatty acid uptake, and of lactate, alanine, glycerol and fatty acid release were constant over a perfusion period of 60 min. Insulin (1 unit//) caused a more than threefold increase in glucose uptake, a stimulation of lactate production, and a 20% increase in the muscular glycogen levels. Fatty acid and alanine release were significantly diminished by insulin, but glycerol release did not change. The uptake of oleate by the rat hind leg was dependent on the medium concentration in a range of 0.7 - 1.9mM oleate, and was stimulated by insulin. Glucose uptake was not influenced by oleate, whether insulin was present or not. When the leg was perfused with [ l-14C]oleate, 75% of the incorporated
fatty acids were found in muscle lipids, 10% were oxidized to C0 2s and 5% were recovered in bone lipids. The absolute amount of oleate oxidation was not altered by insulin. In all experiments with and without glucose in the medium, 70-80% of the 14C label incorporated into muscle lipids was found in the triglyceride fraction. In the presence of glucose, insulin significantly increased the incorporation of [ l-14C]oleate.into muscle triglycerides, whereas no insulin effect, either on fatty acid uptake or on triglyceride formation, could be observed when glucose was omitted from the perfusate. The present results indicate that a "glucose-fatty acid cycle" as found in rat heart muscle does not operate in resting peripheral skeletal muscle tissue. They also demonstrate that the stimulating effect of insulin on muscular fatty acid, uptake and triglyceride synthesis is dependent on glucose supply. This finding can be interpreted as a stimulation of fatty acid esterification by ^«-glycerol 3-phosphate derived from an increased glucose turnover, which is in turn due to insulin.
Die Wirkung von Insulin auf den Glucose- und Fettsäurestoffwechsel Hinterkörpers der Ratte Zusammenfassung: Der Glucose- und Fettsäurestoffwechsel des nicht arbeitenden Skelettmuskels wurde am Modell des hämoglobinfrei perfun-
des isoliert perfundierten
dierten Hinterkörpers der Ratte untersucht. Gemessen am elektronenmikroskopischen Bild, am Sauerstoffverbrauch sowie an den Phospho-
Address: Prof. Dr. 0. H. Wieland, Klinisch-Chemisches Institut und Forschergruppe Diabetes, Krankenhaus München-Schwabing, D-8 München 40, Kölner Platz 1.
Brought to you by | Purdue University Lib Authenticated Download Date | 5/22/15 5:31 AM
1056
F. Reimer, G. Löffler, G. Hennig and O. H. Wieland
kreatin-, ATP- und ADP-Gehalten war das Präparat im Vergleich zum Zustand in vivo intakt. Die Geschwindigkeit von Glucose- und Fettsäureaufnahme sowie die Abgabe von Lactat, Alanin, Glycerin und Fettsäuren waren während 60minütiger Perfusion konstant. l Einh.// Insulin führte zu einer Zunahme der Glucoseaufnahme um mehr als das Sfache, zu einer Steigerung der Lactatbildung und einem Anstieg des Glykogengehaltes um 20%. Fettsäureund Alaninfreisetzung wurden durch Insulin signifikant vermindert, die Glycerinabgabe blieb jedoch unverändert. Die Aufnahmegeschwindigkeit von Ölsäure durch den perfundierten Muskel hing von der Fettsäurekonzentration ab und konnte durch Insulin stimuliert werden. Weder in Anwesenheit noch in Abwesenheit von Insulin wurde die Glucoseaufnahme durch Zugabe von Fettsäuren beeinflußt. 75% der aufgenommenen Fettsäuren fanden sich in den Muskellipiden, 10% wurden zu C02 oxi-
The role of elevated blood fatty acids and ketone bodies during starvation and diabetes in the regulation of muscle glucose metabolism was studied extensively by Rändle and coworkers^1"9^. In their experiments on rat heart and diaphragm muscle, it was found that glucose uptake, glycolysis, and pyruvate oxidation are inhibited when fatty acid oxidation increases due to a greater provision of fatty acids, either externally or from intracellular breakdown of glycerides. The theoretical concept of this reciprocal relationship ("glucose-fatty acid cycle") implies that when fatty acid oxidation is enhanced, the increased acetyl-CoA and citrate levels inhibit the pyruvate dehydrogenase and phosphofructokinase reactions, respectively. The inhibition of the latter leads to an accumulation of glucose 6-phosphate, which is known to be a potent inhibitor of hexokinase. As compared to heart muscle, inhibition of glucose metabolism by fatty acids in rat diaphragm was found to be rather small121, suggesting some fundamental difference in the metabolic properties of the two types of muscle tissues. Furthermore, using skeletal muscle preparations, other investigators were not able to demonstrate an effect of fatty acids or ketone
Bd. 356 (1975)
diert und 5 % in den Knochenmarklipiden wiedergefunden. Insulin war nicht imstande, die Geschwindigkeit der Ölsäureoxidation zu beeinflussen. 70 - 80% der in die Muskellipide aufgenommenen Fettsäuren konnten in der Triglyceridfraktion lokalisiert werden. In Anwesenheit von Glucose wurde der Einbau von Fettsäuren in diese Fraktion durch Insulin signifikant gesteigert, wohingegen bei glucosefreier Perfusion kein derartiger Effekt zu beobachten war. Die vorliegenden Befunde weisen daraufhin, daß die für das Rattenherz gefundene Hemmung des Glucoseumsatzes durch Fettsäuren („glucose-fatty acid cycle") für den nicht arbeitenden Skelettmuskel nicht zutrifft. Sie zeigen außerdem, daß Insulin in Abhängigkeit vom Glucoseangebot sowohl Fettsäureaufnahme als auch Triglyceridsynthese des Muskels stimuliert. Der zugrunde liegende Mechanismus dürfte in einer Stimulierung der Fettsäureveresterung mit tfi-Glycerin-3-phosphat liegen, das der Glykolyse entnommen wird.
bodies on glucose metabolisml10'15'. The present study was undertaken in order to re-investigate the relationship between glucose and fatty acid metabolism in a perfused skeletal muscle preparation.
Material and Methods Animals. Male Sprague-Dawley rats (Wiga, Sulzfeld, Germany) weighing 170 - 230 g were used. The animals were fed on a standard laboratory diet (Labortierfutter 57 Z, J. Zahn II, Hockenheim, Germany). Reagents. Nembutal (sodium pentobarbital) was supplied by the German Abbot GmbH, Ingelheim, Germany. Bovine serum albumin (Cohn fraction V) was obtained from Serva International, Heidelberg, Germany. It was dissolved in Krebs-Henseleit hydrogen carbonate bufferl 161 and dialysed twice against a tenfold volume of the same buffer for 24 h at 4 °C. Bovine insulin (25 units/mg) was kindly supplied by Farbwerke Hoechst, Germany. Oleic acid (distilled, puriss. > 98%) was obtained from Carl Roth OHG, Karlsruhe, Germany, and [l-14Cjoleic acid from the Radiochemical Centre, Amersham, Buckinghamshire, England. Lipid standards were obtained from Serva International, Heidelberg, Germany. Sodium pyruvate and enzymes were purchased from
Brought to you by | Purdue University Li Authenticated Download Date | 5/22/15 5:31 AM
Bd. 356 (1975)
Influence of Insulin
1057
as modified by Guder et alJ 26 l ATP and creatinephosphate were estimated according to Lamprecht and Trautscholdl27!, with the exception that the final glucose concentration in the assay was only 0.0IM. ADP was Perfusion medium. The standard perfusion medium was estimated by the method of Jaworek et aU 28 l, and composed as follows: Krebs-Henseleit hydrogen carbonate glycogen according to Keppler and Decker^29!. The Ο2 buff er gassed with carbogen (95% C>2 and 5% C02) and tension in the hemoglobin-free perfusate was detercontaining 4 g/100 m/ bovine serum albumin, lOmM mined polarographically with a Clark electrode (Radioglucose, O.lSmM pyruvate, and 1.5mM lactate. If oleic meter, Copenhaven, Denmark). The C»2 consumption acid was used, an albumin-oleate complex was prepared was calculated from the arterio-venous differences. At 3 as described by Garland et al.^ J and Guder and Wiethe end of the perfusion, muscle samples (30 - 100 mg) 17 land * ^ and mixed with the perfusion medium. The were obtained by freeze-clamping with a Wollenberger final albumin concentration of the perfusate was always clamp, weighed and homogenized with a high-speed kept at 4 g/100m/. tissue disintegrator (Ultra-Turrax 18/2, Jahnke & Kunkel, Stauffen i.B., Germany) in ice-cold 6% HC104, Surgical preparation. The preparation procedure was (100 mg tissue/m/ HC1 4 for the determination of glycarried out essentially as described by Ruderman et cogen, ATP, ADP, and creatine phosphate). al.l 181. As a modification, the major spinal vessels derivWhen perfused with [l-14C)oleic acid, CO2 was coling dorsally from the aorta and vena cava, and vessels lected in 25% (w/v) phenylethylamine in methanol, originating from the femoral arteries and veins below the inguinal ligament, which all supply the subcutaneous using a small plastic tube attached to the gas outlet of the oxygenator. The samples were placed in Bray's soluregion, were tied up. A ligature was also placed around tioni30! and counted for radioactivity. Tissue lipids were the liver radix. After the canulation of the aorta and extracted with chloroform/methanol (2:1, v/v) as outvena cava, the animal was killed by intrathoracal injeclined by Folch et a 31 J and Sperryl 32 L For radiotion of 40 mg Nembutal. In this way, peripheral vasactivity measurements, the lipids were separated on cular constrictions were largely avoided, and the final 20 χ 20 cm plates coated with silica gel 60 (Merck, perfusion pressure could already be adjusted after Darmstadt, Germany) using a solvent mixture of light 2 - 3 min. petroleum/ethyl ether/acetic acid (80:20:1). The lipidPerfusion system. The perfusion system, in general, was containing areas identified by standards were scraped the same as used for liver perfusions (for references off, transferred into the scintillation mixture (Scintigel, seel 19]) w}t^ t^e exception of a second pump behind Carl Roth OHG, Karlsruhe, Germany), and sonified in the oxygenator for maintaining an arterial pressure of order to secure homogeneous dispersion of the gel parti80 - 90 mm Hg and a constant blood flow of about cles. After the addition of water, stable and homogeneous 35 m//min. In order to get rid of erythrocytes, the first suspensions were obtained, the radioactivity of which was 50 - 60 m/ of the perfusate was discarded. The recircumeasured in a Packard model 3385 Tri-Carb liquid scintillating perfusion was then carried out with 150 ml melation spectrometer. dium. The hematocrit of the perfusate was below 0.1%. Calculations and statistical evaluation. Metabolic rates After 5 min of equilibration, the first medium sample were calculated from metabolite concentration differwas taken to determine the "zero" values. Beginning ences between the "zero" and 60 min values of perfrom "zero", the perfusion was continued for 60 min. fusion, and are given as nmol or μπιοί χ g~ 1 muscle Samples were collected at 15 min intervals. The effect tissue χ min"1. The total mass of the perfused muscle of insulin was tested by addition of a single dose of tissue was calculated using the formula of Ruderman 1 unit//. The temperature in the perfusion box was mainet alJ 18 l, which is based on rat total body weight. Sigtained at 37 °C by means of a thermoregulator. nificance was determined by the Wilcoxon rank sum Analytical methods. The perfusate samples were detest as modified by Mann and Whitney!331. Regression proteinized with ice-cold HQ 4 at a final concentration coefficients were calculated as described by Sachs^34!. of 4.5% (v/v). The extracts were neutralized with solid Electron microscopy. Tissue was excised from the adducKHC 3, and KC104 was removed by centrifugation. tor muscle both in nonperfused animals and after perGlucose was determined by the hexokinase/glucose-620 fusion. The muscle was cut into 0.5-mm cubes and then phosphate dehydrogenase reaction^ !. Lactate was fixed in 2.5 % glutaraldehyde buffered with 0.15M phosdetermined enzymatically as described by Gutmann and phate buffer, pH 7.2. Thereafter the tissue was dehyWahlefeldl21!, pyruvate as described by Czok and Lam22 drated with ethanol and propylene oxide, and then emprechti !, glycerol according to Eggstein and Kuhl35 mann^23!, and alanine according to Karl et alJ24l. Fatty bedded in Epon^ l Sections were cut on a LKB-microtomeNo.HI with a diamond knife, stained with 2% (w/v) acids were determined by the Duncombe procedure^ 25 ! Boehringer Mannheim, Germany. Sodium L-lactate was obtained from Serva International. All other reagents were obtained from Merck, Darmstadt, Germany.
Brought to you by | Purdue University Lib Authenticated Download Date | 5/22/15 5:31 AM
1058
F. Reimer, G. L ffler, G. Hennig and O. H. Wieland
uranyl acetate and lead citrate^ 36 ', and examined in a Siemens Elmiskop 101 electron microscope*.
Results An edema of the perfused hind legs was not observed when albumin concentration was 4 g/ 100 ml, and the perfusion pressure did not exeed 80 - 100 mm Hg. Under these conditions, ultrastructural examination did not reveal any differences in adductor muscle specimens taken in vivo and 60 min after perfusion. The polarographically determined venous oxygen tension was in the range of 120 - 180 mm Hg and remained constant over the 60 min perfusion time. The levels of ATP and creatine phosphate were the same in perfused and nonperfused muscle specimens (Table 1). In this context, it seems noteworthy that we could detect only small changes in the ATP and ADP levels in muscle during anoxia, at least for 60 min, whereas creatine phosphate was already decreased to 55% of the control after 5 min (Table 1). Obviously the resting skeletal muscle is able to maintain a high ATP/ADP ratio over long periods of anoxia at the expense of creatine phosphate. Therefore, the creatine phosphate level is a better control parameter for the evaluation of the metabolic state of the perfused muscle. This is further documented by the observation that during a 10 min perfusion where the medium was gassed with N 2 /C0 2 instead of
Bd. 356(1975)
O 2 /CO 2 , creatine phosphate decreased to 68% of the control, while the adenine nucleotides remained constant. The decrease in creatine phosphate was fully reversible after switching back to O 2 /C0 2 gassing.
Metabolism of the resting muscle. As the rates of glucose (Fig. 1) and fatty acid uptake, lactate production, alanine, glycerol, and fatty acid release (in the oleate-free experiments) were linear over a perfusion period of at least 60 min, it was possible to calculate the substrate turnover rates as nmol or μιηοΐ χ g~ * muscle tissue χ min~ *. Table 2 shows some basal metabolic rates and the effect of insulin (1 unit//) thereon. As may be seen, glucose uptake is stimulated by insulin more than threefold. In the controls, 50% of the extracted glucose was recovered as lactate, whereas in the presence of insulin, this proportion decreased to 27%. The lactate/pyruvate ratio was increased at the end of perfusion in both the control and insulin groups. Fatty acid and alanine release were significantly diminished by insulin, whereas glycerol release was not influenced. In accordance with the findings of Ruderman et aU 181, we observed no difference in the rate of basal lipolysis when the hind legs were skinned and all visible adipose tissue was removed before perfusion. The glycogen content remained constant during 60 min of perfusion and showed values comparable to the muscle in nonperfused animals (32.3 ± 4.95 μτηοΐ/g). Insulin caused a 20% increase in glycogen content. Due to the marked differences * The electron microscopic examination was kindly per- in the glycogen concentrations in different types of muscle, it was necessary always to use the formed by Drs. Maria L. Nestorescu and E. Siess, electron microscopy division of this institute. same muscle for glycogen determination (Table 3). Table 1. ATP, ADP and creatine phosphate (CP) levels of the gastrocnemius muscle in vivo, after perfusion, and during anoxia. Values in μπιοί χ g~ l wet wt.
ATP
In vivo Perfusion Anoxia
60 min 5 min 30 min 60 min
6.03 ± 0.39 (« = 8) 5.76 ± 0.66 (n = 10) 5.46 ± 0.24 C« = 5) 4.86 ±0.14 (w=5) 4.48 ± 0.20 (w=5)
ADP
CP
0.77 ± 0.02 (n = 34)
18.02+ 1.67 («=8) 18.05+ 1.98 (n = 10)
0.94 ± 0.07 (« = 5) 1.20 ± 0.04 (w = 5) 1.55 ±0.17 (n = 5)
9.84 + 0.32 («=5) 2.30 + 0.34 (« = 5) 1.50 ± 0.29 (n=5)
Brought to you by | Purdue University Li Authenticated Download Date | 5/22/15 5:31 AM
1059
t = 60 min
Influence of Insulin
vo
7L-
r-
oo
κ
s:
II
, =0
3.74 ± 0.95 n =6
Fatty acid release Glycerol release
I
I
«a. -Si
ON ro
A 0.05.
poration into triglycerides at increasing fatty acid concentrations, and this is stimulated by insulin. Thus triglyceride synthesis and fatty acid uptake (Fig. 3) exhibit similar dependence on fatty acid concentration with and without insulin. When glucose was omitted from the perfusate, no insulin effect either on fatty acid uptake or on triglyceride formation was demonstrable, whereas a concentration dependence was still observed under these conditions (Figs. 5 and 6). As to the other lipid fractions, 14C incorporation showed no correlation with fatty acid concentration. There was also no effect of insulin, either in the absence or in the presence of glucose (Table 6). Discussion The technique of rat hind quarter perfusion according to Ruderman and coworkers^18' represents a clear improvement over earlier reported methods for the study of skeletal muscle metabolism*37"47'. When we started our experiments, we used the perfusion medium containing human erythrocytes as described by Ruderman. It turned out, however, that the presence of red blood cells raised some complications, mainly due to their glycolytic activity and to difficulties in measuring oxygen consumption. Even 4 - 6 week-old erythrocytes still produced considerable amounts of lactate, which severely inter-
Brought to you by | Purdue University Li Authenticated Download Date | 5/22/15 5:31 AM
Bd. 356(1975)
Influence of Insulin
1063
fered with the net calculation of muscle lactate metabolism. We have therefore omitted the erythrocytes from the medium and found that the muscle remains metabolically and structurally well intact during hemoglobin-free perfusion. The high flow rate of about 35 m//min provides a sufficient oxygen supply to the muscle tissue, at least under resting conditions. Thus, venous oxygen tension remained constant over the 60 min perfusion time, and never fell below 143.5 ± 20.8 mm Hg. The mean arterial oxygen tension was 534.2 ± 48.3 mm Hg. Correspondingly, as shown in Table 1, ATP and creatine phosphate levels remain in the same range as in vivo under these conditions. As reported previously [14,48]^ kasai m(^ insulin-stimulated glucose uptake rates are the same either with or without erythrocytes in the medium. As compared to the data of Ruderman et al.118'491, the present study revealed higher basal and insulin-stimulated glucose uptake rates. This could be explained, at least in part, by the fact that these authors used a glucose concentration of 5 - 6mM for perfusion as compared to 10mM in this communication.
fully confirmed in this laboratory^53!. Thus, it would appear that the mechanism proposed by Rändle et alJ11, ascribing to free fatty acids a regulatory role on glucose uptake and oxidation, reflects the special situation of heart muscle and cannot be generalized to skeletal muscle.
During a perfusion period of 60 min, the glycogen concentration of the muscle was increased by insulin (1 unit//) by 20%. Confirming the results of others^43'50'51\ we have found very marked variations of the glycogen contents on analyzing muscle specimens taken from different sites of the preparation. Thus, an accurate balance of the amount of glucose converted to glycogen cannot be calculated with certainty. Like Garland and Rändle'41, who studied rat diaphragm, we were not able to demonstrate an effect of insulin on glycerol release, whereas fatty acid output was significantly reduced by the hormone (see alsol48!). In contrast, Ruderman et alJ181 have reported an inhibitory effect of insulin on glycerol release in the perfused hind quarter. In the present experiments, fatty acids at different concentrations did not affect glucose uptake and glycogen synthesis of skeletal muscle. This is in accordance with the findings of other investigators, who were not able to demonstrate an inhibitory effect of fatty acids or ketones on glucose uptake by skeletal muscle preparations110114'44'49'521. On the other hand, the decrease in glucose uptake of perfused rat heart in the presence of fatty acids121 could be
In all groups studied, more than 70% of the [l-14C]oleate incorporated into muscle lipids was found in the triglyceride fraction. This is in agreement with the results of Crass III et al.1601 in rat heart perfusion, and with those of Beatty and Bocek1 21 in rhesus monkey sartorius muscle. Increasing the oleate concentration in the medium resulted in a linear increase in triglyceride synthesis. Again, insulin had an additional stimulatory effect (Fig. 4). In rat diaphragm incubation studies, Schonfeld1611 observed that as the palmitate concentration in the incubation medium was elevated, the amount of fatty acids incorporated into muscle triglycerides increased. This observation is qualitatively confirmed by the present data.
As has been found in rat and human heart^ 54 " 57 ', in rat diaphragm1101, in voluntary muscle preparations from the rhesus monkey 112«441} and also in peripheral human skeletal muscle1581, the present investigation demonstrates that muscular extraction of fatty acids is dependent on their extracellular concentration. According to Ruderman and Goodman1591 this applies also to the uptake and utilization of ketone bodies by the perfused skeletal muscle. Previous experiments in our laboratory had already shown that insulin very markedly increased the uptake of oleate by the perfused rat muscle (unpublished data). This is confirmed by the data of Fig. 3 illustrating the stimulation of fatty acid uptake by insulin over the concentration range from 0.7 - 1.9mM oleate. As may be seen, the net increase in fatty acid uptake due to insulin remained nearly constant over this concentration range.
As illustrated in Figs. 5 and 6, the insulin effect on fatty acid uptake and triglyceride synthesis disappeared when glucose was omitted from the medium. In their studies with sartorius muscle biopsies from rhesus monkeys, Beatty and Bocek^121 found that the incorporation of label from [U-14C]glucose into total muscle lipids was enhanced by the addition of palmitate to the in-
Brought to you by | Purdue University Lib Authenticated Download Date | 5/22/15 5:31 AM
1064
F. Reimer, G. Löfflet, G. Hennig and O. H. Wieland
cubation medium. Because the majority (86%) of the label found in muscle .lipids was recovered in the triglycerides, the authors concluded that the incorporation of [14C]glucose into muscle glycerides occurred primarily via sn-glycerol-3phosphate. Our present data would permit the same conclusion: insulin stimulates the transport and utilization of glucose and simultaneously increases fatty acid uptake and triglyceride formation. Because of increased glycolysis, more s/2-glycerol 3phosphate is produced, which serves as an acceptor for the fatty acids. This would then result in an acceleration of both triglyceride synthesis and uptake of extracellular fatty acids. If fatty acid incorporation into muscle triglycerides is plotted against fatty acid uptake, a marked and highly significant correlation is obtained (Fig.7). This finding points to the possibility that the rate of fatty acid uptake may be regulated mainly by the ability of the muscle cell to esterify fatty acids. When the supply of s«-glycerol 3-phosphate is increased due to insulin, the esterification rate and, consequently, the uptake of fatty acids is increased.
Bd. 356(1975)
In the present experiments at different extracellular fatty acid concentrations the absolute amount of fatty acids which underwent oxidation remained constant. There was no effect of insulin on oleate oxidation. The rates of palmitate oxidation reported by Beatty and Bocekl12! in incubated sartorius muscle biopsies of rhesus monkeys are exactly in the same range as in the present study, namely 11 % of the label from [l-14C]palmitate uptake appeared in the C02. The effect of insulin on fatty acid metabolism was not investigated by these authors. Cassens et alJ 44 ^, using the same preparation as Beatty and Bocek^ 12 ', observed that octanoate oxidation increased with increasing concentrations of octanoate in the incubation assay. This is most likely due to the fact that short-chain fatty acids cannot be incorporated into triglycerides, as is known from the liver. Physiologically, the regulation of lipid metabolism in muscle is of special interest in relation to the energy supply of the muscle. Thus, as shown by numerous investigators, intramuscular lipids contribute to an important degree as fuel during exercise^62"66'. From our results, one could speculate that insulin plays an important role for the replenishment not only of muscular glycogen but also of lipid stores.
Literature
^ I
a· W 20 30 50 60 70 Fa tty acid uptake [nmol g -; min -'/ —^Fig. 7. Correlation between fatty acid uptake and fatty acid incorporation into muscle triglycerides. Experimental data from Figs. 3 and 4. Open circles without insulin, filled circles 1 unit insulin//. Correlation coefficient: r = 0.912, < 0.05.
1 Rändle, P. J., Garland, P. B., Hales, C. N. & Newsholme, E. A. (1963) Lancet I, 785 - 789. 2 Rändle, P. L, Newsholme, E. A. & Garland, P. B. (1964) Biochem. J. 93, 652 - 665. 3 Garland, P. B., Newsholme, E. A. & Rändle, P. J. (1964) Biochem. J. 93, 665 - 678. 4 Garland, P. B. & Rändle, P. J. (1964) Biochem. J. 91,6C-7C. 5 Newsholme, E. A. & Rändle, P. J. (1964) Biochem. J. 93,641-651. 6 England, P. J. & Rändle, P. J. (1967) Biochem. J. 105, 907 - 920. 7 Rändle, P. J., Denton, R. M. & England, P. J. (1968) in Metabolic Roles of Citrate (Goodwin, D., ed.) pp. 87 -103, Academic Press, London, New York. 8 Rändle, P. J. (1969) Nature (London] 221, 777. 9 Newsholme, E. A. & Start, C. (1973) in Regulation in Metabolism (Newsholme, E. A. & Start, C., eds.) pp. 88 -137, John Wiley & Sons, Ltd., London, New York.
Brought to you by | Purdue University Li Authenticated Download Date | 5/22/15 5:31 AM
Bd. 356 (1975)
Influence of Insulin
10 Schonfeld, G. & Kipnis, D. M. (1968) Amer. J, Physiol. 215, 513-522. 1l Houghton, C. R. S. & Ruderman, N. B. (1971) Biochem.J. 121, 1 5 p - 16p. 12 Beatty, C. H. & Bocek, R. M. (1971) Amer. J. Physiol. 220,1928-1934. 13 Jefferson, C. S., Koehler, J. O. & Morgan, H. E. (1972) Proc. Nat. Acad. Sei. U.S.A. 69, 816 - 820. 14 Löffler, G., Strohfeldt, P., Reimer, F. & Wieland, O. (1973) ffczsJ. 354,230-231. 15 Berger, M., Goodman, N. N., Hagg, S.A. & Ruderman, N. B. (1974) Diabetes 23, 347. 16 Krebs, H. A. & Henseleit, K. (1932) thisJ. 210, 33-66. 17 Guder, W. G. & Wieland, O. H. (1972) Eur. J. Biochem. 31,69-79. 18 Ruderman, N. B., Houghton, C. R. S. & Hems, R. (1971) Biochem. J. 124, 639 - 651. 19 Ross, B. D. (1972) Perfusion Techniques in Biochemistry - A Laboratory Manual, Clarendon Press, Oxford. 20 Bergmeyer, H. U., Bernt, E., Schmidt, F. & Stork, H. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd. edn., Vol. 2, pp. 1241 - 1250, Verlag Chemie, Weinheim/Bergstr. 21 Gutmann, I. & Wahlefeld, A.W. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 1510 - 1514, Verlag Chemie, Weinheim/ Bergstr. 22 Czok, R. & Lamprecht, W. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 1491 -1496, Verlag Chemie, Weinheim/ Bergstr. 23 Eggstein, M. & Kuhlmann, E. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 1871 - 1878, Verlag Chemie, Weinheim/ Bergstr. 24 Karl, J.E., Pagliara, A. S. & Kipnis, D. M. (1972) /. Lab. Clin. Med. 80, 434 - 441. 25 Buncombe, W. G. (1963) Biochem. J. 88, 7 -10. 26 Guder, W. G., Weiss, L. & Wieland, O. (1969) Biochim. Biophys. Acta 187, 173 -185. 27 Lamprecht, W. & Trautschold, I. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 2151 - 2160, Verlag Chemie, Weinheim/Bergstr. 28 Jaworek, D., Gruber, W. & Bergmeyer, H. U. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 2178 - 2181, Verlag Chemie, Weinheim/Bergstr. 29 Keppler, D. & Decker, K. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 1171 - 1176, Verlag Chemie, Weinheim/ Bergstr. 30 Bray, G. A. (1960) Anal. Biochem. I, 279 - 283. 31 Folch, J., Ascoli, L, Lees, M., Meath, J. A. & Baron, L. A. (1951) /. Biol. Chem. 191, 833 - 840.
1065
32
Sperry, W. M. (1955) Methods Biochem. Anal. 2, 83 - 86. 33 Sachs, L. (1972) Statistische Auswertungsmethoden, 3rd. edn., pp. 244 - 246, Springer-Verlag, Heidelberg, Berlin, New York. 34 Sachs, L. (1974) Angewandte Statistik, SpringerVerlag, Berlin, Heidelberg, New York. 35 Luft, J. H. (1961) /. Biophys. Biochem. Cytol 9, 409-414. 36 Reynolds, E. S. (1963) /. Cell. Biol. 17, 208 - 215. 37 Nakada,H. J. (1956)/. Biol. Chem. 219, 319-326. 38 Kipnis, D. M. & Cori, C. F. (1957) /. Biol. Chem. 224,681-693. 39 Mortimore, G. E., Tietze, F. & Stetten, D. (1959) Diabetes 8, 307-314. 40 Forsander, 0., Räihä, N. & Suomalainen, H. (1960) thisJ. 318, 1-5. 41 Moriwaki, T. & Landau, B. R. (l962) Arch. Biochem. Biophys. 97, 544 - 550. 42 Mahler, R. J., Szabo, D. & Penhos, J. C. (1968) Diabetes 17, l - 7. 43 Chapler, C. K. & Stainsby, W. N. (1968) Amer. J. Physiol. 215, 995 - 1004. 44 Cassens, R. G., Bocek, R. M. & Beatty, C. H. (1969) Amer. J. Physiol. 217, 715 - 719. 45 Chaudry, J. H. & Gould, M. K. (1969) Biochim. Biphys. Acta 177, 527 - 536. 46 Rookledge, K. A. (1971) Biochem. J. 125, 93 - 96. 47 Costin, J. C., Saltin, B., Skinner, S., Jr. & Vastagh,G. (1971) Acta Physiol. Scand. 81, 124 -137. 48 Reimer, F., Löffler, G., Gerbitz, K. & Wieland, 0. H. (1973) thisJ. 354,1232-1233. 49 Berger, M., Hagg, S. and Ruderman, N. B. (1975) Biochem. J. 146,231-238. 50 Kubista, V., Kubista, J. & Pette, D. (1971) Eur. J. Biochem. 18, 553 - 560. 51 Wermers, G. W., Cavert, H. M., Harris, J. 0. & Quelle, C.F. (1970) Amer. J. Physiol. 219, 1434 -1439. 52 Levari, R. & Kornbleuth, W. (1966) Israeli. Med. Sei. 3,513-523. 53 Wieland, O., Funcke, H. J. v. & Löffler, G. (1971) FEBSLett. 15,295-298. 54 Rudolph, W. & Hauer, G. (1970) Klin. Wochenschr. 48,3, 154- 161. 55 Müler, J. H., Yum, K. Y. & Durham, B. C. (1971) Amer. J. Physiol. 220, 589 - 596. 56 Crass, M. F. III (1972) Biochim. Biophys. Acta 280, 71-81. 57 Oram, J. F., Bennetch, S. L. & Neely, J. R. (1973) /. Biol. Chem. 248, 5299 - 5309. 58 Hagenfeldt, L., Wahren, J., Pernow, B. & Räf, L. (1972) /. Clin. Invest. 51, 2324 - 2330. 59 Ruderman, N. B. & Goodman, M. N. (1973) Amer. J. Physiol. 224, 1391 -1397. 60 Crass, M. F. III, McCaskül, E. S., Shipp, J. C & Murthy, V. K. (1971) Amer. J. Physiol. 220, 428 - 435.
Brought to you by | Purdue University Lib Authenticated Download Date | 5/22/15 5:31 AM
1066 61
F. Reimer, G. Löffler, G. Hennig and O. H. Wieland
Schonfeld, G. (1968) /. Lipid Res. 9, 453 - 459. Carlson, L. A., Ekelund, G. G. & Fröberg, S. 0 (1971) Ew. J. Clin. Invest, l, 248 - 254. 63 Hagenfeldt, L. & Wahren, J. (1972) Scand. J. Clin. Lab. Invest. 30,429-436. 64 Askew, E. W., Huston, R. L. & Dohm, G. L. (1973) Metabolism 22, 473 - 480. 65 Therriault, D. G., Beller, G. ., Smoake, J. A. & Hartley, L. H. (1973) /. Lipid Res. 14, 54 - 60. 62
66
Bd. 356 (1975)
Fröberg, S.O. & Morsfeldt, F. (l91l)ActaPhysiol. · > 167 ' 171·
Scand 82
The skilful technical assistance of Miss M. Scheuerecker js greatly acknowledged. This work was supported by the Deutsche Forschungsgemeinschaft, Bad Godesberg, Germany.
Brought to you by | Purdue University Li Authenticated Download Date | 5/22/15 5:31 AM
The Influence of Insulin on Glucose and Fatty Acid Metabolism in the Isolated Perfused Rat Hind Quarter Felix Reimer, Georg Löffler, Gerd Hennig and Otto H. Wieland
(Received 6 February 1975) Dedicated to Prof. Dr. Günther Weitzel on the occasion of his 60th birthday. Summary: Glucose and fatty acid metabolism of resting skeletal muscle were studied by perfusion of the isolated rat hind leg with a hemoglobin-free medium. Tissue integrity was demonstrated by normal ATP, ADP and creatine phosphate levels, by a sufficient oxygen supply, and by a normal appearance of perfused muscle specimens under the electron microscope. The rates of glucose and fatty acid uptake, and of lactate, alanine, glycerol and fatty acid release were constant over a perfusion period of 60 min. Insulin (1 unit//) caused a more than threefold increase in glucose uptake, a stimulation of lactate production, and a 20% increase in the muscular glycogen levels. Fatty acid and alanine release were significantly diminished by insulin, but glycerol release did not change. The uptake of oleate by the rat hind leg was dependent on the medium concentration in a range of 0.7 - 1.9mM oleate, and was stimulated by insulin. Glucose uptake was not influenced by oleate, whether insulin was present or not. When the leg was perfused with [ l-14C]oleate, 75% of the incorporated
fatty acids were found in muscle lipids, 10% were oxidized to C0 2s and 5% were recovered in bone lipids. The absolute amount of oleate oxidation was not altered by insulin. In all experiments with and without glucose in the medium, 70-80% of the 14C label incorporated into muscle lipids was found in the triglyceride fraction. In the presence of glucose, insulin significantly increased the incorporation of [ l-14C]oleate.into muscle triglycerides, whereas no insulin effect, either on fatty acid uptake or on triglyceride formation, could be observed when glucose was omitted from the perfusate. The present results indicate that a "glucose-fatty acid cycle" as found in rat heart muscle does not operate in resting peripheral skeletal muscle tissue. They also demonstrate that the stimulating effect of insulin on muscular fatty acid, uptake and triglyceride synthesis is dependent on glucose supply. This finding can be interpreted as a stimulation of fatty acid esterification by ^«-glycerol 3-phosphate derived from an increased glucose turnover, which is in turn due to insulin.
Die Wirkung von Insulin auf den Glucose- und Fettsäurestoffwechsel Hinterkörpers der Ratte Zusammenfassung: Der Glucose- und Fettsäurestoffwechsel des nicht arbeitenden Skelettmuskels wurde am Modell des hämoglobinfrei perfun-
des isoliert perfundierten
dierten Hinterkörpers der Ratte untersucht. Gemessen am elektronenmikroskopischen Bild, am Sauerstoffverbrauch sowie an den Phospho-
Address: Prof. Dr. 0. H. Wieland, Klinisch-Chemisches Institut und Forschergruppe Diabetes, Krankenhaus München-Schwabing, D-8 München 40, Kölner Platz 1.
Brought to you by | Purdue University Lib Authenticated Download Date | 5/22/15 5:31 AM
1056
F. Reimer, G. Löffler, G. Hennig and O. H. Wieland
kreatin-, ATP- und ADP-Gehalten war das Präparat im Vergleich zum Zustand in vivo intakt. Die Geschwindigkeit von Glucose- und Fettsäureaufnahme sowie die Abgabe von Lactat, Alanin, Glycerin und Fettsäuren waren während 60minütiger Perfusion konstant. l Einh.// Insulin führte zu einer Zunahme der Glucoseaufnahme um mehr als das Sfache, zu einer Steigerung der Lactatbildung und einem Anstieg des Glykogengehaltes um 20%. Fettsäureund Alaninfreisetzung wurden durch Insulin signifikant vermindert, die Glycerinabgabe blieb jedoch unverändert. Die Aufnahmegeschwindigkeit von Ölsäure durch den perfundierten Muskel hing von der Fettsäurekonzentration ab und konnte durch Insulin stimuliert werden. Weder in Anwesenheit noch in Abwesenheit von Insulin wurde die Glucoseaufnahme durch Zugabe von Fettsäuren beeinflußt. 75% der aufgenommenen Fettsäuren fanden sich in den Muskellipiden, 10% wurden zu C02 oxi-
The role of elevated blood fatty acids and ketone bodies during starvation and diabetes in the regulation of muscle glucose metabolism was studied extensively by Rändle and coworkers^1"9^. In their experiments on rat heart and diaphragm muscle, it was found that glucose uptake, glycolysis, and pyruvate oxidation are inhibited when fatty acid oxidation increases due to a greater provision of fatty acids, either externally or from intracellular breakdown of glycerides. The theoretical concept of this reciprocal relationship ("glucose-fatty acid cycle") implies that when fatty acid oxidation is enhanced, the increased acetyl-CoA and citrate levels inhibit the pyruvate dehydrogenase and phosphofructokinase reactions, respectively. The inhibition of the latter leads to an accumulation of glucose 6-phosphate, which is known to be a potent inhibitor of hexokinase. As compared to heart muscle, inhibition of glucose metabolism by fatty acids in rat diaphragm was found to be rather small121, suggesting some fundamental difference in the metabolic properties of the two types of muscle tissues. Furthermore, using skeletal muscle preparations, other investigators were not able to demonstrate an effect of fatty acids or ketone
Bd. 356 (1975)
diert und 5 % in den Knochenmarklipiden wiedergefunden. Insulin war nicht imstande, die Geschwindigkeit der Ölsäureoxidation zu beeinflussen. 70 - 80% der in die Muskellipide aufgenommenen Fettsäuren konnten in der Triglyceridfraktion lokalisiert werden. In Anwesenheit von Glucose wurde der Einbau von Fettsäuren in diese Fraktion durch Insulin signifikant gesteigert, wohingegen bei glucosefreier Perfusion kein derartiger Effekt zu beobachten war. Die vorliegenden Befunde weisen daraufhin, daß die für das Rattenherz gefundene Hemmung des Glucoseumsatzes durch Fettsäuren („glucose-fatty acid cycle") für den nicht arbeitenden Skelettmuskel nicht zutrifft. Sie zeigen außerdem, daß Insulin in Abhängigkeit vom Glucoseangebot sowohl Fettsäureaufnahme als auch Triglyceridsynthese des Muskels stimuliert. Der zugrunde liegende Mechanismus dürfte in einer Stimulierung der Fettsäureveresterung mit tfi-Glycerin-3-phosphat liegen, das der Glykolyse entnommen wird.
bodies on glucose metabolisml10'15'. The present study was undertaken in order to re-investigate the relationship between glucose and fatty acid metabolism in a perfused skeletal muscle preparation.
Material and Methods Animals. Male Sprague-Dawley rats (Wiga, Sulzfeld, Germany) weighing 170 - 230 g were used. The animals were fed on a standard laboratory diet (Labortierfutter 57 Z, J. Zahn II, Hockenheim, Germany). Reagents. Nembutal (sodium pentobarbital) was supplied by the German Abbot GmbH, Ingelheim, Germany. Bovine serum albumin (Cohn fraction V) was obtained from Serva International, Heidelberg, Germany. It was dissolved in Krebs-Henseleit hydrogen carbonate bufferl 161 and dialysed twice against a tenfold volume of the same buffer for 24 h at 4 °C. Bovine insulin (25 units/mg) was kindly supplied by Farbwerke Hoechst, Germany. Oleic acid (distilled, puriss. > 98%) was obtained from Carl Roth OHG, Karlsruhe, Germany, and [l-14Cjoleic acid from the Radiochemical Centre, Amersham, Buckinghamshire, England. Lipid standards were obtained from Serva International, Heidelberg, Germany. Sodium pyruvate and enzymes were purchased from
Brought to you by | Purdue University Li Authenticated Download Date | 5/22/15 5:31 AM
Bd. 356 (1975)
Influence of Insulin
1057
as modified by Guder et alJ 26 l ATP and creatinephosphate were estimated according to Lamprecht and Trautscholdl27!, with the exception that the final glucose concentration in the assay was only 0.0IM. ADP was Perfusion medium. The standard perfusion medium was estimated by the method of Jaworek et aU 28 l, and composed as follows: Krebs-Henseleit hydrogen carbonate glycogen according to Keppler and Decker^29!. The Ο2 buff er gassed with carbogen (95% C>2 and 5% C02) and tension in the hemoglobin-free perfusate was detercontaining 4 g/100 m/ bovine serum albumin, lOmM mined polarographically with a Clark electrode (Radioglucose, O.lSmM pyruvate, and 1.5mM lactate. If oleic meter, Copenhaven, Denmark). The C»2 consumption acid was used, an albumin-oleate complex was prepared was calculated from the arterio-venous differences. At 3 as described by Garland et al.^ J and Guder and Wiethe end of the perfusion, muscle samples (30 - 100 mg) 17 land * ^ and mixed with the perfusion medium. The were obtained by freeze-clamping with a Wollenberger final albumin concentration of the perfusate was always clamp, weighed and homogenized with a high-speed kept at 4 g/100m/. tissue disintegrator (Ultra-Turrax 18/2, Jahnke & Kunkel, Stauffen i.B., Germany) in ice-cold 6% HC104, Surgical preparation. The preparation procedure was (100 mg tissue/m/ HC1 4 for the determination of glycarried out essentially as described by Ruderman et cogen, ATP, ADP, and creatine phosphate). al.l 181. As a modification, the major spinal vessels derivWhen perfused with [l-14C)oleic acid, CO2 was coling dorsally from the aorta and vena cava, and vessels lected in 25% (w/v) phenylethylamine in methanol, originating from the femoral arteries and veins below the inguinal ligament, which all supply the subcutaneous using a small plastic tube attached to the gas outlet of the oxygenator. The samples were placed in Bray's soluregion, were tied up. A ligature was also placed around tioni30! and counted for radioactivity. Tissue lipids were the liver radix. After the canulation of the aorta and extracted with chloroform/methanol (2:1, v/v) as outvena cava, the animal was killed by intrathoracal injeclined by Folch et a 31 J and Sperryl 32 L For radiotion of 40 mg Nembutal. In this way, peripheral vasactivity measurements, the lipids were separated on cular constrictions were largely avoided, and the final 20 χ 20 cm plates coated with silica gel 60 (Merck, perfusion pressure could already be adjusted after Darmstadt, Germany) using a solvent mixture of light 2 - 3 min. petroleum/ethyl ether/acetic acid (80:20:1). The lipidPerfusion system. The perfusion system, in general, was containing areas identified by standards were scraped the same as used for liver perfusions (for references off, transferred into the scintillation mixture (Scintigel, seel 19]) w}t^ t^e exception of a second pump behind Carl Roth OHG, Karlsruhe, Germany), and sonified in the oxygenator for maintaining an arterial pressure of order to secure homogeneous dispersion of the gel parti80 - 90 mm Hg and a constant blood flow of about cles. After the addition of water, stable and homogeneous 35 m//min. In order to get rid of erythrocytes, the first suspensions were obtained, the radioactivity of which was 50 - 60 m/ of the perfusate was discarded. The recircumeasured in a Packard model 3385 Tri-Carb liquid scintillating perfusion was then carried out with 150 ml melation spectrometer. dium. The hematocrit of the perfusate was below 0.1%. Calculations and statistical evaluation. Metabolic rates After 5 min of equilibration, the first medium sample were calculated from metabolite concentration differwas taken to determine the "zero" values. Beginning ences between the "zero" and 60 min values of perfrom "zero", the perfusion was continued for 60 min. fusion, and are given as nmol or μπιοί χ g~ 1 muscle Samples were collected at 15 min intervals. The effect tissue χ min"1. The total mass of the perfused muscle of insulin was tested by addition of a single dose of tissue was calculated using the formula of Ruderman 1 unit//. The temperature in the perfusion box was mainet alJ 18 l, which is based on rat total body weight. Sigtained at 37 °C by means of a thermoregulator. nificance was determined by the Wilcoxon rank sum Analytical methods. The perfusate samples were detest as modified by Mann and Whitney!331. Regression proteinized with ice-cold HQ 4 at a final concentration coefficients were calculated as described by Sachs^34!. of 4.5% (v/v). The extracts were neutralized with solid Electron microscopy. Tissue was excised from the adducKHC 3, and KC104 was removed by centrifugation. tor muscle both in nonperfused animals and after perGlucose was determined by the hexokinase/glucose-620 fusion. The muscle was cut into 0.5-mm cubes and then phosphate dehydrogenase reaction^ !. Lactate was fixed in 2.5 % glutaraldehyde buffered with 0.15M phosdetermined enzymatically as described by Gutmann and phate buffer, pH 7.2. Thereafter the tissue was dehyWahlefeldl21!, pyruvate as described by Czok and Lam22 drated with ethanol and propylene oxide, and then emprechti !, glycerol according to Eggstein and Kuhl35 mann^23!, and alanine according to Karl et alJ24l. Fatty bedded in Epon^ l Sections were cut on a LKB-microtomeNo.HI with a diamond knife, stained with 2% (w/v) acids were determined by the Duncombe procedure^ 25 ! Boehringer Mannheim, Germany. Sodium L-lactate was obtained from Serva International. All other reagents were obtained from Merck, Darmstadt, Germany.
Brought to you by | Purdue University Lib Authenticated Download Date | 5/22/15 5:31 AM
1058
F. Reimer, G. L ffler, G. Hennig and O. H. Wieland
uranyl acetate and lead citrate^ 36 ', and examined in a Siemens Elmiskop 101 electron microscope*.
Results An edema of the perfused hind legs was not observed when albumin concentration was 4 g/ 100 ml, and the perfusion pressure did not exeed 80 - 100 mm Hg. Under these conditions, ultrastructural examination did not reveal any differences in adductor muscle specimens taken in vivo and 60 min after perfusion. The polarographically determined venous oxygen tension was in the range of 120 - 180 mm Hg and remained constant over the 60 min perfusion time. The levels of ATP and creatine phosphate were the same in perfused and nonperfused muscle specimens (Table 1). In this context, it seems noteworthy that we could detect only small changes in the ATP and ADP levels in muscle during anoxia, at least for 60 min, whereas creatine phosphate was already decreased to 55% of the control after 5 min (Table 1). Obviously the resting skeletal muscle is able to maintain a high ATP/ADP ratio over long periods of anoxia at the expense of creatine phosphate. Therefore, the creatine phosphate level is a better control parameter for the evaluation of the metabolic state of the perfused muscle. This is further documented by the observation that during a 10 min perfusion where the medium was gassed with N 2 /C0 2 instead of
Bd. 356(1975)
O 2 /CO 2 , creatine phosphate decreased to 68% of the control, while the adenine nucleotides remained constant. The decrease in creatine phosphate was fully reversible after switching back to O 2 /C0 2 gassing.
Metabolism of the resting muscle. As the rates of glucose (Fig. 1) and fatty acid uptake, lactate production, alanine, glycerol, and fatty acid release (in the oleate-free experiments) were linear over a perfusion period of at least 60 min, it was possible to calculate the substrate turnover rates as nmol or μιηοΐ χ g~ * muscle tissue χ min~ *. Table 2 shows some basal metabolic rates and the effect of insulin (1 unit//) thereon. As may be seen, glucose uptake is stimulated by insulin more than threefold. In the controls, 50% of the extracted glucose was recovered as lactate, whereas in the presence of insulin, this proportion decreased to 27%. The lactate/pyruvate ratio was increased at the end of perfusion in both the control and insulin groups. Fatty acid and alanine release were significantly diminished by insulin, whereas glycerol release was not influenced. In accordance with the findings of Ruderman et aU 181, we observed no difference in the rate of basal lipolysis when the hind legs were skinned and all visible adipose tissue was removed before perfusion. The glycogen content remained constant during 60 min of perfusion and showed values comparable to the muscle in nonperfused animals (32.3 ± 4.95 μτηοΐ/g). Insulin caused a 20% increase in glycogen content. Due to the marked differences * The electron microscopic examination was kindly per- in the glycogen concentrations in different types of muscle, it was necessary always to use the formed by Drs. Maria L. Nestorescu and E. Siess, electron microscopy division of this institute. same muscle for glycogen determination (Table 3). Table 1. ATP, ADP and creatine phosphate (CP) levels of the gastrocnemius muscle in vivo, after perfusion, and during anoxia. Values in μπιοί χ g~ l wet wt.
ATP
In vivo Perfusion Anoxia
60 min 5 min 30 min 60 min
6.03 ± 0.39 (« = 8) 5.76 ± 0.66 (n = 10) 5.46 ± 0.24 C« = 5) 4.86 ±0.14 (w=5) 4.48 ± 0.20 (w=5)
ADP
CP
0.77 ± 0.02 (n = 34)
18.02+ 1.67 («=8) 18.05+ 1.98 (n = 10)
0.94 ± 0.07 (« = 5) 1.20 ± 0.04 (w = 5) 1.55 ±0.17 (n = 5)
9.84 + 0.32 («=5) 2.30 + 0.34 (« = 5) 1.50 ± 0.29 (n=5)
Brought to you by | Purdue University Li Authenticated Download Date | 5/22/15 5:31 AM
1059
t = 60 min
Influence of Insulin
vo
7L-
r-
oo
κ
s:
II
, =0
3.74 ± 0.95 n =6
Fatty acid release Glycerol release
I
I
«a. -Si
ON ro
A 0.05.
poration into triglycerides at increasing fatty acid concentrations, and this is stimulated by insulin. Thus triglyceride synthesis and fatty acid uptake (Fig. 3) exhibit similar dependence on fatty acid concentration with and without insulin. When glucose was omitted from the perfusate, no insulin effect either on fatty acid uptake or on triglyceride formation was demonstrable, whereas a concentration dependence was still observed under these conditions (Figs. 5 and 6). As to the other lipid fractions, 14C incorporation showed no correlation with fatty acid concentration. There was also no effect of insulin, either in the absence or in the presence of glucose (Table 6). Discussion The technique of rat hind quarter perfusion according to Ruderman and coworkers^18' represents a clear improvement over earlier reported methods for the study of skeletal muscle metabolism*37"47'. When we started our experiments, we used the perfusion medium containing human erythrocytes as described by Ruderman. It turned out, however, that the presence of red blood cells raised some complications, mainly due to their glycolytic activity and to difficulties in measuring oxygen consumption. Even 4 - 6 week-old erythrocytes still produced considerable amounts of lactate, which severely inter-
Brought to you by | Purdue University Li Authenticated Download Date | 5/22/15 5:31 AM
Bd. 356(1975)
Influence of Insulin
1063
fered with the net calculation of muscle lactate metabolism. We have therefore omitted the erythrocytes from the medium and found that the muscle remains metabolically and structurally well intact during hemoglobin-free perfusion. The high flow rate of about 35 m//min provides a sufficient oxygen supply to the muscle tissue, at least under resting conditions. Thus, venous oxygen tension remained constant over the 60 min perfusion time, and never fell below 143.5 ± 20.8 mm Hg. The mean arterial oxygen tension was 534.2 ± 48.3 mm Hg. Correspondingly, as shown in Table 1, ATP and creatine phosphate levels remain in the same range as in vivo under these conditions. As reported previously [14,48]^ kasai m(^ insulin-stimulated glucose uptake rates are the same either with or without erythrocytes in the medium. As compared to the data of Ruderman et al.118'491, the present study revealed higher basal and insulin-stimulated glucose uptake rates. This could be explained, at least in part, by the fact that these authors used a glucose concentration of 5 - 6mM for perfusion as compared to 10mM in this communication.
fully confirmed in this laboratory^53!. Thus, it would appear that the mechanism proposed by Rändle et alJ11, ascribing to free fatty acids a regulatory role on glucose uptake and oxidation, reflects the special situation of heart muscle and cannot be generalized to skeletal muscle.
During a perfusion period of 60 min, the glycogen concentration of the muscle was increased by insulin (1 unit//) by 20%. Confirming the results of others^43'50'51\ we have found very marked variations of the glycogen contents on analyzing muscle specimens taken from different sites of the preparation. Thus, an accurate balance of the amount of glucose converted to glycogen cannot be calculated with certainty. Like Garland and Rändle'41, who studied rat diaphragm, we were not able to demonstrate an effect of insulin on glycerol release, whereas fatty acid output was significantly reduced by the hormone (see alsol48!). In contrast, Ruderman et alJ181 have reported an inhibitory effect of insulin on glycerol release in the perfused hind quarter. In the present experiments, fatty acids at different concentrations did not affect glucose uptake and glycogen synthesis of skeletal muscle. This is in accordance with the findings of other investigators, who were not able to demonstrate an inhibitory effect of fatty acids or ketones on glucose uptake by skeletal muscle preparations110114'44'49'521. On the other hand, the decrease in glucose uptake of perfused rat heart in the presence of fatty acids121 could be
In all groups studied, more than 70% of the [l-14C]oleate incorporated into muscle lipids was found in the triglyceride fraction. This is in agreement with the results of Crass III et al.1601 in rat heart perfusion, and with those of Beatty and Bocek1 21 in rhesus monkey sartorius muscle. Increasing the oleate concentration in the medium resulted in a linear increase in triglyceride synthesis. Again, insulin had an additional stimulatory effect (Fig. 4). In rat diaphragm incubation studies, Schonfeld1611 observed that as the palmitate concentration in the incubation medium was elevated, the amount of fatty acids incorporated into muscle triglycerides increased. This observation is qualitatively confirmed by the present data.
As has been found in rat and human heart^ 54 " 57 ', in rat diaphragm1101, in voluntary muscle preparations from the rhesus monkey 112«441} and also in peripheral human skeletal muscle1581, the present investigation demonstrates that muscular extraction of fatty acids is dependent on their extracellular concentration. According to Ruderman and Goodman1591 this applies also to the uptake and utilization of ketone bodies by the perfused skeletal muscle. Previous experiments in our laboratory had already shown that insulin very markedly increased the uptake of oleate by the perfused rat muscle (unpublished data). This is confirmed by the data of Fig. 3 illustrating the stimulation of fatty acid uptake by insulin over the concentration range from 0.7 - 1.9mM oleate. As may be seen, the net increase in fatty acid uptake due to insulin remained nearly constant over this concentration range.
As illustrated in Figs. 5 and 6, the insulin effect on fatty acid uptake and triglyceride synthesis disappeared when glucose was omitted from the medium. In their studies with sartorius muscle biopsies from rhesus monkeys, Beatty and Bocek^121 found that the incorporation of label from [U-14C]glucose into total muscle lipids was enhanced by the addition of palmitate to the in-
Brought to you by | Purdue University Lib Authenticated Download Date | 5/22/15 5:31 AM
1064
F. Reimer, G. Löfflet, G. Hennig and O. H. Wieland
cubation medium. Because the majority (86%) of the label found in muscle .lipids was recovered in the triglycerides, the authors concluded that the incorporation of [14C]glucose into muscle glycerides occurred primarily via sn-glycerol-3phosphate. Our present data would permit the same conclusion: insulin stimulates the transport and utilization of glucose and simultaneously increases fatty acid uptake and triglyceride formation. Because of increased glycolysis, more s/2-glycerol 3phosphate is produced, which serves as an acceptor for the fatty acids. This would then result in an acceleration of both triglyceride synthesis and uptake of extracellular fatty acids. If fatty acid incorporation into muscle triglycerides is plotted against fatty acid uptake, a marked and highly significant correlation is obtained (Fig.7). This finding points to the possibility that the rate of fatty acid uptake may be regulated mainly by the ability of the muscle cell to esterify fatty acids. When the supply of s«-glycerol 3-phosphate is increased due to insulin, the esterification rate and, consequently, the uptake of fatty acids is increased.
Bd. 356(1975)
In the present experiments at different extracellular fatty acid concentrations the absolute amount of fatty acids which underwent oxidation remained constant. There was no effect of insulin on oleate oxidation. The rates of palmitate oxidation reported by Beatty and Bocekl12! in incubated sartorius muscle biopsies of rhesus monkeys are exactly in the same range as in the present study, namely 11 % of the label from [l-14C]palmitate uptake appeared in the C02. The effect of insulin on fatty acid metabolism was not investigated by these authors. Cassens et alJ 44 ^, using the same preparation as Beatty and Bocek^ 12 ', observed that octanoate oxidation increased with increasing concentrations of octanoate in the incubation assay. This is most likely due to the fact that short-chain fatty acids cannot be incorporated into triglycerides, as is known from the liver. Physiologically, the regulation of lipid metabolism in muscle is of special interest in relation to the energy supply of the muscle. Thus, as shown by numerous investigators, intramuscular lipids contribute to an important degree as fuel during exercise^62"66'. From our results, one could speculate that insulin plays an important role for the replenishment not only of muscular glycogen but also of lipid stores.
Literature
^ I
a· W 20 30 50 60 70 Fa tty acid uptake [nmol g -; min -'/ —^Fig. 7. Correlation between fatty acid uptake and fatty acid incorporation into muscle triglycerides. Experimental data from Figs. 3 and 4. Open circles without insulin, filled circles 1 unit insulin//. Correlation coefficient: r = 0.912, < 0.05.
1 Rändle, P. J., Garland, P. B., Hales, C. N. & Newsholme, E. A. (1963) Lancet I, 785 - 789. 2 Rändle, P. L, Newsholme, E. A. & Garland, P. B. (1964) Biochem. J. 93, 652 - 665. 3 Garland, P. B., Newsholme, E. A. & Rändle, P. J. (1964) Biochem. J. 93, 665 - 678. 4 Garland, P. B. & Rändle, P. J. (1964) Biochem. J. 91,6C-7C. 5 Newsholme, E. A. & Rändle, P. J. (1964) Biochem. J. 93,641-651. 6 England, P. J. & Rändle, P. J. (1967) Biochem. J. 105, 907 - 920. 7 Rändle, P. J., Denton, R. M. & England, P. J. (1968) in Metabolic Roles of Citrate (Goodwin, D., ed.) pp. 87 -103, Academic Press, London, New York. 8 Rändle, P. J. (1969) Nature (London] 221, 777. 9 Newsholme, E. A. & Start, C. (1973) in Regulation in Metabolism (Newsholme, E. A. & Start, C., eds.) pp. 88 -137, John Wiley & Sons, Ltd., London, New York.
Brought to you by | Purdue University Li Authenticated Download Date | 5/22/15 5:31 AM
Bd. 356 (1975)
Influence of Insulin
10 Schonfeld, G. & Kipnis, D. M. (1968) Amer. J, Physiol. 215, 513-522. 1l Houghton, C. R. S. & Ruderman, N. B. (1971) Biochem.J. 121, 1 5 p - 16p. 12 Beatty, C. H. & Bocek, R. M. (1971) Amer. J. Physiol. 220,1928-1934. 13 Jefferson, C. S., Koehler, J. O. & Morgan, H. E. (1972) Proc. Nat. Acad. Sei. U.S.A. 69, 816 - 820. 14 Löffler, G., Strohfeldt, P., Reimer, F. & Wieland, O. (1973) ffczsJ. 354,230-231. 15 Berger, M., Goodman, N. N., Hagg, S.A. & Ruderman, N. B. (1974) Diabetes 23, 347. 16 Krebs, H. A. & Henseleit, K. (1932) thisJ. 210, 33-66. 17 Guder, W. G. & Wieland, O. H. (1972) Eur. J. Biochem. 31,69-79. 18 Ruderman, N. B., Houghton, C. R. S. & Hems, R. (1971) Biochem. J. 124, 639 - 651. 19 Ross, B. D. (1972) Perfusion Techniques in Biochemistry - A Laboratory Manual, Clarendon Press, Oxford. 20 Bergmeyer, H. U., Bernt, E., Schmidt, F. & Stork, H. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd. edn., Vol. 2, pp. 1241 - 1250, Verlag Chemie, Weinheim/Bergstr. 21 Gutmann, I. & Wahlefeld, A.W. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 1510 - 1514, Verlag Chemie, Weinheim/ Bergstr. 22 Czok, R. & Lamprecht, W. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 1491 -1496, Verlag Chemie, Weinheim/ Bergstr. 23 Eggstein, M. & Kuhlmann, E. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 1871 - 1878, Verlag Chemie, Weinheim/ Bergstr. 24 Karl, J.E., Pagliara, A. S. & Kipnis, D. M. (1972) /. Lab. Clin. Med. 80, 434 - 441. 25 Buncombe, W. G. (1963) Biochem. J. 88, 7 -10. 26 Guder, W. G., Weiss, L. & Wieland, O. (1969) Biochim. Biophys. Acta 187, 173 -185. 27 Lamprecht, W. & Trautschold, I. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 2151 - 2160, Verlag Chemie, Weinheim/Bergstr. 28 Jaworek, D., Gruber, W. & Bergmeyer, H. U. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 2178 - 2181, Verlag Chemie, Weinheim/Bergstr. 29 Keppler, D. & Decker, K. (1974) in Methoden d. Enzymat. Analyse (Bergmeyer, H. U., ed.) 3rd edn., Vol. 2, pp. 1171 - 1176, Verlag Chemie, Weinheim/ Bergstr. 30 Bray, G. A. (1960) Anal. Biochem. I, 279 - 283. 31 Folch, J., Ascoli, L, Lees, M., Meath, J. A. & Baron, L. A. (1951) /. Biol. Chem. 191, 833 - 840.
1065
32
Sperry, W. M. (1955) Methods Biochem. Anal. 2, 83 - 86. 33 Sachs, L. (1972) Statistische Auswertungsmethoden, 3rd. edn., pp. 244 - 246, Springer-Verlag, Heidelberg, Berlin, New York. 34 Sachs, L. (1974) Angewandte Statistik, SpringerVerlag, Berlin, Heidelberg, New York. 35 Luft, J. H. (1961) /. Biophys. Biochem. Cytol 9, 409-414. 36 Reynolds, E. S. (1963) /. Cell. Biol. 17, 208 - 215. 37 Nakada,H. J. (1956)/. Biol. Chem. 219, 319-326. 38 Kipnis, D. M. & Cori, C. F. (1957) /. Biol. Chem. 224,681-693. 39 Mortimore, G. E., Tietze, F. & Stetten, D. (1959) Diabetes 8, 307-314. 40 Forsander, 0., Räihä, N. & Suomalainen, H. (1960) thisJ. 318, 1-5. 41 Moriwaki, T. & Landau, B. R. (l962) Arch. Biochem. Biophys. 97, 544 - 550. 42 Mahler, R. J., Szabo, D. & Penhos, J. C. (1968) Diabetes 17, l - 7. 43 Chapler, C. K. & Stainsby, W. N. (1968) Amer. J. Physiol. 215, 995 - 1004. 44 Cassens, R. G., Bocek, R. M. & Beatty, C. H. (1969) Amer. J. Physiol. 217, 715 - 719. 45 Chaudry, J. H. & Gould, M. K. (1969) Biochim. Biphys. Acta 177, 527 - 536. 46 Rookledge, K. A. (1971) Biochem. J. 125, 93 - 96. 47 Costin, J. C., Saltin, B., Skinner, S., Jr. & Vastagh,G. (1971) Acta Physiol. Scand. 81, 124 -137. 48 Reimer, F., Löffler, G., Gerbitz, K. & Wieland, 0. H. (1973) thisJ. 354,1232-1233. 49 Berger, M., Hagg, S. and Ruderman, N. B. (1975) Biochem. J. 146,231-238. 50 Kubista, V., Kubista, J. & Pette, D. (1971) Eur. J. Biochem. 18, 553 - 560. 51 Wermers, G. W., Cavert, H. M., Harris, J. 0. & Quelle, C.F. (1970) Amer. J. Physiol. 219, 1434 -1439. 52 Levari, R. & Kornbleuth, W. (1966) Israeli. Med. Sei. 3,513-523. 53 Wieland, O., Funcke, H. J. v. & Löffler, G. (1971) FEBSLett. 15,295-298. 54 Rudolph, W. & Hauer, G. (1970) Klin. Wochenschr. 48,3, 154- 161. 55 Müler, J. H., Yum, K. Y. & Durham, B. C. (1971) Amer. J. Physiol. 220, 589 - 596. 56 Crass, M. F. III (1972) Biochim. Biophys. Acta 280, 71-81. 57 Oram, J. F., Bennetch, S. L. & Neely, J. R. (1973) /. Biol. Chem. 248, 5299 - 5309. 58 Hagenfeldt, L., Wahren, J., Pernow, B. & Räf, L. (1972) /. Clin. Invest. 51, 2324 - 2330. 59 Ruderman, N. B. & Goodman, M. N. (1973) Amer. J. Physiol. 224, 1391 -1397. 60 Crass, M. F. III, McCaskül, E. S., Shipp, J. C & Murthy, V. K. (1971) Amer. J. Physiol. 220, 428 - 435.
Brought to you by | Purdue University Lib Authenticated Download Date | 5/22/15 5:31 AM
1066 61
F. Reimer, G. Löffler, G. Hennig and O. H. Wieland
Schonfeld, G. (1968) /. Lipid Res. 9, 453 - 459. Carlson, L. A., Ekelund, G. G. & Fröberg, S. 0 (1971) Ew. J. Clin. Invest, l, 248 - 254. 63 Hagenfeldt, L. & Wahren, J. (1972) Scand. J. Clin. Lab. Invest. 30,429-436. 64 Askew, E. W., Huston, R. L. & Dohm, G. L. (1973) Metabolism 22, 473 - 480. 65 Therriault, D. G., Beller, G. ., Smoake, J. A. & Hartley, L. H. (1973) /. Lipid Res. 14, 54 - 60. 62
66
Bd. 356 (1975)
Fröberg, S.O. & Morsfeldt, F. (l91l)ActaPhysiol. · > 167 ' 171·
Scand 82
The skilful technical assistance of Miss M. Scheuerecker js greatly acknowledged. This work was supported by the Deutsche Forschungsgemeinschaft, Bad Godesberg, Germany.
Brought to you by | Purdue University Li Authenticated Download Date | 5/22/15 5:31 AM
