Effect of Prolonged Starvation on Glycogen Synthase and Glycogen ...

Effect of Prolonged Starvation on Glycogen Synthase and Glycogen Synthase Phosphatase Activity in Rat Heart MARY C. GANNON1 AND FRANK Q. NUTTALL2

Department of Medicine and The Section of Endocrinology and Metabolism, Minneapolis VA Medical Center and The University of Minnesota, Minneapolis, MN 55417 ABSTRACT Rats were fasted 2A, 48 or 72 hours to determine the effect of several days without food on glycogen synthase and synthase phosphatase activity in heart. The basal percentage of synthase I decreased gradually from approximately 20% in fed animals to approximately 6% in rats starved for 72 hours. Glycogen increased progressively from 4.6 mg/g wet weight in fed rats to 7.6 mg/g wet weight in 72-hour starved rats. Thus, there was an inverse relationship between the glycogen concentra tion and the basal percentage of synthase I. The total synthase phosphatase activity measured at a standardized glycogen concentration decreased 50% by 24 hours of starvation and then was unchanged up to 72 hours. The 50% decrease in phosphatase activity correlated directly with insulin concentration in rats fasted 24-72 hours. The rapid stimulatory effect of insulin on synthase activity observed in fed rats was delayed in rats starved 24 and 48 hours. This correlated with a progressively slower synthase phosphatase response to insulin. The stimulatory effect of insulin was lost completely in 72-hour fasted rats. The proposed mechanism for the delayed response in rats starved 24 and 48 hours and lack of response in rats starved 72 hours is insulin resistance The mechanism remains to be elucidated. J. Nutr. 114: 2147-2154, 1984. INDEXING KEY WORDS rat heart •starvation •fasting •glycogen glycogen synthase •synthase phosphatase •insulin activation •insulin resistance

We have previously demonstrated that level observed in fed rats. Thus, the delayed pharmacological doses of insulin adminis- synthase response was probably due to the tered to intact, fed rats resulted in an in- time required for insulin to stimulate phoscrease in the percent of synthase in the I phatase activity in these animals. (active) form in heart (1). This was associSince starvation for 20 hours resulted in a ated with a modest increase in synthase decrease in the activities of synthase I and phosphatase (glycogen-synthase-D phospha- synthase phosphatase, we were interested in tase, EC 3.1.3.42) activity. In rats fasted 20 determining whether starvation of longer hours, where the basal synthase I was de creased, the synthase response to insulin was present but delayed. @19g4Amerjcan Instjtute ofNutrition. Received forpublication The synthase phosphatase activity was 20Aprii also low in rats fasted 20 hours compared to 'Thedatawerctakenfroma «uœrtation submitted byMaryc. rj ted

ratS.

TT HOWeVer,

. .. inSUlin

j . . administration

.

,_ 15

minutes before killing the rats resulted in a ft stimulation of phosphatase activity to the

Gannon to the Graduate School of the Universityof Minnesota in partial fulfillment of the requirementsfor a Doctor of Philosophydegree in Food

^i6"08andNu,triliâ„¢\. , , Recipient of a Medical Investigatorship Administration. 2147

«i *«.,h.w*.

Award from the Veterans

2148

GANNON AND NUTTALL

duration would further decrease the activi ties of these enzymes and whether the response to insulin would be maintained. Also, we wished to correlate the proportion of synthase in the I form with the increased cardiac glycogen concentration usually ob served in fasted rats. This is of interest be cause in vitro high glycogen concentrations inhibit synthase phosphatase activity (1). MATERIALS AND METHODS

thase, synthase phosphatase, insulin, glu cose, glucagon and glycogen assays were performed as previously described (1, 3). Briefly, glycogen synthase (EC 2.4.1.11) was assayed by the filter paper method of Thomas et al. (4). Synthase D was deter mined by including 10 mM glucose-6-phosphate (Glu-6-P) in the assay mixture. One unit of glycogen synthase activity is defined as the amount of enzyme that catalyzes the incorporation of 1 /tmol of glucose from UDPG into glycogen per minute at 30 °C. Synthase phosphatase was assayed by incu bating a tissue extract for 0-10 minutes at 30 °C.The reaction was stopped by adding an aliquot of incubated extract to 200 ¡A of ice-cold fluoride buffer (1:9 dilution). The buffer contained 200 mM KF, 2.5 mM EDTA, 10 mM KH2PO4, pH 7.8. A glycogen syn thase assay was then carried out on this sam ple. Synthase phosphatase activity is defined as the increase in synthase I or A units of synthase I. Tissue glycogen was determined by digesting ~100 mg of tissue in 30% KOH. Glycogen content of the tissue was determined by using the phenol-sulfuric acid method of DuBois et al. (5). A commercial glucose solution from Fisher Scientific Co. (Pittsburgh, PA) was used as standard. Plas ma glucose was assayed with a Beckman glucose analyzer (Beckman Instruments, Fullerton, CA). Plasma insulin was deter mined by radioimmunoassay with guinea pig antibody prepared in our laboratory. Rabbit anti-guinea pig sera was purchased from Miles-Yeda, Ltd. (Elkhart, IN). Gluca gon was determined by radioimmunoassay with rabbit 30K antibody purchased from Dr. Roger Unger, Dallas, TX. A dextrancharcoal precipitation of the antigen-anti body complex was employed. Statistical analyses were done by analysis of variance by using Dunnett's method for multiple

Labeled UDPG (UDP-[14C]glucose), [MC]glucose-l-P, 125I-labeled insulin and 12sl-labeled glucagon were obtained from New England Nuclear (Boston, MA). Sodi um secobarbital (Seconal), and glucagonfree insulin were obtained from Eli Lilly and Co. (Indianapolis, IN). 2-Mercaptoethanol was from Eastman Kodak Co. (Rochester, NY). Glucagon antibody, 30K, was purchased from Dr. Roger Unger, The University of Texas, Health Science Center at Dallas, Dallas, TX. Other reagents of the greatest possible purity were purchased from Sigma Chemical Company (St. Louis, MO). Type III rabbit-liver glycogen was passed over a mixed-bed ion-exchange resin as described previously (2). Male Holtzman rats, 160-200 g were housed in a temperature- (22.2 °C) and light- (12-hour cycle) controlled animal room. Animals were fed Purina rat chow (Ralston Purina Co., St. Louis, MO) ad libi tum, or starved for 24, 48 or 72 hours. The rats were anesthetized i.p. with Seconal, 50 mg/kg. Approximately 15 minutes later insulin, 6 U/kg, or saline was injected i.p. The rats were killed at timed intervals by quickly opening the chest and removing the heart. Only well-anesthetized, noncyanotic animals were used. For synthase assays, the heart was immediately frozen in liquid nitrogen-cooled aluminum clamps. The comparisons. The criterion of significance atria and great vessels were trimmed away, was a P value of less than 0.05. and the ventricles were stored in liquid nitrogen until they were assayed later the RESULTS same day. For synthase phosphatase assays The administration of insulin to intact, the heart was removed and immediately placed in ice-cold salina Blood was drained fed rats resulted in a rapid increase in the percent of synthase in the active form, from from the chest cavity into beakers contain ing heparin, then transferred to tubes in an approximately 20% in fed animals to 27% 5 ice bath and later centrifuged at 4°C.The minutes after insulin treatment as noted plasma was stored frozen at -20°C. Syn previously (1) (fig. 1). The percent synthase

2149

STARVATION EFFECTS ON GLYCOGEN ENZYMES

hours of fasting ( ~ 1.0 ^mol/minute per gram wet weight) (P < 0.01) (data not shown). 28 There was an inverse correlation between percent synthase I and the glycogen concen 26 tration. Cardiac glycogen concentrations 24 increased progressively with the duration of 22 starvation from 4.6 mg/g wet weight in fed rats to 5.2, 6.9 and 7.6 mg/g wet weight in 20 rats starved 24, 48 and 72 hours, respectively 18 (fig- 2). 16 Unstimulated synthase phosphatase activ 01 s* 14 ity was similar in extracts from rats starved for 24 (1), 48 and 72 hours (figs. 3, 4). In all 12 cases it was decreased approximately 50% IO compared to fed animals. The mean plasma insulin concentration 8 decreased from 61 /tU/ml found in fed rats to 6 42 /ill/ml in 24-hour fasted, 35 /tU/ml in 4 48-hour fasted and 33 jiU/ml in 72-hour 72 hr Starved fasted rats (table 1). Glucose concentrations 5 10 IS were decreased modestly at 24 and 48 hours Minutes after Insulin of fasting and were increased modestly at 72 Fig. 1 Effect of fasting on the synthase response to hours. Thus, the decrease in synthase phos insulin. Numbers in parentheses indicate number of phatase activity correlated with a decrease animals studied at each time point; vertical bars, SEM. in insulin concentration during prolonged Statistical comparisons were done by using Dunnett's starvation. method for multiple comparisons: P < 0.05 signifi 30

TSTF (5)

cant. "Statistically significant compared to the animals studied at the 0 time point.

I remained elevated for 15 minutes after insulin administration, the longest time period studied. In rats starved for 24 hours, the initial percent synthase I was decreased to 13%, and the response to insulin was delayed, reaching 18% by 10 minutes after insulin administration. These results agreed well with previously published data (1). In animals starved for 48 hours, the initial percent synthase I was further decreased to 7%. Ten minutes after insulin administra tion the percent synthase I was significantly increased to 12%. In rats starved 72 hours, the initial percent synthase I was further reduced (5.5%) and insulin administration failed to stimulate an increase in the percent synthase I in these rats. Thus, with pro longed starvation there was a continual decrease in percent synthase I for up to 72 hours. Total synthase activity was similar in fed, 24- and 48-hour starved rats (approxi mately 1.7 /imol/minute per gram wet weight) but it decreased significantly after 72

1*i

FED ,24

48

72

FASTED, hrs

Fig. 2 Effect of starvation on cardiac glycogen. The numbers in the bars indicate the number of ani mals studied in each group, vertical lines, SEM.Glyco gen concentration was significantly increased in hearts from rats fasted 48 and 72 hours.

2150

GANNON AND NUTTALL 1.2

Hysteresis was present in the phosphatase activity profiles from all groups of animals Fed(figs. 3, 4), but was greatest in extracts from 1.0 the starved rats. In fed rats there was a very 09 slight lag in the A units synthase I (phospha 0.8 tase activity) from 0 to 2 minutes, whereas Starved in the 24- and 48-hour starved rats there was + 30' Ins 0.7 a marked lag in phosphatase activity for 2-4 0.6 minutes. The reason for this is unknown. Q5 For comparative purposes, however, phos phatase activity was estimated as the amount 0.4 of substrate converted over the entire 100.3 minute period of incubation as A units syn 0.2 thase 1/10 minutes (fig. 5) in each group of Ql animals. The phosphatase activity in the control animals is represented as 100%. In 2468 10 rats starved 24 hours insulin stimulated Minutes of Incubation (30°) synthase phosphatase activity by 70% when administered 15 minutes before the animals Fig. 3 Effect of insulin on synthase phosphatase activity in 48-hour starved rats. Numbers in paren were killed. In rats starved 48 hours the theses indicate number of animals studied at each time phosphatase activity was stimulated by only point. Because of hysteresis, statistical comparisons were done by using Dunnett's significant difference, 27% 15 minutes after insulin, but by 30 and data were compared at each time point: P < 0.05, minutes the synthase phosphatase activity was 66% greater than in the control ani significant. Phosphatase activity was significantly decreased in 48-hour fasted rats at all time points and mals. Thus, in rats fasted 48 hours, insulin was significantly increased at both 15 minutes and 30 stimulation of synthase phosphatase activity minutes after insulin (Ins) administration at the 2-, 4-, was delayed compared to rats starved 24 6- and 10-minute time points. hours. In rats starved 72 hours insulin had little effect on synthase phosphatase activity. .• (39)

I.I

Synthase phosphatase activity was stimu lated by insulin in 48-hour fasted rats (fig. 3), but the response at 15 minutes was less than previously observed in 24-hour starved ani mals (1). By 30 minutes the phosphatase ac tivity was further increased. It approached that seen in fed rats and was similar to that observed in rats fasted 24 hours at 15 min utes after insulin was given (1). Thus, the response to insulin was slower in the rats fasted 48 hours than in the rats fasted 24 hours. Nevertheless, it is apparent that suf ficient stimulation is present to give a maximal synthase response by 10 minutes (fig- I)In extracts from rats fasted 72 hours, the synthase phosphatase response to insulin was considerably impaired over the entire 30-minute period (fig. 4). There was a modest increase, but this did not reach sta tistical significance The phosphatase results correlated well with the synthase I response to insulin administration in these animals (fig- I)-

1.2 I.I

,•(39)

1.0

Fed-

as as a?

.'

Starved '

ig\

0.6

as 04 0.3 0.2

ai 0

2

4

6

8

10

Minutes of Incubation (30°)

Fig. 4 Effect of insulin in synthase phosphatase activity in 72-hour starved rats. Description is the same as for fig. 3. Phosphatase activity was significantly decreased in 72-hour fasted rats compared to fed rats. Phosphatase activity was not stimulated significantly by insulin administration.

STARVATION EFFECTS ON GLYCOGEN ENZYMES DISCUSSION

In the present studies, it is clear that starvation for up to 72 hours results in a gradual diminution in the basal percent synthase I in heart. There was little change in total synthase activity until 72 hours of starvation. The amount of synthase in the I form is due to the relative activities of synthase phosphatase and synthase kinase (glycogen synthase a kinase EC 2.7.1.37). An increase in synthase kinase activity, a decrease in syn thase phosphatase activity or both, would explain the decreased synthase I observed in the present studies. The number of kinases involved in syn thase regulation and their characteristics are still being identified (6), but cAMPdependent and cAMP-independent protein kinases have been implicated. Preliminary experiments indicated no change in cAMPdependent or cAMP-independent kinase activity with starvation (7). However, changes in specific kinases such as a calcium-depen dent kinase (8, 9) were not studied. An increased kinase activity resulting in a decrease in synthase I activity in hearts from starved rats has not been eliminated. How ever, the current studies have focused on the phosphatase enzyme since it had been shown previously that phosphatase activity decreased with 17-24 hours of starvation (1). The current data clearly indicate that syn thase phosphatase activity also is decreased at 48 and 72 hours of starvation (fig. 3, 4), but is little changed from that observed in 24-hour starved rats. Although the decreased phosphatase activity observed in the extracts

24 hr Starved

48hr Starved

2151 72 hr Starved

Con 15' 30' Ins Ins Fig. 5 Insulin stimulation of synthase phosphatase activity in rat heart. The phosphatase activity was estimated over the entire 10-minute period of incuba tion. Data previously obtained (1) were used for the 24-hour starved rats. Con, synthase phosphatase activ ity in control animals. The number of animals studied is indicated in the bars. 'P < 0.05: a significant stimu lation by insulin.

from rats starved 48 and 72 hours was similar to that in extracts from rats starved 24-hours (réf. 1, fig. 5), the percent synthase I decreased progressively with starvation. Thus, the decrease in measured phosphatase activity could not explain the progressive synthase I decrease This indicated the pos sibility that changes in the concentration of factors known to affect synthase phospha tase were affecting the catalytic activity of the enzyme in vivo, independent of the amount of enzyme present. Two factors which have been implicated in the regula tion of synthase phosphatase activity are citrate and glycogen. By using purified heart synthase as sub strate, 1.4 mM citrate inhibited synthase TABLE 1 phosphatase activity by 32% (10). Also, an Plasma insulin and glucose concentrations1'1 increase in citrate concentration has been reported in fasted rats, from approximately groupFed24-hour Animal 0.4 /tmol/g wet weight in fed rats to approxi mately 0.7 /imol/g wet weight in rats starved 24 and 48 hours (11, 12). However, the (39)106 + 3 (48)42 ±5 (14)106 ±6* starved48-hour citrate concentration as measured in heart (13)35 ±8 (21)33 ±5* (30)116 ±3* starved72-hour tissue represents both mitochondrial and ±6* (12)Glucosemg/dl123 starvedInsulin¡iU/ml61 ±8 (12) cytosolic citrate Approximately 95% of the citrate is mitochondrial (13). Thus, the cyto 'Values are means ±SEfor the number of animals solic citrate concentration is only approxi studied stated in parentheses. 'Significantly differ ent from the fed animals *P < 0.05. Values from the mately 0.04 mM in fasted rats, a concentra 48- and 72-hour starved rats were not significantly dif tion that is not significantly inhibitory. ferent from the 24-hour starved rats. It has been known for decades that the

2152

GANNON AND NUTTALL INSULIN

60

I 9

» u, O

US

40

0.8

20

S o O- \

o,

SYNTHASE' PHOSPHATASE

20 .% SYNTHASE I 8

16

S

_

GLYCOGEN -

.2

4

FED

24

48

S o

72

Hours of Fasting

Fig. 6 Relationship between serum insulin concen tration and cardiac synthase phosphatase activity (top panel) and between glycogen concentration and per cent synthase I (bottom panel). The serum insulin and cardiac synthase phosphatase activity both decrease considerably during the first 24 hours of starvation and are little changed thereafter. Cardiac glycogen in creases progressively with starvation at which time the percent of synthase in the I form decreases, gww, gram wet weight.

cardiac glycogen concentration increased with fasting (14-17). This was demonstrated in the present study as well (fig. 2). Several years ago, Danforth (18) reported that the glycogen concentration in rat diaphragm tissue was inversely related to the percent synthase I. We also showed an inverse cor relation between the glycogen concentration and the percent synthase I in heart tissue from fasted rats (fig. 6). That glycogen should be involved in controlling its own synthesis has long been an attractive hy pothesis (19). We (1), as well as others (20) have pre viously reported that glycogen in the physio logical range inhibited synthase phospha tase in a concentration-dependent manner in vitro. In the present studies (fig. 2-4) glycogen was routinely added to the homogenate in the synthase phosphatase assay in order to standardize the concentration near

7-8 mg/ml. Therefore, although progres sively increasing glycogen concentrations observed with fasting would not detectably influence the phosphatase activity as as sayed, it could inhibit the synthase phospha tase activity in vivo. Thus, in vivo, glycogen may be inhibiting synthase phosphatase activity and consequently reducing the pro portion of synthase in the I form. Synthase phosphatase activity in rats starved 24 hours was decreased approxi mately 50% compared to fed animals, but with further starvation, it remained un changed. The reason for the decrease is uncertain. However, synthase phosphatase activity correlated with plasma insulin con centrations (fig. 6). Both decreased rapidly by 24 hours and then tended to stabilize Thus, the decrease in synthase phosphatase activity with starvation may be due to a reduced insulin level. It is also possible that there are two forms of phosphatase (21), only one of which is affected by insulin and stimulants. In rats starved 48 hours, the synthase I response to insulin was slower than the response observed in rats fasted 24 hours. This could be explained by a decreased stimulation of the phosphatase by insulin or perhaps to an increased glycogen concentra tion. Even though the insulin desensitizes the enzyme to glycogen inhibition (1), the glycogen concentration may still have been sufficient to delay the response After 72 hours of starvation neither syn thase phosphatase nor synthase I responded to insulin. Both the slow response at 48 hours and the absence of a response at 72 hours may be due to a starvation-induced insulin resistance (22). Insulin resistance is not well understood. An association between insulin resistance and decreased receptor number has been noted in cases of hyperinsulinemia and obesity. This mechanism is not likely to be important, however, since insulin binding to only 10% of the receptors has been reported to result in maximal biological action, at least in adipocytes (23). In addition, with starvation-induced insulin resistance, insu lin binding to fat cell receptors has been reported to be increased (24). It has been proposed that insulin pro duces a chemical signal at the membrane,

STARVATION EFFECTS ON GLYCOGEN ENZYMES

resulting in phosphorylation and proteolysis of the receptor. The proteolytic product has been called the insulin mediator (25). Whether receptor phosphorylation is af fected by starvation or whether a reduction in mediator production has occurred in the present studies is unknown. Begum et al. (26) have presented evidence from adipocyte plasma membranes that rats adapted to a high fat diet have a reduced capacity to generate the mediator substance in response to insulin. If a high fat diet simulates starvation metabolically (27), it is possible that the insulin resistance observed in rats starved 48 and 72 hours is due to decreased generation of the insulin mediator. LITERATURE CITED 1. Nuttall, F. Q., Gannon, M. C., Corbett, V. A. & Wheeler, M. P. (1976) Insulin stimulation of heart glycogen-synthase-D phosphatase (protein phosphatase). J. Biol. Chem. 251, 6724-6729. 2. Huijing, F., Nuttall, F. Q., Villar-Palasi, C. & Lamer, J. (1969) UDPGlucose:a-l,4-glucan a4-glucosyltransferase in heart. Regulation of the activity of the transferase in vivo and in vitro in rat. A dissociation in the action of insulin on transport and on transferase conversion. Biochim. Biophys. Acta 177, 204-212. 3. Nuttall, F. Q., Gannon, M. C. & Bergstrom, W. J. (1975) Insulin and epinephrine effects on heart glycogen synthase and phosphorylase activity. Am. J. Physiol. 228, 1815-1820. 4. Thomas, J. A., Schlender, K. K. & Larner, J. (1968) Rapid filter paper assay for UDPGlucoseglycogen glucosyltransferase, including an im proved biosynthesis of UDP-'4C-glucose. Anal. Biochem. 25, 486-499. 5. DuBois, M., Gilles, K. A., Hamilton, J. R., Rebers, P. A. & Smith, F. (1956) Colorimetrie method for determination of sugars and related substances. Anal. Chem. 28, 350-356. 6. Gannon, M. C. (1983) Effect of diet or starva tion on glycogen synthase phosphatase activity in rat heart. Ph.D. Thesis, University of Minnesota, Minneapolis, MN. 7. Nuttall, F. Q. (1970) Studies on the hormonal and non-hormonal control of the rat heart glyco gen transferase system. Ph.D. Thesis, University of Minnesota, Minneapolis, MN. 8. Payne, E. M. & Soderling, T. R. (1980) Calmodulin-dependent glycogen synthase kinase J. Biol. Chem. 255, 8054-8056. 9. Ahmad, Z., DePaoli-Roach, A. A. & Roach, P. J. (1982) Purification and characterization of rab bit liver calmodulin-dependent protein kinase able to phosphorylate glycogen synthase. J. Biol. Chem. 257, 8348-8355.

2153

10. Nakai, C. & Thomas, J. A. (1974) Properties of a phosphoprotein phosphatase from bovine heart with activity on glycogen synthase, phosphorylase and histone J. Biol. Chem. 249, 6459-6467. 11. Williamson, D. H. & Brosnan, J. T. (1974) Concentrations of metabolites in animal tissues. In: Methods of Enzymatic Analysis, vol. 4, (Bergmeyer, H. U., ed.) 2nd English éd., pp. 2266-2302, Verlag Chemie Weinheim, Academic Press, New York. 12. Adrouny, G. A. (1969) Differential patterns of glycogen metabolism in cardiac and skeletal mus cles. Am. J. Physiol. 217, 686-693. 13. Kauppinen, R. A., Hiltunen, J. K. & Hassinen, I. E. (1982) Compartmentation of citrate in relation to the regulation of glycolysis and the mitochon dria! transmembrane proton electrochemical po tential gradient in isolated perfused rat heart. Biochim. Biophys. Acta 681, 286-291. 14. Evans, G. & Bowie, M. A. (1936) Cardiac gly cogen in diabetic animals. Proc. Soc. Exp. Biol. Med. 35, 68-71. 15. Russell, J. A. & Bloom, W. (1956) Hormonal control of glycogen in the heart and other tissues in rats. Endocrinology 58, 83-94. 16. Adrouny, G. A. & Russell, J. A. (1956) Effects of growth hormone and nutritional status on cardiac glycogen in the rat. Endocrinology 59, 241-251. 17. Hazelwood, R. L. (1968) Growth hormone, plasma glucose and ketone bodies as determinants of cardiac glycogen in normal and diabetic rats. Proc. Soc. Exp. Biol. Med. 127, 450-458. 18. Danforth, W. H. (1965) Glycogen synthetase activity in skeletal muscle. Interconversion of two forms and control of glycogen synthesis. J. Biol. Chem. 240, 588-593. 19. Larner, J. (1972) Insulin and glycogen syn thase Diabetes 21, 428-438. 20. Thomas, J. A. & Nakai, C. (1973) Control of glycogen synthase phosphatase from rat heart. The role of substrate. J. Biol. Chem. 248, 2208-2213. 21. Nuttall, F. Q., Gilboe, D. P., Tan, A. W. H., Doorneweerd, D. D., Theen, J. W., Gannon, M. C. & Chou, B. B. (1981) Regulation of liver glyco gen synthesis. In: The Regulation of Carbohydrate Formation and Utilization in Mammals (Veneziale, Carlo M., éd.),pp. 315-343, University Park Press, Baltimore, MD. 22. Truheart, P. A. & Herrera, M. G. (1971) De creased response to insulin in adipose tissue during starvation. Diabetes 20, 46-50. 23. Ip, C., Tepperman, H. M., Holohan, P. & Tepperman, J. (1976) Insulin binding and insulin response of adipocytes from rats adapting to fat feeding. J. Lipid Res. 17, 588-599. 24. Smith, L. H., Jr. & Thier, S. O. (1981) Fuelhormone interactions. Starvation. In: Pathophysiology. The Biological Principles of Disease (Samiy, A. H., Smith, L. H., Jr. & Wyngaarden, J. B., eds.), pp. 548-556, W. B. Saunders Co., Phila delphia, PA. 25. Larner, J., Cheng, K., Schwartz, C., Kikuchi, K.,

2154

GANNON AND NUTTALL

Tuiimra, S., Creacy, S., Dubler, R., Calasko, G., Pullin, C. & Katz, M. (1982) Insulin mediators and their control of metabolism through protein phosphorylation. Ree. Prog. Horm. Res. 38, 511-556. 26. Begum, N., Tepperman, H. M. & Tepperman, J. (1983) Effects of high fat and high carbohydrate

diets on liver pyruvate dehydrogenase and its activation by a chemical mediator released from insulin-treated liver particulate fraction: effect of neuraminidase treatment on the chemical mediator activity. Endocrinology 112, 50-59. 27. Canili, G. F., Jr. (1970) Starvation in man. N. Engl. J. Med. 282, 668-675.